International Journal of Coal Geology · Posidonia Shale samples from the Hils Syncline, Germany, using BIB-SEM. In addition, the bulk porosity is estimated and compared to MIP. Here

International Journal of Coal Geology xxx (2012) xxx–xxx

COGEL-02026; No of Pages 14

Contents lists available at SciVerse ScienceDirect

International Journal of Coal Geology

j ourna l homepage: www.e lsev ie r .com/ locate / i j coa lgeo

BIB-SEM study of the pore space morphology in early mature Posidonia Shale fromthe Hils area, Germany

Jop Klaver a,⁎, Guillaume Desbois a, Janos L. Urai a, Ralf Littke b

a Structural Geology, Tectonics and Geomechanics, Energy and Mineral Resources Group (EMR), RWTH Aachen University, Germany, Lochnerstrasse 4–20, 52056, Aachen, Germanyb Institute of Geology and Geochemistry of Petroleum and Coal, Energy and Mineral Resources Group (EMR), RWTH Aachen University, Lochnerstrasse 4–20, D, 52056 Aachen, Germany

⁎ Corresponding author. Tel.: +49 241 80 98441; faxE-mail address: [email protected] (J. Klav

0166-5162/$ – see front matter © 2012 Elsevier B.V. Alldoi:10.1016/j.coal.2012.06.012

Please cite this article as: Klaver, J., et al., BGermany, Int. J. Coal Geol. (2012), doi:10.1

a b s t r a c t
a r t i c l e i n f o
Article history:Received 14 December 2011Received in revised form 29 June 2012Accepted 30 June 2012Available online xxxx

Keywords:Posidonia ShaleBroad-ion-beam millingSEM imagingPore morphologyPore size distributionMercury porosimetry

This contribution reports on the study of the pore space morphology in two early mature (VRr=0.59 and0.61) samples of Posidonia Shale from the Hils Syncline in Germany, using Broad Ion Beam (BIB) polishingand high resolution Scanning Electron Microscopy (SEM). This allows imaging pores with resolution downto 10 nm in equivalent diameter (deq), and quantitative estimation of porosity. Using a combination of BSEand SE detectors and semi-automatic segmentation of the gigapixel images, the representative elementaryarea of the samples, on the scale of a few mm, is inferred to be about 140×140 μm2. Pore morphologiesand pore sizes are clearly related to the mineral fabric, with large differences: very large (typically severalmicrons) pores with internal faceted crystal morphology in recrystallized calcite clasts, and smaller pores(deqb1024 nm) in clay-rich matrix and in cf. Schizosphaerella nanofossils (typically 200 nm). Pores are lesscommon in organic material and in pyrite aggregates. Pore characteristics are very similar for both samples,and porosity resolvable by BIB-SEM is 2.75 and 2.74%. Pore size distribution can be described by a power-law,with an exponent about 2.0 and 2.2, respectively, for the pore population excluding the fossils. Pores in thecarbonate fossils show dual-power-law distribution with power-law exponents of about 1 and 3. By extrap-olating the power-law distribution for each sample, total porosity is estimated to be 4.8% (−0.9, +1.7%), and6.5% (−2.7, +7.2%), respectively. This estimate can be compared with 3.4–3.7 and 3.3–3.6% as measured bymercury injection porosimetry. We interpret this difference to reflect the unconnected (for mercury) part ofthe porosity. Comparison of imaged pores and mercury injection porosimetry suggest a very high pore bodyto pore throat ratio. This results in a pore model where large pores, represented mainly by pores in fossils andcalcite grains, are connected via a low-porous (and low-permeable) clay-rich matrix with pore throats below10 nm in both samples.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

Petrophysical properties of organic-rich shales are of special inter-est due to the growth in exploration and production of gas shales. Incentral and western Europe, one of the potential gas shales is theLower Jurassic (Toarcian) Posidonia Shale which is wide-spreadand known as the principal source rock for petroleum in the NorthGerman Basin, the Upper Rhine Valley and the Paris Basin. A gas shaleis both the source and the reservoir of the gas, and pores are the reser-voirs of free gas (Bustin et al., 2008). Decrease of shale porosity withdepth is well known (Aplin et al., 2006; Broichhausen et al., 2005) butthe relative roles of mechanical compaction and diagenesis are stillnot well understood. Quantifying the pore structure is still challengingin low porous and low permeable rocks, due to a lack of an appropriatemethod to investigate directly sub-micrometer structures in represen-tative area and volume. The most popular conventional bulk porosity

: +49 241 80 92358.er).

rights reserved.

