spectrum of pore types and networks in mudrocks and a ...€¦ · ventional sandstone (i.e.,...

AUTHORS

Robert G Loucks Bureau of EconomicGeology Jackson School of Geosciences Univer-sity of Texas at Austin University Station Box XAustin Texas bobloucksbegutexasedu

Robert G Loucks is a senior research scientist atthe Bureau of Economic Geology He receivedhis BA degree from the State University ofNew York at Binghamton in 1967 and his PhDfrom the University of Texas at Austin in 1976

Spectrum of pore types andnetworks in mudrocks and adescriptive classification formatrix-related mudrock poresRobert G Loucks Robert M Reed Stephen C Ruppeland Ursula Hammes

His general research interests include carbonateand siliciclastic sequence stratigraphy deposi-tional systems diagenesis and reservoir char-acterization His present research includes porenetworks in carbonates sandstones and mud-rocks evaporite and carbonate paleokarst deepto ultradeep reservoirs in the Gulf of Mexicoand shale-gas systems

Robert M Reed Bureau of EconomicGeology Jackson School of Geosciences Uni-versity of Texas at Austin University StationBox X Austin Texas robreedbegutexasedu

RobM Reed is a research scientist associate at theBureau of Economic Geology He received hisBS degree and his PhD in geologic sciencesfrom the University of Texas at Austin and hisMS degree in geology from the University ofMassachusetts His current research focuses onvarious aspects of the microstructure of rocks

Stephen C Ruppel Bureau of EconomicGeology Jackson School of Geosciences Uni-versity of Texas at Austin University StationBox X Austin Texasstephenruppelbegutexasedu

Stephen C Ruppel is a senior research scientistat the Bureau of Economic Geology in Austinwhere he specializes in sedimentologic charac-terization of carbonate and mudrock successionsand is currently director of the Mudrock SystemsResearch Laboratory Before coming to the Bu-reau in 1981 he worked at McGill Universityin Montreal and Chevron Oil Company in NewOrleans He received his PhD from the Uni-

ABSTRACT

Matrix-related pore networks in mudrocks are composed ofnanometer- tomicrometer-size pores In shale-gas systems thesepores along with natural fractures form the flow-path (perme-ability) network that allows flow of gas from the mudrock toinduced fractures during production A pore classification con-sisting of threemajor matrix-related pore types is presented thatcan be used to quantify matrix-related pores and relate themto pore networks Two pore types are associated with the min-eral matrix the third pore type is associated with organic matter(OM) Fracture pores are not controlled by individual matrixparticles and are not part of this classification Pores associatedwith mineral particles can be subdivided into interparticle(interP) pores that are found between particles and crystals andintraparticle (intraP) pores that are located within particlesOrganic-matter pores are intraP pores located within OMInterparticle mineral pores have a higher probability of beingpart of an effective pore network than intraP mineral poresbecause they are more likely to be interconnected Althoughthey are intraP OM pores are also likely to be part of an inter-connected network because of the interconnectivity of OMparticles

In unlithifed near-surface muds pores consist of interP andintraP pores and as the muds are buried they compact andlithify During the compaction process a large number ofinterP and intraP pores are destroyed especially in ductile grain-rich muds Compaction can decrease the pore volume up to

versity of Tennessee and his BS and MS de-grees from the University of Illinois and Uni-versity of Florida respectively His currentresearch is focused on integrating sedimento-logic geochemical and stratigraphic data fromoutcropping and subsurface Paleozoic andCopyright copy2012 The American Association of Petroleum Geologists All rights reserved

Manuscript received May 16 2011 provisional acceptance July 25 2011 revised manuscript receivedAugust 2 2011 final acceptance August 17 2011DOI10130608171111061

AAPG Bulletin v 96 no 6 (June 2012) pp 1071ndash 1098 1071

Mesozoic carbonate and shale successions todevelop process and response models of de-position and diagenesis

Ursula Hammes Bureau of EconomicGeology Jackson School of Geosciences Uni-versity of Texas at Austin University StationBox X Austin Texasursulahammesbegutexasedu

Ursula Hammes obtained her diploma in geol-ogy from the University of Erlangen in Germanyin 1987 and her PhD from the University ofColorado at Boulder in 1992 She spent 10 yrworking as a consultant performing postdoc-toral research at the Bureau of EconomicGeology and as an exploration geologist in theindustry She joined the Bureau of EconomicGeology in 2001 as a research associate Hermain research focus is in clastic and carbonatesequence stratigraphy depositional systemsand carbonate and clastic diagenesis Her recentresearch objective is in shale-gas systems inTexas and Germany

ACKNOWLEDGEMENTS

This research was funded by the Mudrock Sys-tems Research Laboratory and the State of TexasAdvanced Resource Recovery (STARR) Projectat the Bureau of Economic Geology Industrymembers include Anadarko BP ChesapeakeChevron Cima Cimarex ConocoPhillips CypressDevon Encana EOG Resources EXCO ResourcesHusky Marathon Pangaea Penn Virginia PennWest Pioneer Shell StatOil Texas AmericanResources The Unconventionals US EnerCorpValence and YPF (Yacimientos PetroliferousFiscales) Editing was provided by Lana DieterichBureau of Economic Geology The authors ac-knowledge reviews by David Awwiller ShirleyDutton Terri Olson and Arthur Trevena RuarriDay-Stirrat provided an ion-milled sample fromthe Ursa Basin in the Gulf of Mexico Publicationwas authorized by the director of Bureau ofEconomic Geology Jackson School of Geosci-ences University of Texas at AustinThe AAPG Editor thanks the following reviewersfor their work on this paper David N AwwillerTerrilyn M Olson and Arthur Trevena

1072 Spectrum of Pore Types and Networks In

88 by several kilometers of burial At the onset of hydrocar-bon thermal maturation OM pores are created in kerogen Atdepth dissolution of chemically unstable particles can createadditional moldic intraP pores

INTRODUCTION

Nanometer- to micrometer-size connected matrix-related poresform flow networks along with fractures where present thatcontribute to the natural permeability pathways for gas flowin unconventional mudrock reservoirs (the term ldquomudrockrdquo isused to includemudstones and shales) (Javadpour 2009 Curtiset al 2010) Research during the last several years has identi-fied a variety of pore types in thesemudrocks (ie Davies et al1991 Desbois et al 2009 Loucks et al 2009 2010 Curtiset al 2010 Diaz et al 2010Milner et al 2010 Passey et al2010 Schieber 2010 Sisk et al 2010 Lu et al 2011 Slatt andOrsquoNeal 2011) We propose a simple relatively objective de-scriptive approach to grouping mudrock matrix-related poresinto three basic categories on the basis of their relationships toparticles (Figure 1) Pore types include interparticle (interP)mineral pores intraparticle (intraP) mineral pores and intraPorganic-matter (OM) pores This type of pore classificationis consistent with schemes of pore subdivisions used in con-ventional sandstone (ie Pittman 1979) and carbonate (ieChoquette and Pray 1970) reservoir systems Although de-scriptive it provides a first step in relating pores to reservoirproperties For example type size and arrangement of poresaffect storage and calculations of hydrocarbons in place (ieAmbrose et al 2010) as well as sealing capacity (Dawson andAlmon 2002 Dewhurst et al 2002 Schieber 2010) Alsodifferences in the origins and distribution of pore types havebeen shown to have different effects on permeability and wet-tability (ie McCreesh et al 1991 Passey et al 2010) Thesethree categories of pores can be plotted on a ternary diagram tocompare different scales of sampling such as different areasof a scanning electron microscope (SEM) sample different in-tervals of an individual shale play or summary comparisons ofdifferent shale plays in different basins

The major goals of this article are to document in a simpleand systematic manner the wide spectrum of matrix-relatedpores that exist in mudrocks and classify these pores on thebasis of basic descriptive attributes Specific objectives are to(1) present a pore classification that summarizes the spec-trum of pore types seen in mudrock reservoirs and that alsocaptures some important petrophysical properties of the pores

Mudrocks

(2) introduce a ternary classification that allowscomparison of mudrocks that contain matrix-related pores (3) compare pores within unlithi-fied and shallow muds with pores within ancientmore deeply buried lithified mudrocks and (4) dis-cuss how different pore types contribute to effec-tive matrix-flow pathways in mudrocks

DATA AND METHODS

Data for this investigation of pore types inmudrockscome from core and outcrop samples collected froma variety of formations Many of the data were gen-erated by the Mudrock Systems Research Labora-tory at the Bureau of Economic Geology JacksonSchool of Geosciences University of Texas at Aus-tin Some data are from published studies thatcontain SEMphotomicrographs ofmudrock poresTable 1 presents a summary of data showing theage unit name general location pore types accord-ing to present classification and source of dataThe 26 geologic units included in this survey rangein age from Cambrian to PliocenendashPleistocene indepth from outcrop to 15950 ft (4752 m) and inmaturity from less than 05 to 317 vitrinite re-flectance (Ro)

As pointed out by Loucks et al (2009) most ofthe small pores in mudrocks are generally smallerthan a few micrometers in diameter and cannot beseen using light microscopy These workers alsonoted that polished thin sections produced surfacetopographic irregularities (their figure 3) becauseof differential hardness of components making thinsections inadequate for pore identification usingSEM-based imaging techniques Pittman (1992)also noted that grains are plucked out during thin-section preparation producing artifact pores Rockchips have a tendency to produce a pull-out ofgrains (grain plucking) that may be misidentifiedas pores and this dilemma gives a low confidencelevel of correct identification of artifacts from ac-tual pores Some authors (ie Milner et al 2010)disagreed with the inadequacies of rock chips butdetailed work by the present authors has substan-tiated abundant pull-out artifacts as has the workby Sondergeld et al (2010)

Hildenbrand and Urai (2003) used Woodrsquosmetal injection technique to create pore casts inmudrocks They were able to image a pore cast assmall as 40 nm in diameter and showed larger poresconnected by smaller pore throats (their figure 11)Theirs is an interesting and informative techniqueand should be considered in the study of mudrockpore networks However we think that our meth-od (described herein) displays pores more clearlyMethods that indirectly attempt analysis of porestructures in mudrocks also exist such as by con-ducting high-pressure CH4 isotherm analysis (Rossand Bustin 2009) These techniques were not usedin this study

Our samples were all prepared using ion-millingtechniques (Reed and Loucks 2007 Loucks et al2009) The technique is similar to a commonmethodfor preparation of electron-transparent transmis-sion electronmicroscope samples (eg Hover et al1996) although the milling is confined to one sur-face instead of two Argon (Ar)-ion-beam millingproduces surfaces showing only minor topographicvariations unrelated to differences in hardness ofthe sample but rather to slight variations in thepath of the Ar-ion beam Tomutsa et al (2007)promoted using this ion-milling technique to im-age pore systems in diatomite and chalk These flatsurfaces allow the quantification of types and num-ber of pores in two dimensions Pores viewedwithinthree-dimensional (3-D) rock chips cannot be quan-tified because of the irregular surface and associatedpull-outs A limitation of using Ar-ion-milled sam-ples is the inherently small viewing area providedby this technique The milled area measures onlyapproximately 15 times 05 mm Although this area issmall relative to the nanometer to micrometer sizeof pores in mudstones many pores can nonethelessbe observed