IB-SEM study of the pore spa016/j.coal.2012.06.012

measurements are performed by mercury injection porosimetry (MIP)and gas adsorption porosimetry. MIP measures pore size down to3 nm in diameter but only from the connected part of porosity andinterpretation of the measurements is based on a simplified model ofcylindrical pore tubes (Washburn, 1921) which does not reflect thecomplexity of natural pore network. Therefore, pore sizes inferredfrom MIP are underestimated due to the ink-bottle effect (Münch andHolzer, 2008) and give only information about pore throat size. More-over, during the omnidirectional injection at high pressure, pore col-lapse is possible when an effective stress is created by the capillarypressure and gradients in saturation in the sample (Hildenbrand andUrai, 2003; Hildenbrand et al., 2005). With gas adsorption BET, surfacearea can bemeasured (Brunauer et al., 1938), and pore size distributioncan be calculated (Barrett et al., 1951; Schull, 1948) down to 0.3 nm indiameter. However, these data are still only related to the connectedpo-rosity and are based on a simplifiedmodel. Bothmethods lack direct in-formation about pore morphologies and the relation of porosity tomineralogy and microstructures. In contrast, recent developments ofion beam milling allow study of porosity and microstructure on high

ce morphology in early mature Posidonia Shale from the Hils area,

http://dx.doi.org/10.1016/j.coal.2012.06.012

mailto:[email protected]


http://www.sciencedirect.com/science/journal/01665162


2 J. Klaver et al. / International Journal of Coal Geology xxx (2012) xxx–xxx

quality flat surfaces in SEM (Desbois et al., 2009). This research onnanopores in low porous rocks has grown rapidly in recent years (e.g.,Bernard et al., 2012; Chalmers et al., 2009, 2012; Curtis et al., 2011a;Holzer and Cantoni, 2011; Milner et al., 2010; Passey et al., 2010;Schieber, 2010; Schneider et al., 2011; Slatt and O'Brien, 2011; Wangand Reed, 2009), see Loucks et al. (2012) for a thorough overview andpore classification. Serial cross-sectioning with focused ion beam (FIB)milling in combination with SEM is able to deliver a 3D model of porespace (Ambrose et al., 2010; Curtis et al., 2012; Desbois et al., 2009;Heath et al., 2011; Keller et al., 2011; Sisk et al., 2010; Sondergeld etal., 2010) but the studied volume is limited, typically 10×10×10 μm3

and usually not representative. Complementary to these methods, acombination of broad-ion-beam milling and scanning electron micros-copy (BIB-SEM) allows imaging of large (greater than mm2) planar,undamaged surfaces. This technique is suitable for the qualitative andquantitative study of microstructures and porosity, in representative el-ementary areas. This combined BIB-SEM technique to quantify thepores was used on claystones from reference sites for waste disposal(Desbois et al., 2009; Hemes et al., 2011), salt (Desbois et al., 2012),tight gas sandstones (Desbois et al., 2011) and on other, organic-rich,shales (Loucks et al., 2009).

The major goal of this study is to obtain pore size distributions,quantify the porosity and to study the pore morphology in represen-tative areas, on the scale of the BIB cross-section, of two early maturePosidonia Shale samples from the Hils Syncline, Germany, usingBIB-SEM. In addition, the bulk porosity is estimated and comparedto MIP. Here the physical porosity (Pearson, 1999) is studied in themeso- to macropores range (Nelson, 2009; Rouquerol et al., 1994).The BIB-SEM results are compared with MIP data to infer the proper-ties of the pore network, and to formulate a conceptual pore model.

2. Samples and geological setting

Both samples are from a Toarcian shale interval known as thePosidonia Shale (or Lias Epsilon), from the Hils Syncline in northernGermany (Fig. 1). The Posidonia Shale in the Hils Syncline was drilledand completely cored in several shallow boreholes in the 80's andcomprehensive studies were done by several authors (Bernard et al.,2012; Jochum et al., 1995; Littke et al., 1988, 1991a, 1991b; Mann,1987; Mann and Müller, 1988; Rullkötter et al., 1988). The coredshale intervals represent a large maturity range from very early ma-ture to overmature gas window. The reason for the partly high matu-rity in this area is discussed and attributed to either a Late Cretaceousmagmatic heating or to deep burial during the Late Jurassic and EarlyCretaceous (Bartenstein et al., 1971; Petmecky et al., 1999). ThePosidonia Shale consists of two units: an upper calcareous shale,and a lower marlstone (Littke and Rullkötter, 1987). Mann (1987)found, based on XRD, that a typical calcareous shale consists of 43%clay minerals, 37% calcite, 15% quartz and feldspar and 5% pyrite;whereas a typical lower marlstone consists of 35% clay minerals,50% calcite, 11% quartz and feldspar and 4% pyrite. The origin of thecalcite are mainly coccoliths and other plankton-derived microfossils(Littke et al., 1991a), of which a part is recrystallized depending onthe thermal maturity (Rullkötter et al., 1988). Vitrinite reflectancevaries from 0.48 in the southeast to 1.45 in the northwest(Rullkötter et al., 1988). MIP porosities have been measured andrange between 2.4 and 22% (Mann, 1987) and a weak dependencyof VRr is seen. One geochemical scanning transmission X-ray micros-copy (STXM) and transmission electron microscopy (TEM) study hasbeen done recently on the Posidonia Shale (Bernard et al., 2012) indi-cating intra-particle pores of 1–50 nm in the organic matter for themature samples which is consistent with work done by others(Chalmers and Bustin, 2008; Chalmers et al., 2012; Curtis et al.,2011b; Loucks et al., 2009; Ross and Bustin, 2009; Sondergeld et al.,2010). This contribution is based on two samples, known as

Please cite this article as: Klaver, J., et al., BIB-SEM study of the pore spGermany, Int. J. Coal Geol. (2012), doi:10.1016/j.coal.2012.06.012

RWEP06 and RWEP08. Both samples are thought to be dry as theywere stored under normal atmospheric conditions.

3. Methods and approaches

3.1. XRD analysis, incident light organic petrography and Rock Evalpyrolysis

X-ray powder diffraction (XRD) analysis was performed on bothRWEP06 and RWEP08 using a Bruker D5000 at the Geological Insti-tute Aachen (GIA) RWTH Aachen University. Quantification of thedifferent minerals was done using the Rietveld Method (TOPAS),with special attention for clay (Kahle et al., 2002). In order to studythe samples in incident white light and in an incident light fluores-cence mode, polished sections of whole rocks were prepared in ori-entation perpendicular to bedding following the proceduredescribed in Sachse et al. (2011). Vitrinite reflectance was measuredin oil immersion with magnification of 50 times following standardprocedures. An Yttrium-Alluminium-Garnet (R=0.89%) was usedfor calibration. Reflectance measurements followed standard proce-dures as described in Taylor et al. (1998) and details of the micro-scopic equipment are described in Littke et al. (2012). Vitrinitereflectance was measured on 49–58 points of RWEP06 andRWEP08 respectively, and mean values were calculated. It shouldbe noted that autochthonous vitrinite is rare in Posidonia Shale,whereas resedimented vitrinite and inertinite are much more com-mon (Littke et al., 1988). However, the dominant macerals areliptinites (alginite). Rock-Eval pyrolysis was done according toEspitalié et al. (1977).

3.2. Sample preparation

Core fragments were stored at room temperature in a plasticair-tight container. Subsamples (~0.5×0.5×0.5 cm3) were cut witha saw blade and pre-polished using silicon carbide (SiC) sandpaperto reduce roughness from sawing down to 10 μm. Surface damage in-duced by SiC polishing was removed by argon BIB polishing which re-moved a 100 μm thick layer from the surface. The size of a typical BIBpolished cross-section is 1–2 mm2 (Fig. 2A). Samples were BIB argonpolished in a JEOL SM-09010 cross-section polisher (8 h, 1×10−3–

1×10−4 Pa, 6 kV, 150 μA) to produce a high quality, planarcross-section with topography less than ±5 nm as measured byAFM (Fig. 2B) and very sharp edges at pore boundaries.

The BIB polished cross-sections were Au-coated and imaged in aZeiss Supra 55 SEM with a back scatter detector (BSE) for phase con-trast imaging and a secondary electron detector (SE2) for topographyinvestigation up to a magnification of 30,000 corresponding to a pixelsize of 10 nm. From BIB cross-sections, large areas were selected to beimaged (SE2) at magnifications of 10,000, 20,000 and 30,000 timesusing 10–20% of overlap to create a large representative mosaicmade of hundreds of single images (Table 1) to study the poresdown to the resolution of SEM. A minimum of 10 pixels was shownto be the practical pore resolution, i.e., all pores with a minimumsize of 10 pixels are detected (Fig. 2C). Mosaics made with the BSEdetector were imaged at a magnification of 1500, 2500, 5000 and15,000 times to gain qualitative information about the mineralogy.An energy-dispersive X-ray spectroscopy (EDX) detector was usedfor semi-quantitative chemical composition analysis.