Artifacts associatedwith sample preparation andAr-ion milling exists but they are easy to recognizeSamples with clay-rich matrix and large clay par-ticles can desiccate and produce shrinkage cracks(Figure 2AndashC) on drying The most common arti-facts associated with ionmilling are (1) redepositionof milled material within some pores (Figure 2D)and (2) curtaining (Figure 2E F) Curtaining whichis typically expressed as flaring structures associated

Loucks et al 1073

1074 Spectrum of Pore Types and Networks In Mudrocks

with minor relief is primarily a minor annoyance inphotomicrographs it is easily identified and presentsno major problems for pore identification Redepo-sition of milled material is a more serious problemif not recognized Although not documented thisprocess could completely fill some of the smallerpore population the redeposited material can berecognized however in backscatter by contrastin density Redeposited material appears as anom-alous lighter colored areas around the edges ofpores this material commonly fills the upflow partof larger pores (Figure 2D)

Ion-milled sampleswere imaged on several typesof SEMs Samples were examined using two differ-ent field-emission SEMs One was a Zeiss Supra 40VP and the other was an FEI Nova NanoSEM 430Both systems are at the University of Texas at Aus-tin Use of these field-emission SEMs equippedwith in-lens secondary electron (SE) detectors pro-vides greatly increased detail of nanometer-scalefeatures Lower accelerating voltages (1ndash10kV)weretypically used on these systems to prevent beamdamage of the sample Working distances were 3to 7 mm

Secondary electron images were acquired fordocumenting topographic variation and backscat-tered secondary electron (BSE) images were ac-quired for delineating compositional variation En-ergy dispersive spectroscopy analyses of specificgrains were also conducted for mineral identifica-tion These analyses and images provided the basicdata for lithologic determination and types and lo-cations of pores within the sample

Several publications (Table 1) have providedexcellent photographs of mudrock pores acquiredby SEM imaging Nearly all the samples including

those from the literature were prepared using theion-milling technique These published data pro-vided an additional range of sample age and re-gional extent to our own data Other authors haveused different terminologies to name and classifypore types Tables 1 and 2 compare their termi-nology with the terminology used in this article

The upper pore-size range in mudrocks is gen-erally less than a few micrometers with most poresbeing less than 1 mm Chalmers et al (2009) recom-mended that geoscientists working on mudrocksuse the pore-size terminology of the InternationalUnion of Pure and Applied Chemistry as devel-oped by Rouquerol et al (1994) Micropores asdefined by Rouquerol et al (1994) have widthsless than 2 nm mesopores have widths between 2and 50 nm and macropores have widths greaterthan 50 nm In this classification nearly all mud-rock pores would be grouped with larger pores incarbonates and sandstones asmacropores (Figure 3)Although the pore-size classification of Rouquerolet al (1994) may be appropriate for chemical prod-ucts such as describing membranes it is insufficientfor reservoir systems Choquette and Pray (1970)provided a useful pore classification for carbonaterocks within which they provided size modifiersThey definedmicropores as being less than 625 mmmesopores as being between 625 mm and 4 mmand megapores as being between 4 and 256 mmThese authors included no subdivisions of poressmaller than 625 mm

Becausemanymudrock pores commonly rangefrom a few nanometers to several micrometers indiameter it would be useful to have well-definednames for pores in this size range and for poressmaller than this size range We propose the terms

Figure 1 Pore type classes (A) Spectrum of pore types occurring within mudrocks General pore types include interparticle andintraparticle pores associated with mineral matrix organic-matter pores associated with organic matter and fracture pores that crosscutmatrix and grains Examples of common pore types are presented but are not meant to indicate formal subtypes or terminology Fracturepores are presented in this figure only to emphasize their importance However they are not considered matrix-related pores and are notpart of the matrix-related classification presented in this study (B) The mudrock pore classification ternary diagram was modified fromLoucks et al (2010) If a pore network is dominated by 50 or more of one pore type it assumes the name of that pore type If no onepore type dominates then it is a mixed-pore network Examples of pore networks (shown in Figures 5A 13) from the Barnett Shale (redcircle) Bossier Shale (green circle) Pearsall Formation (blue circle) and a PliocenendashPleistocene mudstone (orange circle) are plottedThe proportion of pore types for the Barnett Bossier and Pearsall was derived by point counting 2000 points and for the PliocenendashPleistocene by point counting 1000 points

Loucks et al 1075

Table 1 Data Used to Develop Pore Types

Age Unit Name General Location Pore Type (this article) Pore Type as Described by Author Sourcedagger

PliocenendashPleistocene Nankai accretionaryprism offshore Japan

InterP dominant andintraP present

Intergranular pores (authorsrsquo figure 11) Milliken and Reed(2010)

PliocenendashPleistocene Ursa Basin Gulf ofMexico


No data provided Day-Stirrat et al(2010b)

Oligocene Boom Clay Mol Belgium InterP and intraP Three types Type I elongated pores between similarlyoriented clay sheets (lt100 nm) type II crescent-shaped pores in saddle reefs of folded sheet of clay(100 nmndash1 mm) and type III large jagged poressurrounding clast grains (gt1 mm) (authorsrsquo figures 3 4)

Desbois et al (2009)

Eocene Wilcox Group West Baton RougeParish Louisiana

InterP and intraP Type I pores are aligned parallel to clay-mineralcleavage (authorsrsquo figure 10A B) type II pores arebedding-parallel cleavage cracks through severalgrains (authorsrsquo figure 10C) type III wedge-shapedpores are voids at terminations of stacked clayplatelets (authorsrsquo figure 11A) type IV pores are atface-edge mineral contacts (authorsrsquo figure 11B)type V pores occur at interfaces between clays anddetrital grains (authorsrsquo figure 11C) and type VImicrofractures occur within rigid grains (authorsrsquofigure 11C)

Kwon et al (2004)

Late Cretaceous Tuscaloosa Group Mississippi IntraP and OM Pores within pyrite framboids (authorsrsquo figure 9A)inorganic matter (authorsrsquo figure 9B) fluid-inclusionspores (authorsrsquo figure 9C) and pores in dissolvedTiO2 grain (authorsrsquo figure 9D)

Lu et al (2011)

Late Cretaceous Eagle Ford Formation South Texas shelf InterP and intraPdominant lesser OM

Loucks et al (2010)and MSRL study

Late Cretaceous Eagle Ford Formation South Texas shelf IntraP and OM Phyllosilicate and organophyllic pores (authorsrsquofigure 8)

Curtis et al (2010)

Early Cretaceous Mancos Shale Book Cliffs Utah InterP and intraP Phyllosilicate framework (authorrsquos figure 4AndashC) Schieber (2010)Early Cretaceous Pearsall Formation South Texas Maverick

BasinInterP and intraPdominant lesser OM


Late Jurassic Kimmeridge Clay Europe OM dominant Organophyllic pores (authorsrsquo figure 8) Curtis et al (2010)Late Jurassic Haynesville Formation East Texas Basin InterP and intraP with

minor OMLoucks et al (2010)and MSRL study

1076Spectrum

ofPoreTypes

andNetworks

InMudrocks

Late Jurassic Haynesville Formation East Texas Basin InterP and intraP withminor OM

Matrix intercrystalline and organic pores (authorsrsquofigure 1AndashC) and intraP pore (authorsrsquo figure 3)

Milner et al (2010)


Phyllosilicate pores (authorsrsquo figures 8 10 16 17) Curtis et al (2010)

Late Jurassic Bossier Shale East Texas Basin InterP and intraP withminor OM


Pennsylvanian Atoka Formation Permian Basin Rare interP and intraPno OM


Carboniferous Germany InterP intraP and OM MSRL studyMississippian Barnett Shale Fort Worth Basin

and Permian BasinOM dominant with intraPin pyrite framboids

Loucks et al (2009)Loucks et al (2010)and MSRL study

Mississippian Barnett Shale Forth Worth Basin OM dominant Organophyllic pores (authorsrsquo figure 8) Curtis et al (2010)Mississippian Barnett Shale Fort Worth Basin OM dominant Organic matter pores (authorsrsquo figure 2AndashC) Milner et al (2010)Mississippian Barnett Shale Fort Worth Basin OM dominant with interP Authorsrsquo figures 8 14 18 and 20 Sondergeld et al (2010)Mississippian Fayetteville Shale Arkoma Basin

ArkansasIntraP and OM Phyllosilicate and organophyllic pores

(authorsrsquo figure 8)Curtis et al (2010)

Mississippian Floyd Shale Black Warrior BasinAlabama

OM dominant Organophyllic pores (authorsrsquo figure 8) Curtis et al (2010)


OM and interP Authorsrsquo figure 13 Sondergeld et al (2010)

MississippianDevonian

Barnett Shale andWoodford Shale

Texas OM interP and intraP Organic porosity interP pores between flocculatedparticles pores within pellets and microchannelsin matrix

Slatt and OrsquoNeal (2011)

Devonian Woodford Shale Permian Basin OM dominant MSRL studyDevonian Woodford Shale OM dominant Organophyllic pores (authorsrsquo figures 6 and 8) Curtis et al (2010)Devonian New Albany Shale Illinois Basin InterP and intraP Loucks et al (2010)

and MSRL studyDevonian New Albany Shale Illinois Basin InterP and intraP Phyllosilicate framework (authorrsquos figures 1ndash3) and

carbonate-dissolution pores (authorrsquos figure 5A B)Schieber (2010)

Devonian Marcellus Shale Northeast USA OM dominant withminor intraP

Organic pores (authorsrsquo figure 2DndashF) and intraparticlepores

Milner et al (2010)

Devonian Marcellus Shale Northeast USA OM dominant Organophyllic pores (authorsrsquo figure 8) Curtis et al (2010)Devonian Geneseo Shale New York InterP intraP and OM Phyllosilicate framework (authorrsquos figure 4D E)

carbonate dissolution pores (authorrsquos figure 5E F)and organic matter pores (authorrsquos figure 6CndashE)

Schieber (2010)

Devonian Horn River Shale Canada OM dominant Organic pores (authorsrsquo figure 1DndashF) Milner et al (2010)Devonian Horn River Shale Canada OM dominant Authorsrsquo figures 8 9 Curtis et al (2010)

Loucksetal

1077

Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica

(authorsrsquofigure3A)

Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica

(authorsrsquofigure3D

)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM

Phyllosilicateframew

ork(authorrsquos

figure4F)

carbonatedissolutionpores(authorrsquos

figure5CD

)andorganicmatterpores(authorrsquos

figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP

=interparticleintraP=intraparticleO

M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger


ldquonanoporerdquo for pores less than 1 mm (1000 nm) andgreater thanor equal to 1 nanometer and ldquopicoporerdquofor pores less than 1 nm In our system picroporesare less than 1 nm nanopores range from 1 nm toless than 1 mmmicropores range from1 to less than625 mm mesopores range from 625 mm to lessthan 4 mm and macropores are larger than 4 mmFigure 3 compares our pore-scale schemewith thatof Rouquerol et al (1994)