3.3. Image processing and pore segmentation

Single images were stitched together with a bicubic interpolationalgorithm in Autopano 2 to build large mosaics preserving the pixelresolution. Pores were segmented semi-automatically in ArcGIS 10from mosaics of SE2 images by using the contour tool in the 3D Ana-lyst toolbar in ArcGIS 10 (Fig. 2C). With this tool the pore boundary of

ace morphology in early mature Posidonia Shale from the Hils area,


Fig. 1. Geological setting of our samples. (A) Location of the Hils area, south of Hannover, Germany. (B) Geological map of the Hils Syncline after Mann (1987). Isovitrinite reflec-tance contours after Bartenstein et al. (1971) and Koch and Arnemann (1975). Vitrinite reflectances from well data are plotted as dots after Littke et al. (1988). Names of strati-graphic units are given in the legend.

3J. Klaver et al. / International Journal of Coal Geology xxx (2012) xxx–xxx

each single pore is selected by the user and a contour at that specificgray value is drawn around the pore by ArcGIS. If the result is not sat-isfactory, the contour can be deleted or edited. Using this method,segmentation is faster than full manual interpretation. After segmen-tation pores were analyzed in MATLAB with PolyLX (Lexa, 2010).Cracks were interpreted as core damage due to stress relaxationafter sample collection and were removed during analysis.

Segmentation of mineral phases was achieved by using gray valuepatterns from BSE images validated by a combination of EDX and XRD

Fig. 2. BIB cross-sectioning. (A) Typical overview of a BIB polished cross-section made in sathe cross section. (B) The topography mapmeasured by Atomic Force Microscopy (AFM) on tthe BIB cross-section allows imaging pores into 2D flat surface down to the SEM ressemi-automated segmentation. Arrows indicate a resolved pore close to the practical pore

Please cite this article as: Klaver, J., et al., BIB-SEM study of the pore spaGermany, Int. J. Coal Geol. (2012), doi:10.1016/j.coal.2012.06.012

investigations. Because of low gray value contrast, quartz, mica andcarbonate fossils were segmented manually (Fig. 3).

3.4. Representative elementary area

The representative elementary area (REA) was determined withthe box counting method (Kameda et al., 2006) on the classifiedBSE mosaic of samples. Thus, (as shown in Fig. 4), gray values fromBSE mosaics that were ranked in ArcGIS10 were first attributed to

mple RWEP06 with area of about 1 mm2. The dashed line indicates the contour area ofhe BIB-polished shale surface from (A) shows an amplitude of ±5 nm. (C) Zoom in intoolution. Comparison of pore segmentation methods based on thresholding versusresolution of the SEM.



Fig. 3. Mineral fabric and classification of mineralogy in sample RWEP06. (A) BSE mo-saic shows mineral phase contrast. (B) Identification of minerals by using the EDX de-tector which provides semi-quantitative chemical analysis of the BIB cross-section.Identified chemical elements are given in the legend. (C) Interpreted segmented min-eralogy based on the BSE mosaic and EDX analysis, mineralogy is given in the legend.

Table 1Mosaic properties for samples RWEP06 and RWEP08.

Sample Mosaic Magnification #Images

Pixel size[nm2]

Practical poreresolution [nm]

RWEP06 I 30,000 120 10×10 =36II 10,000 169 29×29 =104

RWEP08 I 20,000 144 15×15 =54


the phases organic matter, clay matrix, calcite and pyrite. Then, astepwise growing box was applied to the classified BSE image. Ateach increasing box size the amount of the different phases presentin that box was calculated in the box. This process was repeated forseveral different starting points in the mosaic. At a given box sizethe individual contribution of each phase to the overall compositiondoes not significantly change. This area is interpreted to be the REAon the scale of the BIB cross-section.

3.5. Pore size by mercury intrusion and extrusion

For comparison with the BIB-SEM-based pore data, MIP wasperformed onboth samples atMicromeritics Analytical Services Europe,Aachen, Germany. Mercury intrusion and extrusion analysis wasdone in the 360–0.003 μm pore throat range. Samples to be analyzedwere dried at room temperature in vacuum (1×10−3–1×10−4 Pa)for 30 h. No significant change in weight was measured, as expected.Pore throat radius distribution calculations are based on theWashburnequation (Washburn, 1921). Pressures used for intrusion were up to414 MPa corresponding to a pore throat diameter of 3 nm (Hg surfacetension=485 dynes/cm, contact angle=130°). Extrusion of the mer-cury takes place at decreasing pressure steps. Difference in intrusionand extrusion mercury volumes give trapped mercury volumes dueto the pore body to pore throat differences. Pore body to pore throatratios can be calculated from MIP (Cerepi et al., 2002; Janssen et al.,2011; Webb, 2001). Measurements were corrected for blank errorsand conformance errors (Sigal, 2009). After drainage, the sampleswere weighed to measure the amount of mercury remaining in thesamples.