MUDROCK PORE-TYPE CLASSES

Mudrock matrixndashrelated pores range from simpleto complex in shape and origin They form by bothdepositional and diagenetic processes Also poresmay show a multiple-stage origin related to depo-sition compaction cementation and dissolutionPores and associated pore networks are also partlya function of original matrixmineralogy (Figure 4)fabric (arrangement of grains) texture (size andsorting of grains) and OM composition

A classification ofmudrockmatrixndashrelated poresshould be simple and should also incorporate asbest as possible some inherent characteristics suchas permeability and wettability Carbonate (ieChoquette and Pray 1970 Lucia 1999) and coarsergrained siliciclastic (ie Pittman 1979 Dutton andLoucks 2010) pore classifications generally dividepores into those that occur between particles (interP)and those that occur within particles (intraP) Inter-particle pores include intergranular (between grains)and intercrystalline pores (between crystals) intraPpores include moldic pores and pores in fossil cav-ities Mudrocks contain pore types similar to thoseseen in carbonates and coarser grained siliciclasticsalthoughmuch smaller in scale However they alsocontain OM pores which have not yet been rec-ognized in carbonate or coarser grained siliciclasticrocks Intraparticle OM pores are created duringhydrocarbonmaturation (Jarvie et al 2007 Louckset al 2009) Loucks et al (2009) Ambrose et al(2010) and Curtis et al (2010) suggested and dem-onstrated that OM pores form the major connectedor effective pore network in some shale-gas sys-tems such as the Mississippian Barnett Shale

The following sections present a simple andobjective classification of mudrock matrixndashrelatedpores consisting of three basic types (1) interPpores between grains and crystals (2) intraP poreswithin mineral particles and (3) OM pores withinOM (Figure 1) A preliminary introduction to thesepore types was presented by Loucks et al (2010)

Interparticle Pores

Interparticle pores are abundant in young or shallow-buried sediments (Figure 5) and are commonlywell connected forming an effective (permeable)pore network However this pore network evolveswith burial as overburden stress and diagenesis

Figure 2 Common samplepreparation artifacts (A) Shrink-age of clay-rich matrix by desic-cation 10460 ft (3188 m) Penn-sylvanian Atoka interval AndrewsCounty Texas (B) Possibleshrinkage cracks within clay par-ticle by desiccation 10190 ft(3106 m) Pennsylvanian Atokainterval Andrews County Texas(C) Possible shrinkage of organicmatter showing elongated crackswith bridges across the cracks9700 ft (2957 m) Lower Creta-ceous Pearsall Formation Atasco-sa County Texas (D) Pores partlyfilled by milled material during theion-milling process Note thatlight-colored material is depositedon the sides of pores oriented to-ward the bottom of the photo-graph 13818 ft (4212 m) UpperCretaceous Eagle Ford Shale DeWitt County Texas (E) Differentialabrasion by ion beam producescurtains with sharp points orientedin an upflow direction of beam13384 ft (4079 m) Upper JurassicBossier Shale San AugustineCounty Texas (F) Differential re-lief striations formed by irregularion-bean current 1887 ft (575 m)Devonian New Albany ShaleHancock County Kentucky det =detector WD = working distancemag = magnification SE2 = sec-ondary electron detector TLD =through-lens detector HV = highvoltage (accelerating voltage)HFW = horizontal frame widthspot = spot size BSED = back-scattered electron detector EHT =

electron high tension (accelerat-ing voltage) TLD = through-lensdetector

Loucks et al 1079


increase At the time of deposition interP pores oc-cur between grains that range from soft and ductileto hard and rigid Examples of ductile grains in-clude clay flocculates peloids (micritic grains ofuncertain origin) fecal pellets and OM examplesof rigid grains include quartz feldspars authigenicpyrite and skeletal material During burial ductilegrains can distort and close interP pore space andplug pore throats

In young shallow-buried muds interP poreshave a variety of shapes (Figure 5) (also see Desboiset al 2009 their figures 3 4 Milliken and Reed2010 their figure 11) In an example of a relativelyshallow (11615 ft [354 m]) PliocenendashPleistocenemudrock sample shown in panels A to C of Figure 5interP pores are elongate and appear not to show astrong preferential orientation Some of the poresespecially the larger pores are concentrated aroundrigid grains (Figure 5AndashC) The long dimensions ofvisible pores range from about 30 nm to 2 mmwhich are similar to dimensions of interP mudrockpores shown by Milliken and Reed (2010) fromthe PleistoceneNankai accretionary prism In theirphotomicrograph (their figure 11) interP poreshapes range from elongated to rounded to angu-lar Many of the pores appear to be concentratedaround rigid grains Pores range in size from about100 nm to 3 mm In contrast the interP pores of

Table 2 Comparison of Pore Terminology from Other Publications with Terminology Suggested by the Present Classification

Pore Types (this article)

Terminology Used by Other Authors

Reference

Interparticle

Intergranular

Milliken and Reed (2010)

Type III large jagged


Phyllosilicate

Curtis et al (2010)

Phyllosilicate framework

Schieber (2010) Milner et al (2010)

Type IV occurs at face-to-edge arrangement of clay platestype V pores at interfaces between clay and rigid grains

Kwon et al (2004)

Intraparticle

Type I elongate type II crescent shape


Phyllosilicate

Curtis et al (2010)

Carbonate dissolution phyllosilicate framework


Type I cleavage-parallel pore type II cracks through severalgrains type III pore at terminations of stacked clay plateletstype VI intragranular fractures

Kwon et al (2004)

Intraparticle (primary in fecal pellets)

Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter


Figure 3 Pore-size classification for mudrock pores Classifi-cation is modified from the Choquette and Pray (1970) classifi-cation New pore classes include a picopore defined as being lessthan 1 nm and a nanopore defined as being equal to or greaterthan 1 nm and less than 1 mm Rouquerol et al (1994) pore-sizeclassification is also presented because this classification hasbeen suggested as a pore classification for mudrocks The sizes ofmethane and water molecules are shown for reference

Desbois et al (2009) from the Oligocene BoomClay are developed between slightly compactedclay platelets and are elongated roughly parallel tobedding except where they bend around a rigidgrain Their figure 4D shows a photomicrograph ofmudwith204porosity the pores having lengthsof about 35 nm to 2 mm

In older more deeply buried mudrocks interPpores are typically reduced in abundance by com-paction and cementation (Figures 6 7) Pores arecommonly scattered and show little to no preferen-tial orientation except in localized areas (Figure 6F)Many are triangular and are interpreted to be theremaining pore space between compacted and ce-mented rigid grains (Figures 6 7E F) Others oc-cur as linear pores interpreted to be remnant spacebetween larger clay platelets (Figure 7A) Mostpores have lengths in the 1-mm range but can rangefrom 50 nm to several micrometers Some interPpores may have developed where more ductilegrains bend around a rigid grain or preservedwhere a cluster of rigid grains formed a shelteringeffect inhibiting compaction of the ductile grains(Desbois et al 2009 Schieber 2010) The PliocenendashPleistocene mudstone in Figure 5 displays someexcellent examples of this phenomenon Interpar-ticle pores are not reduced or destroyed only bycompaction but also by cementation around grainssuch as quartz calcite and feldspar (Figure 6A)

Pore-filling cements include authigenic clay min-erals (Figure 6B) quartz and carbonate as well asless common albite and pyrite After extensive ce-mentation identification of individual grain bound-aries may be difficult

We use the term ldquointerparticlerdquo descriptivelyto categorize pores that occur between particlesand crystals The origin of these pores is varied andtheir geometries differ significantly as a function ofboth primary pore preservation and diagenetic al-teration Accordingly they commonly require a de-tailed study of their history so that their origin maybe deciphered and it is best not to use subjectiveterms requiring this knowledge to categorize themIn Figure 1A we diagrammatically illustrate someof themost common interP pore types Pores foundbetween particles are commonly triangular or elon-gate in cross section (Figure 6A B) Many of thesepores are primary in origin Grain-edge pores arefound along the edge of particles and are possiblyrelated tomatrix separating from the particle duringcompaction This is also commonly noted adjacentto large OM particles Interparticle pores are com-monly found between clay and mineral particles orbetween ductile clay and rigid particles (Figures 5AndashC 6F 8AndashD) Intercrystalline pores occur betweencrystals (Figure 6B) Interparticle pores also occurbetween clay aggregates that may be compactedclay flocculates (Figure 5)

Figure 4 Compositional diagram formudstones showing stability relationshipsbetween mineral end members

Loucks et al 1081

Intraparticle Pores

Intraparticle pores are found within particles(Figures 1A 8 9) most of these are intragranularbut intercrystalline pores are also locally commonSome of these pores are primary in origin althoughmost are probably diagenetic Examples include(1) moldic pores formed by partial or complete dis-solution (2) preserved intrafossil pores (3) intercrys-talline pores within pyrite framboids (4) cleavage-


plane pores within clay and mica-mineral grainsand (5) intragrain pores (ie within peloids and fe-cal pellets)

The relative abundance of intraP pores appearsto vary with age For example in young muds andmudrocks they can be very abundant In oldermudrocks they become more uncommon Exam-ples of intraP pores in young muds and mudrocksare (1) body-cavity pores in fauna and flora (2)cleavage-plane pores in clay particles (Figure 5)

Figure 5 Example of a relativelyshallow-buried (1162 ft [354 m])PliocenendashPleistocene clay-richmudstone from the Ursa Basin(Gulf of Mexico) Integrated OceanDrilling Program Expedition 308Site U1324B (Day-Stirrat et al2010b) (A) Silty clay-rich mud-stone with abundant interparticle(interP) and intraparticle (intraP)pores Some cleavage intraP poresare enhanced by bending relatedto compaction Day-Stirrat et al(2010b) reported this sample tohave 38 porosity A 2000-pointcount on this photomicrographgives values of 131 interP poresand 145 intraP pores for totalvisible point-count porosity of276 far less than the measuredporosity The difference is attrib-uted to pores too small to becounted in the photomicrograph(B) Similar to A (C) Interparticlepores are concentrated aroundthe rim of rigid grains (D) Intra-particle pores are located alongcleavage planes of clay particles(E) Some clay intraP pores areproduced during compaction bydistortion of clay palettes (F) Intra-particle pores in the low nano-meter size range det = detectorWD = working distance mag =magnification SE2 = secondaryelectron detector TLD = through-lens detector HV = high voltage(accelerating voltage) HFW =horizontal frame width spot =spot size BSED = backscatteredelectron detector