3.6. Power-law distribution and porosity nomenclature

For comparison of the pore size distributions in the different mo-saics and samples, pore data are analyzed as normalized discrete fre-quency as a function of pore area;

Ni

biSmosaic¼ CS−D

pore ð1Þ

or alternatively:

LogNi

biSmosaic

� �¼ −D:Log Spore

� �þ LogC ð2Þ

where Ni is the number of pores with pore area Spore within bin bi,where bi doubles each next bin, except for the first bin (1, 1, 2, 4, 8,etc., with corresponding boundaries at 1, 2, 4, 8, 16, etc.). Smosaic isthe surface area of the current mosaic, C is a constant and D is thepower-law exponent. When the pore size distributions follow apower-law we interpret this as self-similar. The pore size distribu-tions are plotted and the best fit is calculated between the largestpore and the practical pore resolution (Fig. 5), which bounds thepower-law distribution (Eq. (2)). This fit is later used for the extrap-olation of the pore size distribution down to the pore sizes, which arenot detectable at the resolution of the SEM. The 95% confidence levels


and the outermost linear fits of the log–log data are plotted as a mea-sure of uncertainty for the best fit.

We adopted the following nomenclature. The porosity (of phase(θ) in sample (N) inferred by the method (M) at the given resolution(X) is given by: ϕN;θ

M;X, if the porosity is extrapolated from data and notmeasured, the same porosity is then indicated by the symbol (*).Thus, in this case the symbol ϕN;θ

M;X*. Therefore, for example, the total(T) porosity measured by BIB-SEM (BIB) method down to 36 nm insample RWEP06 (06) made of organic matter (OM), clay matrix(CM), calcite (C), pyrite (P) and fossils (F) is written as follows:

ϕBIB;3606;T ¼ ϕBIB;36

06;OM þ ϕBIB;3606;CM þ ϕBIB;36

06;C þ ϕBIB;3606;P þ ϕBIB;36

06;F

¼ ϕBIB;3606;NF þ ϕBIB;36

06;F

ð3Þ

Where ϕ06;NFBIB;36 represents the porosity associated with non-carbonate

fossil phases. A list of symbols used to specify all possible M, X, N andθ terms used in this paper is given in Table 2.




4. Results

4.1. XRD analyses, organic petrography, organic carbon content andRock-Eval pyrolysis

XRD analyses show that both samples RWEP06 and RWEP08 con-sist, respectively, mainly of calcite (44% and 57%), quartz (18% and17%) and pyrite (4% and 4%), together with phyllosilicates of the mus-covite/illite (14% and 8%) and kaolinite/clinochlore (9% and 6%). Theother components consist of a mixture of albite, anhydrite, nacrite,chlorite, anorthite and siderite counting for about 10% of the totalcontent (11% for RWEP06 and 8% for RWEP08). RWEP06 has a TOCof 6.05 with mean vitrinite reflectance (VRr) of 0.59% whereasRWEP08 has a TOC of 8.14 with average VRr of 0.61%. Rock-Eval S1and S2 peaks indicate a type I-II kerogen (Fig. 6A). Table 3 summa-rizes all compositional data for the two samples. Moreover, both sam-ples are rich in brightly fluorescing alginite, but only sample RWEP06contains Tasmanales algae as a common constituent (Fig. 6B).

Fig. 4. Principle of representative elementary area calculation (REA). (A) BSE mosaic is segmgrid is placed on the segmented BSE mosaic in (B) to perform the box counting method. (CEvolution of mineralogy content with increasing box size and for different starting points forDashed line in (D–G) indicates the box size where the overall composition is stable, pointin


4.2. Representative elementary area

BSE gigapixel mosaic (Fig. 4A) combined with EDX chemical anal-yses allow identification of 5 different mineral phases: calcite, organicmatter, clay rich matrix, pyrite and others (Fig. 4B). A first boxcounting analysis of the above 5 different mineral phases from a seg-mented BSE mosaic shows that the REA is between 100×100 and200×200 μm2 (Fig. 4C). Based on a more detailed approach, the min-imum REA for both samples was interpreted to be 140×140 μm2

(Fig. 4B–F). Below areas of 140×140 μm2, the relative content ofeach mineral phase does change significantly for both samples(Fig. 4C–G). At the scale of the REA, the only feature that may not berepresentative is the organic matter, which shows some larger fluctua-tions (up to 5%) in some analysis (Fig. 4D, F) probably because the or-ganic matter-rich regions are mostly distributed as rare large particleswithin the matrix (blue line Fig. 4F, for example). Above the area of140×140 μm2 and at the scale of BIB-milled cross-section, the averagemineral contents of sample RWEP08 are: pyrite=2.3%±0.38%, organic

ented according to the different gray values and EDX analysis (B). A stepwise growing) Box counting analysis indicating REA is between 100×100 and 200×200 μm2. (D–G)organic matter (D, F) and Pyrite (E, G) phases and both samples RWEP06 and RWEP08.g to REA.



Fig. 5. Illustration and interpretation of power-law distribution of pore area represented in the form of the logarithm of normalized frequency of pores as a function of the logarithmof binned pore area (Eq. (2)). Black-filled circles are pores fully resolved whereas white-filled circles are pores still detectable but not fully resolvable at SEM resolution. SEM pixelresolution and the largest pore visible in the mosaic bound the data set; the practical pore resolution marks the change between fully resolved and under resolved pores. In thedomain bounded by the practical pore resolution and the largest pore, the data are power-law distributed suggesting the self-similarity of pore area distribution. The solid line in-dicates the best fit of the fully resolved pores defined by the slope D (i.e. the power-law exponent) and the y-coordinate at origin Log [C]. The 95% confidence levels, dotted line, andthe outermost linear fits, dashed line, of the log–log data are plotted as a measure of uncertainty for the best fit.


matter=9%±1.2%, calcite 30%±2.1%, clay-rich matrix=33%±0.9%and other 26%±1.7. For sample RWEP06 themineralogical compositionis: pyrite=1.6%±0.36%, organic matter=14%±2.2%, calcite 18%±3.6%, clay-richmatrix=39%±3.4% and other 28%±4.8. The differencesare due to the intrinsic heterogeneity in the shale.

4.3. Description of mosaics

Two mosaics with different surface areas were imaged for sampleRWEP06: mosaic-I covers 0.3×REA (105×65 μm2), whereas themosaic-II covers 1.1×REA (231×94 μm2). In mosaic-I, close to16,500 pores were detected at the magnification of 30,000 resultingin a visible porosity of 2.62% from SE image analysis (Fig. 7A). TheBSE mosaic shows clearly the carbonate fossils, organic matter, quartzand calcite (Fig. 7B). The largest pores (>500 nm, up to a few μm) areseen in the carbonate fossils and calcite grains (Fig. 7C, D). In betweencarbonate fossils and calcite grains there appears to be a low-porousmatrix showing typical pore sizes about a few 10 nm (Fig. 7C andD). In the mosaic-II, over 26,000 pores were segmented resulting ina visible porosity of 2.75% at a magnification of 10,000, with similarporosity fabric as seen for the mosaic-I: i.e. large pores in carbonatefossils and calcite grains surrounded by a low porous matrix.

One mosaic of 1.0×REA area (163×119 μm2) was made of sampleRWEP08, in which a total of about 55,000 pores were detected at themagnification of 20,000 times resulting in a visible porosity of 2.74%

Table 2Definitions of symbols and terms used to describe porosity.

Symbol Name Range

φ Porosity 0–100%N Sample nr. RWEP06–RWEP08θ Phase T=Total, OM=organic matter, CM=clay-rich matrix,

C=Calcite, F=Fossil, P=Pyrite, NF=Non-FossilM Method BIB=BIB SEM or MIP=MIPx Resolution

resolvable36, 104, 54 (for BIB-SEM), 3 (for MIP) nm

* Extrapolated -


(Fig. 8A). Though in RWEP08, the BSE mosaic (Fig. 8C and B) shows alarger amount of calcite (Fig. 7B) and the carbonate fossils are morefragmentedwithin the lowporousmatrix (Fig. 8C, D), the overall poros-ity fabric has similar characteristics than those observed for RWEP06.

4.4. Pore morphologies

Poremorphologies are very similar in the two samples RWEP06 andRWEP08. Therefore, descriptions below are valid for both samples.