Figure 6 Example of interparticle (interP) pores within mudrocks (A) Interparticle pores between quartz (Qtz) and calcite grains withcement overgrowths The interP pores have straight edges and one pore has a triangular cross section Intraparticle (intraP) and organic-matter (OM) pores are also present 8427 ft (2569 m) vitrinite reflectance (Ro) = 15 Lower Cretaceous (lower Bexar Shale Member)Pearsall Formation Maverick County Texas (B) Sample contains intercrystalline-appearing interP pores Intraparticle pores are also presentin the center of the coccolith and along cleavage planes of distorted clay grains 6900 ft (2103 m) 09 Ro Upper Cretaceous Austin ChalkLa Salle County Texas (C) Common interP pores in a mixture of rigid calcite grains and compacted clay grains 11734 ft (3577 m) 18 RoLower Cretaceous (lower Bexar Shale Member) Pearsall Formation Maverick County Texas (D) Interparticle pore-dominated calcareousmudstone 13701 ft (4176 m) 15 Ro Upper Cretaceous Eagle Ford Shale Dewitt County Texas (E) Argillaceous calcareous mudstonehas interP pore network with pores characterized by straight boundaries 14381 ft (4383 m) 266 Ro Upper Jurassic Bossier ShaleSabine County Texas (F) Organic-rich argillaceous mudstone with abundant interP pores and some intraP pores The InterP pores have alarge range in size (nanometer to micrometer) and some appear to have formed by bending of clay particles Kerogen is immature and didnot form OM pores 1899 ft (579 m) 05 Ro Devonian New Albany Shale Hancock County Kentucky det = detector WD = workingdistance mag = magnification SE2 = secondary electron detector TLD = through-lens detector HV = high voltage (accelerating voltage)HFW = horizontal frame width spot = spot size BSED = backscattered electron detector EHT = electron high tension (accelerating voltage)

Loucks et al 1083

Figure 7 Example of interparticle (interP) pores within mudrocks (A) Linear interP pore along clay grains Organic-matter (OM) andintraparticle (intraP) pores present 10190 ft (3106 m) vitrinite reflectance (Ro) = 085 Pennsylvanian Atoka interval Andrews CountyTexas (B) Linear interP pores at contact between quartz crystals and calcite 10190 ft (3106 m) 085 Ro Pennsylvanian Atoka intervalAndrews County Texas (C) Linear interP pores parallel to edges of calcite crystals 7103 ft (2165 m) 145 Ro Posidonia Shale (LowerJurassic) Germany (D) Grain-edge interP pores along rigid clay-size grains Organic-matter pores present 8402 ft (2561 m) 15 RoLower Cretaceous (lower Bexar Shale Member) Pearsall Formation Maverick County Texas (E) Interparticle pores between framboidsof pyrite 1879 ft (573 m) 05 Ro Devonian New Albany Shale Hancock County Kentucky (F) Complex interP pore This relativelylarge mudstone pore appears to have collapsed after being partly occluded by clay and pyrite cement Cementation commonly breaks uplarger pores into smaller pores 10522 ft (3207 m) 116 Ro Pennsylvanian Atoka interval Midland County Texas det = detector WD =working distance mag = magnification SE2 = secondary electron detector TLD = through-lens detector HV = high voltage (acceleratingvoltage) HFW = horizontal frame width spot = spot size BSED = backscattered electron detector TLD = through-lens detector


Figure 8 Example of intraparticle (intraP) pores within mudrocks (A) Sample contains microfossils with interparticle (interP) body-cavitypores Interparticle and organic-matter (OM) pores are present 8470 ft (2582m) vitrinite reflectance (Ro) = 15 Lower Cretaceous (lowerBexar Shale Member) Pearsall Formation Maverick County Texas (B) Intercrystalline intraP pores within a pyrite framboid IntercrystallineinterP pores between calcite crystals have sharp triangular outlines 6900 ft (2103 m) 09 Ro Upper Cretaceous Austin Chalk La SalleCounty Texas (C) Cleavage-sheet intraP pores within a clay particle 11209 ft (3417 m) 13 Ro Upper Jurassic Haynesville FormationHarrison County Texas (D) Cleavage wedgendashshaped intraP pores along boundary of clay particle 10460 ft (3188 m) 089 RoPennsylvanian Atoka interval Andrews County Texas (E) Sample showing several grains containing cleavage-sheet intraP pores 13344 ft(4369 m) 124 Ro Upper Jurassic Bossier Shale San Augustine County Texas (F) Biotite flake containing cleavage-sheet intraP poresOrganic-matter pore present 8470 ft (2582 m) 15 Ro Lower Cretaceous (lower Bexar Shale Member) Pearsall Formation MaverickCounty Texas det = detector WD = working distance mag = magnification SE2 = secondary electron detector TLD = through-lensdetector HV = high voltage (accelerating voltage) HFW = horizontal frame width spot = spot size BSED = backscattered electron detector

Loucks et al 1085

and micas (3) pore spaces within clay flocculates(Figure 5) and (4) pore spaces within fecal pellets

Desbois et al (2009) showed excellent exam-ples of intraP pores within clay particles (theirfigures 3 and 4) of immature mudstones Becausethe clay particles in their mudstone example areoriented roughly parallel to bedding the sheet-


like pores tend to line up parallel to bedding ex-cept where they are bent around a rigid grain(their figure 3A) The shallow-buried PliocenendashPleistocene claystone presented in Figure 5 showsexcellent examples of intraP pores along distortedcleavage planes within clay particles Some of theintraP pores appear to be within compacted clay

Figure 9 Example of intrapar-ticle (intraP) pores within mud-rocks (A) Intraparticle poreswithin phosphate grain (pellet)8845 ft (2696 m) vitrinite reflec-tance (Ro) = approximately 12Lower Cretaceous (lower BexarShale Member) Pearsall Forma-tion Maverick County Texas(B) Fluid-inclusion intraP poreswithin large calcite grain 13344 ft(4369 m) 124 Ro Upper Ju-rassic Bossier Shale San AugustineCounty Texas (C) Dissolution-rim moldic pores after dissolvedfossils 15934 ft (4857 m) ap-proximately 18 Ro Lower Cre-taceous (Pine Island Shale Mem-ber) Pearsall Formation BeeCounty Texas (D) Dissolutionpores after dolomite crystals andfossils Interparticle (interP) porespresent 15934 ft (4857 m) ap-proximately 18 Ro Lower Cre-taceous (Pine Island Shale Mem-ber) Pearsall Formation BeeCounty Texas (E) Dissolutionpores in feldspar-silt grain alongtwin planes 2400 ft (732 m)approximately 06 Ro Mississip-pian Barnett Shale Wise CountyTexas (F) Dissolution-rim andmoldic pores in dolomite crystals1879 ft (573 m) 05 Ro Devo-nian New Albany Shale HancockCounty Kentucky OM = organicmatter det = detector WD =working distance mag = magnifi-cation SE2 = secondary electrondetector TLD = through-lens de-tector HV = high voltage (accel-erating voltage) HFW = horizontalframe width spot = spot size

BSED = backscattered electron de-tector TLD = through-lens detector

particles that were composed of loosely aggregatedclay platelets (Figure 5D) The intraP pores rangein size from 10 nm (Figure 5F) to 1 mm (Figure 5A)

In older mudrocks many of the early formedintraP pores are absent presumably destroyed bymechanical compaction or they have been filled inwith cement However other intraP pores (disso-lution intraP pores) develop in the subsurface bycorrosive fluids Milner et al (2010) and Schieber(2010) also recognized intraP dissolution poresThis dissolution process is addressed in the Dis-cussion section The shape of an intraP pore iscommonly controlled by its origin In clay particlesand micas pores are sheetlike (Figure 8CndashF) infossils pores are controlled by the shape of thebody cavity (Figure 8A) and in grain and crystaldissolution moldic pores the pores can take theshape of the precursor (Figure 9CndashF) As withother pore types intraP pores can contain cement(Figure 8A C) that occludes pore space

Common types of intraP pores are shown inFigure 1A Pyrite intercrystalline pores betweenpyrite crystals are locally common in pyrite fram-boids (Figure 8B) Many of these pores are oc-cluded with OM or with authigenic clay plateletsIntraparticle pores within clay sheets that may beflocculates are generally linear and are parallel toone another (Figure 8CndashE) Similar pores are seenin micas (Figure 8F) Intraparticle pores alongcrystal rims or grains are most likely formed bypartial dissolution of carbonate crystals and grains(Figure 9F) Crystal-mold pores are related porescreated by the complete dissolution of a crystal(Figure 9D) Dissolution pores also occur in feld-spar grains (Figure 9E) Body-cavity pores includeprimary pore chambers within faunal and floralfossil grains (Figures 6A 8A) Fossil molds are theproduct of diagenetic dissolution of fossil skeletons(Figure 9C D) Figure 9A shows pores within apossible fecal pellet that are similar to pores de-scribed by Milner et al (2010) Figure 9B showsfluid-inclusion pores within a crystal

Organic-Matter Intraparticle Pores

Organic-matter pores (Figures 1A 10 11) are intraPpores that are found within OMWe have observed

that thermal maturation has to reach an Ro level ofapproximately 06 or higher before OM poresbegin to developwhich according toDow (1977)is the beginning of peak oil generation At levels ofmaturity less than 06 Ro OM pores are absentor extremely rare (Figure 11C) Reed and Loucks(2007) Ruppel and Loucks (2008) and Louckset al (2009) were the first to describe theOMporesand associated pore network in the Barnett mud-stone play in the FortWorth Basin Since then otherstudies have also documented the abundance of thispore type in Barnett mudstones (ie Sondergeldet al 2010) Ambrose et al (2010) and Sondergeldet al (2010) used SEMndashfocused ion beam (FIB)analysis to demonstrate that these OM pores formeda connected 3-D effective pore network through thetouching of grains a relationship previously sug-gested by Loucks et al (2009) The abundance ofthis pore type is nowwell established inmost othermudrock systems by the present authors and otherresearchers (Table 1) including the Eagle FordPearsall Kimmeridge Floyd Woodford MarcellusHorn River and Maquoketa systems

Organic-matter pores have irregular bubble-like elliptical cross sections and generally rangebetween 5 and 750 nm in length They commonlyappear isolated in two dimensions but in threedimensions they display connectivity (Figure 10BC) which has now been illustrated using SEM-FIBanalysis (Ambrose et al 2010 Sondergeld et al2010) The amount of porosity within an OM par-ticle in a single sample ranges from0 to 40 (Louckset al 2009 this study) The large piece of OM inFigure 10A has 41 porosity (by point-countmethod) A broad range can even occur within asingle OM grain (Figure 11D) Curtis et al (2010)identified organic particles having as much as 50porosity SomeOM has an inherited structure thatcontrols the development and distribution of thepores within a grain (Figure 11A B)

Not all OM types appear to be prone to thedevelopment of OM pores Limited data collectedduring this study suggest that type II kerogenmay be more prone to the development of OMpores than type III kerogen The organic particle inFigure 11E on the basis of appearance and Rock-Eval data is interpreted as type III kerogen from