Pores in the organic matter (Fig. 9A) are mainly at interfaces be-tween organic matter and the low porous matrix. These elongatedpores have high aspect ratios (3±1.7). Straight and sharp edgesmatching both sides and lack of cementation suggest that they wereformed by micro cracking after coring (Fig. 9A, B). Pores in and aroundorganicmatter are sub-parallel to bedding. The fewmesopores detectedwere close to the pixel resolution (inset in Fig. 9A). It is well known thatorganic matter has pore sizes which are often below 3 nm (Chalmersand Bustin, 2008; Chalmers et al., 2012; Curtis et al., 2011a, 2012;Prinz and Littke, 2005; Ross and Bustin, 2009; Sondergeld et al., 2010)and thus below the resolution of the methods applied here.

Carbonate fossils can be porous (Fig. 9C–D) or non-porous (Fig. 9F).Porous fossils are usually U-shaped fragmented with sizes of up to10–20 μm(Figs. 7 and 8). The non-porous carbonate fossils have shapesfrom round- to U-shaped and are generally smaller than porous carbon-ate fossils with sizes of up to 5 μm (Fig. 9F). Pores inside the carbonatefossils have sizes below500 nmand are typically polygonswith straightand sharp edges (Fig. 9D). Porous carbonate fossils in RWEP08 havesome tendency to be more fragmented and may be more cementedwith recrystallized calcite (Fig. 8D). The porous carbonate fossils areinterpreted to be cf. Schizosphaerella (Bour et al., 2007; Veiga deOliveira et al., 2007).

Pores related to calcite grains are relatively large (1–4 μm)with fac-eted calcite crystals growing towards the center of the open pore(Fig. 9G–J). These large pores have no preferred shape or orientation.They contribute 24 and 15% to the total visible porosity for RWEP06and RWEP08, respectively. These large pores in calcite contribute signif-icantly to the porosity (Table 4). However, most of the pores in calcitehave sizes below 1 μm. Pores between two mica sheets are elongated



Fig. 6. Rock Eval pyrolysis and organic petrography. (A) Hydrogen Index versus Tmax diagram with measurements of RWEP06 and RWEP08 reported. (B) Image of Tasmanalas algeain RWEP06 under fluorescing light.


and parallel to the longest axis of the mica (Fig. 9I, J). These pores arerare and do not significantly contribute to the total visible porosity. Py-rite framboids in RWEP06 have low porosities with equidimensionalshapes controlled by the arrangement of each single pyrite grain(Fig. 9F), whereas in RWEP08 the pyrite framboids are non-porousand calcite-cemented (Fig. 9K, L).

4.5. Pore size distributions from BIB-SEM data

Because of the limited spatial resolution of the BSE and EDX detec-tors (about few μm) and the sub-micron grains in the clay matrix,pores in the clay matrix cannot be accurately classified by mineralogylike we did for pores in the carbonate fossils. Therefore, for the calcu-lation of pore size distributions, the total porosity seen by BIB-SEM iswritten as the sum of the porosity associated with carbonate fossilsand the porosity associated with non-fossil phases (see Eq. (3)).

Moreover, because of different magnifications used for SEM imag-ing, the practical pore resolution is different for each mosaic (Table)and this affects the pore statistics described below. The largestpores seen in a mosaic are between 500 and 4000 nm in diameter(Fig. 5).

4.5.1. Pore size distributions of carbonate fossilsIn both samples, regardless of themagnification and themosaic area,

pore areas in carbonate fossils follow a dual-power-law distribution(Fig. 10A-C). From bin 1500, 12,300 or 3000 nm2 to bin 24,600 nm2,pore areas are power-law distributed with an exponent around 1(D=0.95, 1.04 and 1.08 with Log C=−7.08, −6.96 and −6.87, forRWEP06_Mosaic-I, RWEP06_Mosaic-II and RWEP08; respectively).From bin 24,600 nm2 to bin 196,600, 393,200 or 1,572,900 nm2, thepores are also power-law distributed but with an exponent around 3(D=3.02, 3.34 and 2.65with Log C=−2.06,−3.46 and−0.17, respec-tively). Pores below bin 24,600 nm2, contribute 32, 15 and 24% to theporosity associated with fossils whereas pores with area from bin

Table 3Mineralogical composition obtained by XRD analyses, TOC, vitrinite reflectance and Rock-E

Sample Calcite [%] Quartz [%] Muscovite/illite [%] Kaolinite/clinochlore [%] Pyrite

RWEP06 44 18 14 9 4RWEP08 57 17 8 6 4


24,600 nm2 to larger pore areas contribute 68, 85 and 76% to the poros-ity associated with carbonate fossils, respectively.