Loucks et al 1087

Figure 10 Example of organic-matter (OM) pores within mudrocks (A) Large OM particle with OM pores On the basis of pointcounting 1000 points this grain has a porosity value of 41 7625 ft (2324 m) vitrinite reflectance (Ro) = approximately 16 Mis-sissippian Barnett Shale Wise County Texas (B) Organic-matter pores slightly aligned and showing complexity in third dimension 7625 ft(2324 m) approximately 16 Ro Mississippian Barnett Shale Wise County Texas (C) Organic-matter pores showing range in pore sizefrom 10 to 300 nm 8402 ft (2561 m) 15 Ro Lower Cretaceous (lower Bexar Shale Member) Pearsall Formation Maverick CountyTexas (D) Compacted OM with a range of pore sizes 8470 ft (2582 m) 15 Ro Lower Cretaceous (lower Bexar Shale Member) PearsallFormation Maverick County Texas (E) Scattered OM containing numerous OM pores 7138 ft (2176 m) approximately 135 Ro Mis-sissippian Barnett Shale Wise County Texas (F) Large organic flake with numerous pores Original ductile particle shows flowage betweenrigid grains 12923 ft (3939 m) 131 Ro Mississippian Barnett Shale Pecos County Texas InterP = interparticle intraP = intraparticle det =detector WD = working distance mag = magnification SE2 = secondary electron detector TLD = through-lens detector HV = high voltage(accelerating voltage) HFW = horizontal frame width spot = spot size BSED = backscattered electron detector


Figure 11 Example of organic-matter (OM) pores within mudrocks (A) Organic particle with relatively large OM pores showingalignment The original particle had inherent variation in its internal structure 7625 ft (2324 m) vitrinite reflectance (Ro) = approximately16 Mississippian Barnett Shale Wise County Texas (B) Organic-matter pores showing alignment controlled by original structure ofOM 8545 ft (2605 m) 317 Ro Mississippian Barnett Shale Hill County Texas (C) Immature OM in shallow-buried Barnett Shaleshowing no onset of OM pore development 647 ft (197 m) less than 05 Ro Mississippian Barnett Shale Lampasas County Texas(D) A large piece of kerogen showing areas of OM pore development and other areas free of OM pores This examples shows elongategrain-edge pores that may be associated with separation of the grain from thematrix during compaction 15934 ft (4857 m) approximately18 Ro Lower Cretaceous (Pine Island Shale Member) Pearsall Formation Bee County Texas (E) Woody-type kerogen showing nodevelopment of OM pores This type of kerogen may not be prone to OM pore development 10460 ft (3188 m) 089 Ro Penn-sylvanian Atoka interval Andrews County Texas (F) Interparticle (interP) pore (arrow) may have been originally filled with oil A possibleshrinkage bubble formed in the center of the pore after drying Pores associated with residue oil and could be mistaken for kerogen-typeOM pores 9700 ft (2957 m) 078 Ro Lower Cretaceous (Pine Island Shale Member) Pearsall Formation Atascosa County Texas det =detector WD = working distance mag = magnification SE2 = secondary electron detector TLD = through-lens detector HV = high voltage(accelerating voltage) HFW = horizontal frame width spot = spot size BSED = backscattered electron detector EHT = electron high tension(accelerating voltage) TLD = through-lens detector

Loucks et al 1089

the Atoka in the Midland Basin and it shows nodevelopment of OM pores despite having an Ro of089 No OM pores were observed in this unitSchieber (2010) also noted that the type of OMmay control OM pore formation in thermally ma-ture rocks

Note that imaging pores with residual oil canlead to the mistaken identification of OM poresFigure 11F shows what we think may be residualoil within an interP pore The darker material inthe pore is some form of organic material Becausethe bubble-shaped pore is in the center it appearsto be a desiccation pore resulting from oil in thepore drying out and therefore not an OM pore

Comparisons with Previous Pore Naming andClassification Schemes

Previous authors have used a variety of both de-scriptive and interpretive names to characterizemudrock pores Essentially all of these pore typescan readily fit into our mudrock pore classificationscheme (Tables 1 2) Many of the names giventhese pores by previous authors are descriptiveand thus useful for describing specific pore sub-typeswithin the threefold systemwe propose hereHowever we think that interpretive names (egorganophyllic organophobic carbonate dissolutionpores primary secondary) although useful whensupported by pore genesis data should be avoidedin any pore classification scheme

Fracture-Related Pores

Fracture-related pores are not part of the matrix-related pore classification presented in this articlehowever these nonmatrix pores can be very im-portant in shale-gas systems where they exist andare not completely cemented Where natural frac-tures are present in mudrock reservoirs they canhave a significant effect on hydrocarbon produc-tion (Eichhubl and Boles 1998 Pitman et al 2001Clarke 2007 Gale and Holder 2010) Some mud-rock reservoirs have fractures that are cemented andimpermeable Nevertheless they can still have astrong influence on induced fracture propagation(Gale et al 2007) Pearsallmudstones have opening-


mode natural fractures that contain pores betweencalcite cement fill (Figure 12A) Early successfulwells in the Pearsall mudstones have been attrib-uted to open fractures (Clarke 2007) Open micro-fractures have also been thought to be present inmudrocks (Kanitpanyacharoen et al 2011 theirfigures 4 5) and as a storage and transport mech-anism for hydrocarbons inmudrocks (egDewhurstet al 1999) However no quantity of open micro-fractures has been found in the hundreds of samplesthat we have investigated using thin-section andSEM-based imaging One cementedmicrofracturehas been noted in the Barnett Shale (Loucks et al2009) and one partly cementedmicrofracture wasidentified in an Atokan mudstone (Figure 12B)Although large permeable natural fractures maybe present in some mudrocks abundant openmicrofractures have yet to be well documented atthe thin-section or SEM scale by this study of hun-dreds of samples

MUDROCK PORE NETWORKS ANDASSOCIATED PORE ABUNDANCEAND DISTRIBUTION

Mudrocks may contain one dominant pore type ora combination of pore types Understanding theabundance distribution and spatial interrelation-ships of each pore type is important because eachtype may contribute differently to permeabilityleading to the need to understand the distributionof pores and establish their connectivity Mudrockreservoirs have inherently low permeability andrequire fracture stimulation to improve hydrocar-bon flow We must therefore recognize the per-meability pathways within the mudrock matrix toinduced fractures

To address the questions of pore distributionand connectivity we point counted pores of ion-milled samples from four typical mudrocks usingphotomicrographs takenon theSEMThree samplescome from major producing mudstone reservoirsystems (Barnett Pearsall and Bossier) one comesfrom a shallow young nonproductive PliocenendashPleistocene claystone (Figures 5A 13) Proportions

of pore types are plotted on the ternary diagram inFigure 1B

The PliocenendashPleistocene sample (Figure 5A)is a silty claystone with an estimated Ro of less than03 Visible porosity on the basis of 1000 pointcounts is 276 Pores are nearly equally dividedbetween interP (475) and intraP (525) Ac-cording to Day-Stirrat et al (2010b) this samplehas a measured porosity of 38 (measured by mer-cury injection capillary pressure analysis) whichis higher than the visible point-count porosity Thedifference between these two measurements islikely the result of pores being too small to bepoint counted on the photomicrograph

TheBarnett sample (Figure 13A) is a calcareoussiliceous mudstone with abundant OM (original to-tal organic content by point count [volume percent]was sim124) and about 135 Ro Visible porosityon the basis of 2000 point counts is 42 Pores aredominantly within OM (952) with some intraPpores (48) Trace amounts of interP pores are alsopresent Most of the OM contains well-developedpores and it displays extensive compaction anddeformation around rigid grains

Although similar in maturity (sim124Ro) theBossier sample an argillaceous calcareousmudstone(Figure 13B) differs significantly in visible porosityand pore abundance from the Barnett samplePorosity on the basis of 2000 point counts is 72Intraparticle pores are most abundant (755)followed by interP (196) and OM pores (49)Most of the intraP pores are associated with dis-

torted cleavage planes inmicas and clay-rich grainsSome of the interP pores are grain-edge pores

The Pearsall sample (Figure 13C) is a siliceouscalcareousmudstone with 15Ro Porosity on thebasis of 2000 point counts is only 18 Most poresare interP (694) with the rest beingmostly intraP(306) Organic-matter pores are uncommonSome of the interP pores are grain-edge pores Intra-particle pores occur within micas or as fluid inclu-sions in calcite grains

These four mudstone examples illustrate thebroad variety of pore networks that can exist inmudrocks depending on age mineralogy and OMtype and content Plotting themon the ternary poreclassification (Figure 1) provides a quick compar-ison of the pore networks

DISCUSSION

Pore Types and Effective Pore Networks

That the permeability of carbonate and coarse-grained siliciclastic pore networks are related to poretypes iswell known(ie Pittman 1979Lucia 1999)Interparticle pores are generally better connectedto one another than intraP pores Therefore higherpermeability is generally related to interP pore net-works than to intraP pore networks Intraparticlepores are commonly connected to the overall poresystem but associated pore throats may be smallerand relatively fewer than those associated with

Figure 12 Examples of frac-ture pores (A) Opening-modefracture partly filled with calcitecement Some fracture poresopen 8524 ft (2598 m) LowerCretaceous (lower Bexar ShaleMember) Pearsall FormationMaverick County Texas (B) Ex-ample of a rare natural micro-fracture approximately 30 mmlong 01 mm wide and partlyhealed 10140 ft (3091 m)Pennsylvanian Atoka intervalAndrews County Texas

Loucks et al 1091

Figure 13 Examples of mudrock-pore networks Percent matrix total organic content (TOC) organic-matter (OM) pores interparticle(interP) pores and intraparticle (intraP) pores calculated by point counting 2000 points (A) Organic-rich calcareous siliceous mudstone withOM porendashdominated pore network Present TOC = 84 and original TOC = 124 Total point-count porosity = 42 of which OM pores =952 interP pores = 0 and intraP pores = 48 of pore volume 7111 ft (2167 m) vitrinite reflectance (Ro) = approximately 135Mississippian Barnett Shale Wise County Texas (B) Siliceous calcareous mudstone with intraP porendashdominated pore network PresentTOC = 09 and original TOC = 13 Total point-count porosity = 72 of which OM pores = 49 interP pores = 196 and intraPpores = 755 of pore volume 13344 ft (4067 m) 124 Ro Upper Jurassic Bossier Shale San Augustine County Texas (C) Siliceouscalcareous mudstone with interP porendashdominated pore network Present TOC = 03 Total point-count porosity = 18 of which OMpores = 0 interP pores = 694 and intraP pores = 306 of pore volume 8427 ft (2569 m) approximately 15 Ro Lower Cretaceouslower Bexar Shale Member of the Pearsall Formation Maverick County Texas det = detector WD = working distance mag = magni-fication SE2 = secondary electron detector TLD = through-lens detector HV = high voltage (accelerating voltage) HFW = horizontal framewidth spot = spot size BSED = backscattered electron detector ETD = Everhart-Thomley detector


interP pores (McCreesh et al 1991) Some intraPpores such as moldic pores may be nearly totallycut off from the effective pore network We sug-gest that the same principles of pore connectivelyapply to pores in mudrocks Interparticle pores con-tribute more to the effective pore network than dointraP pores

The relative connectivity of OM pores a formof intraP pores do not parallel intraP pores in themineral matrix As previously noted several stud-ies have shown that OM pores in a continuous or-ganic framework can form an effective pore systemand can form the dominant flow pathway (Louckset al 2009Ambrose et al 2010Curtis et al 2010)

Effect of Mineralogy and Fabric on PoreDevelopment and Preservation

Figure 4 summarizes the importance of mineral-ogical composition on the development and pres-ervation of mudrock pores This ternary diagramgroupsmajormudrockminerals bymechanical andchemical stability The apices are (1) phyllosilicateclay (2) silica +pyrite and (3) carbonate + feldspar +phosphate Each apex reflects different mechanicaland chemical stability parameters during diagenesisClays are ductile and compact and deform easilymaking them mechanically unstable Many typesof clay especially smectite are chemically unstableand will transform to mixed-layer clays and illitewith increasing temperature associated with burial(ie Pytte and Reynolds 1988) Silica + pyrite arerigid grains that resist compaction and act as focalpoints where more ductile grains such as clay andOM compact around them They are also relativelychemically stable and generally do not undergodissolution Carbonate + feldspar + phosphate arerigid grains and resist compaction They can alsoform focal points where more ductile grains willbend and deformHowever they can be chemicallyunstable and undergo dissolution forming intraPmoldic pores

The above three end members are importantcontrols on mudrock pore types but the additionof reactive kerogen is an additional parameter thatcan greatly affect the pore network system Givenabundant kerogen a continuous pore networkmay

develop such as seen in the Barnett shale-gas sys-tem (Loucks et al 2009 Curtis et al 2010)

Several authors have noted that admixtures ofrigid and ductile grains can affect pore develop-ment and preservation (Dawson andAlmon 2002Desbois et al 2009 Schieber 2010 this study)Desbois et al (2009) described how clay sheetsfolded around rigid grains can form crescent-shapedpores Schieber (2010) recognized that pores can bepreserved in the compaction-protected shadowadjacent to rigid grains The PliocenendashPleistocenesample shown in Figure 5 demonstrates the inter-action between ductile and rigid grainsMicrometer-size interP pores developed around the larger rigidgrains where ductile clay grains were distortedSimilar pore types are seen in older more deeplyburiedmudrocks (Figure 6F) Also observed in thissample are the bending of clay grains and the as-sociated development of intraP sheetlike pores

Evolution of Mudrock Pores through Timeand with Burial

It has beenwell established that as muds are buriedand exposed over time to higher temperatures andpressures they decrease in volume and becomelithified Initial porosities of muds range from 60 to80 at the time of deposition (Bridge andDemicco2008) and these porosities drop to as low as a fewpercent in the subsurface As an example of po-rosities in producing shale-gas reservoirs TXCO(2009) noted porosity ranges in several shale-gassystems (1) Barnett Shale from 40 to 96 (2)Haynesville Formation from 8 to 15 (3) Fayet-teville Shale from 40 to 50 (4) Pearsall For-mation from 60 to 115 and (5) Eagle Ford from34 to 146No informationwas provided onhowthe porosity ranges were calculated The burialdiagenesis and transformation ofmuds tomudrocksis a complex process composed of many variablesand controls including original mineralogy fabrictexture organic content fluids hydrology andrate and depth of burial Figure 14 illustrates someof the major processes involved in the evolution ofmud to mudrocks

Perhaps the most significant process in muddiagenesis is mechanical compaction Bridge and

Loucks et al 1093

Demicco (2008) noted that the fastest rates ofcompaction occur in the first kilometer (06 mi) ofburial which is before major mineralogical diagen-esis starts except for limited carbonate phosphateand early pyrite diagenesis This rapid rate of com-paction has been documented bymany studies (ieRieke and Chilingarian 1974 Fruehn et al 1997Van Sickel et al 2004) Most compaction occursbefore the initiation of any significant volume ofhydrocarbon generation Bridge andDemicco (2008)also noted that major large-scale mineral transfor-mations do not begin until temperatures of ap-proximately 100degC are reached Therefore dur-ing much of the early history of burial interP andintraP pores are primarily lost through compactionDuring this time ductile grains are distorted andflow into interP pores between rigid grains addingto the loss of interP pore space

Some early cementation can occur in near-surface settings in the form of carbonate (ieRaiswell 1976) pyrite (ie Wilkin et al 1996


1997) and phosphate (ie Foumlllmi 1996) miner-als These diagenetic products commonly form asconcretions and hardgrounds (ie Weeks 1957)and effectminimum volumes of sediment Loucksand Ruppel (2007) recognized each of these earlydiagenetic products in the Barnett Shale Sili-ceous ooze (ie diatoms radiolarians silicoflag-ellates and sponges) with an opal mineralogy willundergo a transformation through opal A to opalCT to cryptocrystalline quartz (ie Kastner et al1977 Williams and Crerar 1985) This solution-redepositionmechanismmay simply transform thesiliceous particles in place or reprecipitate-dissolvedsilica in pores Work by Matheney and Knauth(1993) based on isotope evidence from theMioceneMonterey Formation stated that it is a relativelylow-temperature process starting at 17 to 21degC

The process of diagenetic transformation ofsmectite to illite during burial is well known (ieBoles and Franks 1979) During this process ele-ments can be released and volume changes occur

Figure 14 Generalized parasequencesummary diagram of the major stages inmudrock burial diagenesis and the rela-tionship of these stages to the evolution ofpore types and abundances in mudrocksThis sequence of diagenetic events does notreflect any particular case history of a spe-cific mudrock

(Day-Stirrat et al 2010a c) Silica calcium andothers can be released from the clays forming newauthigenic clays andminerals (Hayes 1991 Surdamet al 1991 Land and Milliken 2000 Day-Stirratet al 2010c Schieber 2010) that occlude porespace Figure 14 suggests the relative timing of thistransformation event

Dissolution of carbonates (both aragoniteand calcite) and feldspars are noted in mudrocksAragonite is unstable with burial and dissolvesSchieber (2010) suggested that fluids for the dis-solution process in mudrocks are the fluids asso-ciated with decarboxylation of kerogen whichproduces an increase of carboxylic and phenolicacids He suggested that this occurred in the tem-perature range of 80 to 120degC following the con-cept developedbyMacGowan and Surdam (1990)These late aggressive fluids could dissolve calciteand feldspar The dissolved material is reprecipi-tated in the mudrock unless it migrates out of thesystem

General pore evolution history in mudrocksrelated to burial diagenesis is summarized inFigure 14 Given our review of relatively youngmudrocks (ie Figure 5) pores at and in the shal-low subsurface are interP and intraP During initialburial to several kilometers interP and intraP arecompacted and affected by chemical diagenesisand the volume of interP and intraP pores greatlydecreases Porosity versus depth curves (Fruehnet al 1997 Van Sickel et al 2004) for mud-dominated strata show that porosity can drop toless than 10 by 25 km (16 mi) of burial sug-gesting that 50 to 70 porosity units or 83 to 88volume is lost by compactional and diageneticprocesses Cementation associated with burial isvariable It can take the form of precipitation ofclays in pores (Figure 8D) cement overgrowths ongrains by quartz carbonate or feldspar (Figure 7A)or as pore cement fill As already discussed abun-dant pores can be created in OM undergoing ther-mal maturation Organic-matter pore developmentstarted with the onset of organic thermal matura-tion but we found no linear correlation with in-creasedmaturation This observation is based on ananalysis of numerous Barnett Shale samples cov-ering a range of Ro values (05ndash317 Ro) The

final pore network in a mudrock can be any com-bination of pore types

SUMMARY AND CONCLUSIONS

A mudrock matrixndashrelated pore classification is pre-sented that can be used to categorize different poretypes into three general classesmineralmatrix interPpores that developed between grains and crystalsmineralmatrix intraPpores that are containedwithina particle boundary and intraP OM pores Youngshallow-buried muds contain only interP and intraPpores that are greatly diminished during burial andcompactionWith deeper burialOMpores developduring hydrocarbon thermal maturation and intraPdissolution pores may develop by acidic fluid gen-erated during decarboxylation of kerogen

Mudrocks have a variety of pore networks thatcan be any combination of pore types Develop-ment of the pore network depends on originalmineralogy fabric and texture which can vary atthe millimeter lamination level The OM contentand type are important factors in the developmentand abundance of OM pores Pores of all types ap-pear to be able to stay open into the deep subsur-face (gt15000 ft [gt4570 m])

The practical function of this pore classifica-tion is its application in characterizing pore net-works in different rock types for reservoir charac-terization Pore-type categories were developed tobe relatively simple and objective while carryingsome implications for permeability Interparticleand OM pores have been demonstrated to havegenerally better connectivity than intraP pores Thelatter pores will provide storage and some perme-ability but will not have the same level of con-nectively as the other pore types This classificationof pore types depends on imaging of pores at theSEM level Multiple two-dimensional images areneeded to characterize each rock type quantita-tively for pore networks and several 3-D imageanalyses are necessary to substantiate that the poresare connected and form an effective pore network

The major goal of imaging is to quantify andcapture the strong heterogeneity observed at thesubmillimeter scale in mudrocks A database of

Loucks et al 1095

combining pore-network types with mudrock typesmight show that certain mudrock types producesimilar pore networks Characterization of eachmudrock type relative to its Ro level would alsobe essential

The mudrock matrixndashrelated pore classifica-tion presented here offers a simple but robust sys-tem for categorizing pores on the basis of their re-lationships with grains Although pore types varywidely in origin shape and size all pores can beassigned to one of the three categories defined hereby visual examination at the SEM scale Becausethe interrelationships between grains and pores thatform the basis for this classification scheme relatedirectly to permeability grouping of pores using thissystem may be a future aid in modeling and pre-dicting reservoir quality relative to porosity andpermeability However much more research anddata collection are needed from a broad variety ofpore networks fromdifferentmineral compositionsand at different levels of maturity These data needto be related to measured porosity and permeabil-ity analyses We must also more fully understandthe evolution of pore networks relative to increasesin time temperature and pressure before reliablereservoir-quality prediction can be possible An-other application might be using this pore classifi-cation for defining general nanopore to microporenetwork types that can then be modeled for hy-drocarbon flow The types of pores within the net-work are a major controlling factor for storagepermeability and wettability

REFERENCES CITED

Ambrose R J R C HartmanMDiaz-Campos I Y Akkutluand C H Sondergeld 2010 New pore-scale consid-erations for shale gas in place calculations Society of Pe-troleum Engineers Unconventional Gas ConferencePittsburgh Pennsylvania February 23ndash25 2010 SPEPaper 131772 17 p

Boles J R and S G Franks 1979 Clay diagenesis inWilcoxsandstones of southwest Texas Implication of smectitediagenesis on sandstone cementation Journal of Sedi-mentary Research v 49 p 55ndash70

Bridge J S and R V Demicco 2008 Earth surface pro-cesses landforms and sediment deposits New YorkCambridge University Press 830 p

Chalmers G R M Bustin and I Powers 2009 A pore by


any other name would be as small The importance ofmeso- and microporosity in shale gas capacity (abs)AAPG Search and Discovery article 90090 1 p httpwwwsearchanddiscoverycomabstractshtml2009annualabstractschalmershtm (accessed March 142011)

Choquette P W and L C Pray 1970 Geologic nomencla-ture and classification of porosity in sedimentary carbon-ates AAPG Bulletin v 54 p 207ndash244

Clarke R 2007 Basin focus Maverick Basin Oil and GasInvestor v 27 p 87ndash90

Curtis M E R J Ambrose C H Sondergeld and C S Rai2010 Structural characterization of gas shales on themicro- and nano-scales Canadian Unconventional Re-sources and International Petroleum Conference Cal-gary Alberta Canada October 19ndash21 2010 SPE Pa-per 137693 15 p

Davies D K W R Bryant R K Vessell and P J Burkett1991 Porosities permeabilities and microfabrics of De-vonian shales in R H Bennett W R Bryant and M HHulbert eds Microstructure of fine-grained sedimentsFrom mud to shale New York Springer-Verlag p 109ndash119

Dawson W and W R Almon 2002 Top seal potential ofTertiary deep-water shales Gulf of Mexico Gulf CoastAssociation of Geological Societies Transactions v 52p 167ndash176

Day-Stirrat R J S P Dutton K L Milliken R G LoucksA C Aplin S Hillier and B E van der Pluijm 2010aFabric anisotropy induced by primary depositional varia-tions in the silt Clay ratio in two fine-grained slope fancomplexes Texas Gulf Coast and northern North SeaSedimentary Geology v 226 p 42ndash53 doi101016jsedgeo201002007

Day-Stirrat R J P B Flemings A C Aplin and A MSchleicher 2010b The fabric and compaction of mud-stones in the Gulf of Mexico (abs) American Geophys-ical Union Fall Meeting San Francisco December 13ndash172010 Abstract H43A-1213 1 p

Day-Stirrat R J K Milliken S P Dutton R G Loucks SHillier A C Aplin and AM Schleicher 2010c Open-system chemical behavior in deep Wilcox Group mud-stones Texas Gulf Coast USA Marine and PetroleumGeology v 27 p 1804ndash1818 doi101016jmarpetgeo201008006

Desbois G J L Urai and P A Kukla 2009 Morphology ofthe pore space in claystonesmdashEvidence from BIBFIBion beam sectioning and cryo-SEM observations Earthv 4 p 15ndash22 doi101051epjconf2010062205

Dewhurst D N Y Yang and A C Aplin 1999 Permeabil-ity and fluid flow in natural mudstones in A C AplinA J Fleet and J H S Macquaker eds Muds andmudstones Physical and fluid flow properties Geo-logical Society (London) Special Publication 158 p 23ndash43

Dewhurst D N RM Jones andMD Raven 2002Micro-structural and petrophysical characterization of MuderongShale Application to top seal risking Petroleum Geo-science v 8 p 371ndash383 doi101144petgeo84371

Diaz E C Sisk and A Nur 2010 Quantifying and linking

shale properties at a variable scale 44th US Rock Me-chanics Symposium and 5th US-Canada Rock Me-chanics Symposium Salt Lake City Utah June 27ndash302010 American Rock Mechanics Association Paper 10-272 3 p

Dow W G 1977 Kerogen studies and geological interpre-tations Journal of Geochemical Exploration v 7 p 79ndash99 doi1010160375-6742(77)90078-4

Dutton S P and R G Loucks 2010 Diagenetic controls onevolution of porosity and permeability in lower TertiaryWilcox sandstones from shallow to ultradeep (200ndash6700 m) burial Gulf of Mexico Basin USA Marineand Petroleum Geology v 27 p 69ndash81 doi101016jmarpetgeo200908008

Eichhubl P and J R Boles 1998 Vein formation in relationto burial diagenesis in theMioceneMonterey FormationArroyo Burro Beach Santa Barbara California in PEichhubl ed Diagenesis deformation and fluid flowin theMioceneMonterey Formation Pacific Section SEPMSpecial Publication 83 p 15ndash36

Foumlllmi K B 1996 The phosphorus cycle phosphogenesisand marine phosphate-rich deposits Earth-Science Re-views v 40 p 55ndash124 doi1010160012-8252(95)00049-6

Fruehn J E S White and T A Minshull 1997 Internaldeformation and compaction of the Makran accretionarywedge Terra Nova v 9 p 101ndash104

Gale J F W and J Holder 2010 Natural fractures in someUS shales and their importance for gas production Pe-troleum Geology Conference Series 7 p 1131ndash1140

Gale J F W R M Reed and J Holder 2007 Natural frac-tures in theBarnett Shale and their importance for hydrau-lic fracture treatments AAPG Bulletin v 91 p 603ndash622 doi10130611010606061

Hayes J B 1991 Porosity evolution of sandstones related tovitrinite reflectance Organic Geochemistry v 17 p 117ndash129

Hildenbrand A and J L Urai 2003 Investigation of the mor-phology of pore space in mudstones First results Marineand Petroleum Geology v 20 p 1185ndash1200

Hover V C D R Peacor and L M Walter 1996 STEMAEM evidence for preservation of burial diagenetic fab-rics in Devonian shales Journal of Sedimentary Researchv 66 p 519ndash530 doi101306D4268397-2B26-11D7-8648000102C1865D

Jarvie DM R J Hill T E Ruble and EM Pollastro 2007Unconventional shale-gas systems TheMississippian Bar-nett Shale of north-central Texas as one model for ther-mogenic shale-gas assessment AAPG Bulletin v 91p 475ndash499 doi10130612190606068

Javadpour F 2009 Nanopores and apparent permeabilityof gas flow in mudrocks (shales and siltstone) Journalof Canadian Petroleum Technology v 48 p 16ndash21

Kanitpanyacharoen W W Hans-Rudolf F Kets C Lehrand R Wirth 2011 Texture and anisotropy analysis ofQusaiba shales Geophysical Prospecting v 59 p 536ndash556

Kastner M J B Keene and J M Gieskes 1977 Diagenesisof siliceous oozes I Chemical controls on the rate ofopal-A to opal-CT transformationmdashAn experimental

study Geochimica et Cosmochimica Acta v 41 p 1041ndash1059

Kwon O A K Kronenberg A F Gangi B Johnson andB E Herbert 2004 Permeability of illite-bearing shale1 Anisotropy and effects of clay content and loading Jour-nal of Geophysical Research v 109 no B10205 19 pdoi1010292004JB003052

Land L S and K L Milliken 2000 Regional loss of SiO2

and CaCO3 and gain of K2O during burial diagenesis ofGulf Coast mudrocks USA in R H Worden and SMorad eds Quartz cementation in sandstones OxfordBlackwell Science International Association of Sedimen-tologists Special Publication 29 p 183ndash198

Loucks R G and S C Ruppel 2007 Mississippian BarnettShale Lithofacies and depositional setting of a deep-water shale-gas succession in the Fort Worth BasinTexas AAPG Bulletin v 91 p 579ndash601 doi10130611020606059

Loucks RG RMReed SCRuppel andDM Jarvie 2009Morphology genesis and distribution of nanometer-scalepores in siliceous mudstones of the Mississippian BarnettShale Journal of Sedimentary Research v 79 p 848ndash861 doi102110jsr2009092

Loucks R G R M Reed S C Ruppel S C and UHammes 2010 Preliminary classification of matrix poresin mudrocks Gulf Coast Association of Geological Socie-ties Transactions v 60 p 435ndash441

Lu J K L Milliken R M Reed and S Hovorka 2011 Dia-genesis and sealing capacity of the middle Tuscaloosamudstone at the Cranfield carbon dioxide injection siteMississippi USA Environmental Geosciences v 18p 35ndash53 doi101306eg09091010015

Lucia F J 1999 Carbonate reservoir characterization NewYork Springer-Verlag 226 p

MacGowan D B and R C Surdam 1990 Carboxylic acidanions in formation waters San Joaquin Basin and Louisi-anaGulfCoastUSA Implications for clastic diagenesisApplied Geochemistry v 5 p 687ndash701

Matheney R K and L P Knauth 1993 New isotopic tem-perature estimates for early silica diagenesis in beddedcherts Geology v 21 p 519ndash522 doi1011300091-7613(1993)021lt0519NITEFEgt23CO2

McCreesh C A R Erlich and S J Crabtree 1991 Petrogra-phy and reservoir physics II Relating thin section porosityto capillary pressure the association between pore typesand throat size AAPG Bulletin v 75 p 1563ndash1578

Milliken K L and R M Reed 2010 Multiple causes of dia-genetic fabric anisotropy in weakly consolidated mudNankai accretionary prism IOPD Expedition 316 Jour-nal of Structural Geology v 32 p 1887ndash1898 doi101016jjsg201003008

Milner M R McLin and J Petriello 2010 Imaging textureand porosity in mudstones and shales Comparison ofsecondary and ion milled backscatter SEMmethods Ca-nadian Unconventional Resources and International Pe-troleum Conference Calgary Alberta Canada October19ndash21 2010 Canadian Society for Unconventional GasSociety of Petroleum Engineers Paper 138975 10 p

Passey Q R KM BohacsW L Esch R Klimentidis and SSinha 2010 From oil-prone source rock to gas-producing

Loucks et al 1097

shale reservoir Geologic and petrophysical characteriza-tion of unconventional shale-gas reservoirs InternationalOil and Gas Conference and Exhibition Beijing ChinaJune 8ndash10 2010 SPE Paper 131350 29 p

Pitman J K L C Price and J A LeFever 2001 Diagenesisand fracture development in the Bakken FormationWillis-ton Basin Implications for reservoir quality in the middlemember US Geological Survey Professional Paper 165319 p

Pittman E D 1979 Porosity diagenesis and productivecapability of sandstone reservoirs in P A Scholle andP R Schluger eds Aspects of diagenesis SEPM SpecialPublication 26 p 159ndash173

Pittman E D 1992 Artifact porosity in thin sections of sand-stones Journal of Sedimentary Research v 62 p 734ndash737

Pytte A M and R C Reynolds 1988 The thermal trans-formation of smectite to illite inN D Naeser and T HMcCulloh eds Thermal histories of sedimentary ba-sins New York Springer-Verlag p 133ndash140

Raiswell R 1976 The microbiological formation of car-bonate concretions in the Upper Lias of NE EnglandChemical Geology v 18 p 227ndash244

Reed R M and R G Loucks 2007 Imaging nanoscalepores in the Mississippian Barnett Shale of the northernFort Worth Basin (abs) AAPG Annual ConventionAbstracts v 16 p 115

Rieke H H and G V Chilingarian 1974 Compaction ofargillaceous sediments Developments in sedimentologyAmsterdam Elsevier v 16 424 p

Ross D J K and R M Bustin 2009 The importance of shalecomposition and pore structure upon storage potentialof shale gas reservoirs Marine and Petroleum Geologyv 26 p 916ndash927 doi101016jmarpetgeo200806004

Rouquerol J D Avnir C W Fairbridge D H EverettJ H HaynesN Pernicone JD F Sing andKKUnger1994 Recommendations for the characterization of po-rous solids Pure and Applied Chemistry v 66 p 1739ndash1758 doi101351pac199466081739

Ruppel S C and R G Loucks 2008 Black mudrocks Les-sons and questions from the Mississippian Barnett Shalein the southern mid-continent The Sedimentary Re-cord v 6 p 4ndash8

Schieber J 2010 Common themes in the formation andpreservation of porosity in shales and mudstones Illus-trated with examples across the Phanerozoic Societyof Petroleum Engineers Unconventional Gas Confer-ence Pittsburgh Pennsylvania February 23ndash25 2010SPE Paper 132370 12 p

Sisk C D Diaz J Walls A Grader and M Suhrer 2010


3-D visualization and classification of pore structure andpore filling in gas shales Society of Petroleum EngineersAnnual Technical Conference and Exhibition FlorenceItaly September 19ndash22 2010 SPE Paper 134582 4 p

Slatt EM and N R OrsquoNeal 2011 Pore types in the Barnettand Woodford gas shale Contribution to understand-ing gas storage and migration pathways in fine-grainedrocks (abs) AAPG Annual Convention Abstracts v 20p 167

Sondergeld C H R J Ambrose C S Rai and J Moncrieff2010 Microstructural studies of gas shales Society ofPetroleum Engineers Unconventional Gas ConferencePittsburgh Pennsylvania February 23ndash25 2010 SPEPaper 131771 17 p

Surdam R C D B MacGowan and T L Dunn 1991 Pre-dictive models for sandstone diagenesis Organic Geo-chemistry v 17 p 242ndash253

Tomutsa L D Silin and V Radmilovic 2007 Analysis ofchalk petrophysical properties by means of submicron-scale pore imaging and modeling Society of PetroleumEngineers Reservoir Evaluation and Engineering v 10p 285ndash293

TXCO Resources 2009 The emerging resource companyTXCO Resources Howard Weil 37th Annual EnergyConference New Orleans March 22ndash29 2009 35 phttpwwwscribdcomdoc20128412The-Emerging-Resource-Company (accessed March 25 2011)

Van Sickel W A M A Kominz K G Miller and J VBrowning 2004 Late Cretaceous and Cenozoic sea lev-el estimates Backstripping analysis of borehole data on-shore New Jersey Basin Research v 16 p 451ndash465doi101111j1365-2117200400242x

Weeks L G 1957 Origin of carbonate concretions inshales Magdalena Valley Colombia Geological Societyof America Bulletin v 88 p 95ndash102 doi1011300016-7606(1957)68[95OOCCIS]20CO2

Wilkin R T H L Barnes and S L Brantley 1996 The sizedistribution of framboidal pyrite in modern sedimentsAn indicator of redox conditions Geochimica et Cos-mochimica Acta v 60 p 3897ndash3912 doi1010160016-7037(96)00209-8

Wilkin R T M A Arthur and W E Dean 1997 Historyof water column anoxia in the Black Sea indicated bypyrite framboids size distributions Earth and PlanetaryScience Letters v 148 p 517ndash525 doi101016S0012-821X(97)00053-8

Williams L A and D A Crerar 1985 Silica diagenesis IIGeneral mechanisms Journal of Sedimentary Petrologyv 55 p 312ndash321 doi101306212F86B1-2B24-11D7-8648000102C1865D

Mesozoic carbonate and shale successions todevelop process and response models of de-position and diagenesis

Ursula Hammes Bureau of EconomicGeology Jackson School of Geosciences Uni-versity of Texas at Austin University StationBox X Austin Texasursulahammesbegutexasedu

Ursula Hammes obtained her diploma in geol-ogy from the University of Erlangen in Germanyin 1987 and her PhD from the University ofColorado at Boulder in 1992 She spent 10 yrworking as a consultant performing postdoc-toral research at the Bureau of EconomicGeology and as an exploration geologist in theindustry She joined the Bureau of EconomicGeology in 2001 as a research associate Hermain research focus is in clastic and carbonatesequence stratigraphy depositional systemsand carbonate and clastic diagenesis Her recentresearch objective is in shale-gas systems inTexas and Germany

ACKNOWLEDGEMENTS

This research was funded by the Mudrock Sys-tems Research Laboratory and the State of TexasAdvanced Resource Recovery (STARR) Projectat the Bureau of Economic Geology Industrymembers include Anadarko BP ChesapeakeChevron Cima Cimarex ConocoPhillips CypressDevon Encana EOG Resources EXCO ResourcesHusky Marathon Pangaea Penn Virginia PennWest Pioneer Shell StatOil Texas AmericanResources The Unconventionals US EnerCorpValence and YPF (Yacimientos PetroliferousFiscales) Editing was provided by Lana DieterichBureau of Economic Geology The authors ac-knowledge reviews by David Awwiller ShirleyDutton Terri Olson and Arthur Trevena RuarriDay-Stirrat provided an ion-milled sample fromthe Ursa Basin in the Gulf of Mexico Publicationwas authorized by the director of Bureau ofEconomic Geology Jackson School of Geosci-ences University of Texas at AustinThe AAPG Editor thanks the following reviewersfor their work on this paper David N AwwillerTerrilyn M Olson and Arthur Trevena

1072 Spectrum of Pore Types and Networks In

88 by several kilometers of burial At the onset of hydrocar-bon thermal maturation OM pores are created in kerogen Atdepth dissolution of chemically unstable particles can createadditional moldic intraP pores

INTRODUCTION

Nanometer- to micrometer-size connected matrix-related poresform flow networks along with fractures where present thatcontribute to the natural permeability pathways for gas flowin unconventional mudrock reservoirs (the term ldquomudrockrdquo isused to includemudstones and shales) (Javadpour 2009 Curtiset al 2010) Research during the last several years has identi-fied a variety of pore types in thesemudrocks (ie Davies et al1991 Desbois et al 2009 Loucks et al 2009 2010 Curtiset al 2010 Diaz et al 2010Milner et al 2010 Passey et al2010 Schieber 2010 Sisk et al 2010 Lu et al 2011 Slatt andOrsquoNeal 2011) We propose a simple relatively objective de-scriptive approach to grouping mudrock matrix-related poresinto three basic categories on the basis of their relationships toparticles (Figure 1) Pore types include interparticle (interP)mineral pores intraparticle (intraP) mineral pores and intraPorganic-matter (OM) pores This type of pore classificationis consistent with schemes of pore subdivisions used in con-ventional sandstone (ie Pittman 1979) and carbonate (ieChoquette and Pray 1970) reservoir systems Although de-scriptive it provides a first step in relating pores to reservoirproperties For example type size and arrangement of poresaffect storage and calculations of hydrocarbons in place (ieAmbrose et al 2010) as well as sealing capacity (Dawson andAlmon 2002 Dewhurst et al 2002 Schieber 2010) Alsodifferences in the origins and distribution of pore types havebeen shown to have different effects on permeability and wet-tability (ie McCreesh et al 1991 Passey et al 2010) Thesethree categories of pores can be plotted on a ternary diagram tocompare different scales of sampling such as different areasof a scanning electron microscope (SEM) sample different in-tervals of an individual shale play or summary comparisons ofdifferent shale plays in different basins

The major goals of this article are to document in a simpleand systematic manner the wide spectrum of matrix-relatedpores that exist in mudrocks and classify these pores on thebasis of basic descriptive attributes Specific objectives are to(1) present a pore classification that summarizes the spec-trum of pore types seen in mudrock reservoirs and that alsocaptures some important petrophysical properties of the pores

Mudrocks


DATA AND METHODS






Loucks et al 1073










Loucks et al 1075













Kwon et al (2004)


Lu et al (2011)




Curtis et al (2010)






1076Spectrum

ofPoreTypes

andNetworks

InMudrocks



Milner et al (2010)


























Milner et al (2010)



Schieber (2010)


Loucksetal

1077

Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica


Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica


)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM


ork(authorrsquos

figure4F)


figure5CD


figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP


M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger







Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097



























DATA AND METHODS






Loucks et al 1073










Loucks et al 1075













Kwon et al (2004)


Lu et al (2011)




Curtis et al (2010)






1076Spectrum

ofPoreTypes

andNetworks

InMudrocks



Milner et al (2010)


























Milner et al (2010)



Schieber (2010)


Loucksetal

1077

Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica


Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica


)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM


ork(authorrsquos

figure4F)


figure5CD


figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP


M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger







Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097



































Loucks et al 1075













Kwon et al (2004)


Lu et al (2011)




Curtis et al (2010)






1076Spectrum

ofPoreTypes

andNetworks

InMudrocks



Milner et al (2010)


























Milner et al (2010)



Schieber (2010)


Loucksetal

1077

Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica


Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica


)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM


ork(authorrsquos

figure4F)


figure5CD


figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP


M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger







Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097






































Kwon et al (2004)


Lu et al (2011)




Curtis et al (2010)






1076Spectrum

ofPoreTypes

andNetworks

InMudrocks



Milner et al (2010)


























Milner et al (2010)



Schieber (2010)


Loucksetal

1077

Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica


Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica


)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM


ork(authorrsquos

figure4F)


figure5CD


figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP


M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger







Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097




























Milner et al (2010)


























Milner et al (2010)



Schieber (2010)


Loucksetal

1077

Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica


Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica


)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM


ork(authorrsquos

figure4F)


figure5CD


figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP


M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger







Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097


























Devonian

Antrim

Shale

Mich

igan

Basin

IntraP

Cleavage

poreswithinmica


Hoveretal(1996)

Devonian

Seneca

Mem

ber

ofOnondaga

Limestone

NewYork

IntraP

Cleavage

poreswithinmica


)Ho

veretal(1996)

Ordovician

MaquoketaGroup

Indiana

InterPintraPandOM


ork(authorrsquos

figure4F)


figure5CD


figure6F)

Schieber

(2010)

Cambrian

EauClaire

Form

ation

Indiana

InterP

andintraP


ork(authorrsquos

figure4G

)Schieber

(2010)

Datainclu

dematerialgenerated

bythisstu

dyandmaterialfrom

theliterature

InterP


M=organicmatter

dagger MSRL=samples

from

theMudrock

Syste

msResearch

Laboratory

attheBureau

ofEconom

icGeologyUn

iversity

ofTexasatAustin

Table1

Continued

Age

UnitNa

me

GeneralLocation

Pore

Type

(thisarticle)

Pore

Type

asDescribed

byAuthor

Source

dagger







Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097



























Interparticle Pores




Loucks et al 1079







Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097
































Reference

Interparticle

Intergranular




Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)

Intraparticle



Phyllosilicate

Curtis et al (2010)




Kwon et al (2004)


Milner et al (2010)

Organic matter

Organophyllic

Curtis et al (2010)

Organic matter








Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097































Loucks et al 1081

Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097


























Intraparticle Pores







Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097



























Loucks et al 1083




Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097





























Loucks et al 1085















Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097








































Loucks et al 1087




Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097





























Loucks et al 1089



















DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097












































DISCUSSION




Loucks et al 1091













Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097






































Loucks et al 1093
















Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097









































Loucks et al 1095



REFERENCES CITED

















































Loucks et al 1097




























REFERENCES CITED

















































Loucks et al 1097
























































Loucks et al 1097


























spectrum of pore types and networks in mudrocks and a ...€¦ · ventional sandstone (i.e.,...

Documents