4.5.2. Pore size distributions of non-fossil poresSize distributions were analyzed for the pore population excluding

both the pores, which were interpreted as microcracks formed aftercoring and the pores in the carbonate fossils (Fig. 10). In both sam-ples, regardless of the magnification and the mosaic area, pores innon-fossil phase follow a power-law distribution (Fig. 10D-F). Frombin center 1500, 12,300 or 3000 nm2 (corresponding to their practi-cal pore resolutions) to bin center 6,291,500 or 12,582,900 nm2

pore area, pore sizes are power-law distributed with an exponentaround 2 (D=1.9±0.15, 2.02±0.07 and 2.2±0.12 with Log C=−3.3±0.76, −2.4±0.40 and −1.8±0.66, respectively).

4.6. Pore throat distributions inferred from MIP

Mercury porosimetry data indicate a 3D bulk connected porosityof 5.44% and 7.33% for RWEP06 and RWEP08, respectively (Fig. 11A,B). A significant amount of mercury is injected in the 60–240 μm(RWEP06) and 45–300 μm (RWEP08) pore throats range followedby a low intrusion curve (Fig. 11C, D) contributing to the total intrud-ed porosity (Fig. 11A, B). However, these large pore sizes were neverseen in the microstructural investigations on the BIB cross-sections.Injection volumes in these pore throat ranges are interpreted ascracks, filling of the container and surface roughness, also known asconformance errors (Sigal, 2009). MIP curves are thus corrected bysubtracting the first part of the intrusion data from the total inorder to provide realistic MIP porosity (Fig. 11A, B and C, D). Thepoint from which the data should be subtracted is in the range be-tween the largest pore imaged by the SEM and the start of the low in-trusion seen in the MIP data (60 and 45 μm, respectively). The largestimaged equivalent pore diameters are deq=4.3 μm in RWEP06 anddeq=4.5 μm in RWEP08. Thus, the MIP data are corrected by

val results of RWEP06 and RWEP08.

[%] Other [%] TOC [%] VR, [–] S1 [mg/g] S2 [mg/g] Tmax [°C] HI [–] PI [–]

11 6.05 0.59 2.92 32.8 433 542 0.0828 8.14 0.61 3.88 51.66 434 635 0.070



Fig. 7. SEM analysis of cross-section RWEP06, mosaic-I. (A) SE mosaic overview (30kx, 0.3×REA) with pores segmented in blue. (B) BSE mosaic overview (15kx) of the same area in(A) highlighting mineral phases. (C and D) Details of (A) showing large pores in calcite, in fragmented and U-shaped carbonate fossils surrounded by low porous matrix and poresin pyrite framboid. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


excluding data in the range from 60 to 4.5 μm and 45 to 4.5 μm forRWEP06 and RWEP08, respectively. Therefore, the porosity providedfrom corrected MIP data is in the range of 3.4–3.7% for RWEP06 and3.3–3.6% for RWEP08. For sample RWEP06, the highest mercury po-rosity contribution (8%) is at a pore throat diameter of 6 nm and67% of the mercury is injected in pore throats below 11 nm in diam-eter (Fig. 11C). Sample RWEP08 shows similar results with a highestmercury porosity contribution of 8.5% at a pore throat diameter of7.2 nm and also 67% of the mercury is injected in pore throatsbelow 11 nm in diameter (Fig. 11D). Comparison of intrusion and ex-trusion MIP curves (Fig. 11E, F) indicates that about half of the mercu-ry remained in sample RWEP06 (Fig. 11E) and two third in sampleRWEP08 (Fig. 11F). Probably the mercury remained in the poresdue to the ink-bottle effect (Moro and Böhni, 2002). This is confirmedby the sample weight increase measured after the mercuryintrusion-extrusion cycle, indicating a residual filled porosity of 1.13%(RWEP06) and 1.50% (RWEP08). Pore throat to pore body ratios are cal-culated (Fig. 11G, H) based on Janssen et al. (2011)) and Webb (2001).Differences in mercury intrusion and extrusion curves point to porebody-pore throat ratios between 1:1 and 2000:1 for the 3 to 7.2 nmpore throat diameter interval (Fig. 11G, H), both for RWEP06 andRWEP08, respectively. By plotting pore frequencies as a function ofpore throat diameters (Fig. 11I, J), we show that pore throat diametersare power-law distributed over the entire MIP pore size resolution(i.e., from 3 nm to 4622 nm pore throat diameter) with power-lawexponents similar for both samples (D=2.42±0.07 for RWEP06 andD=2.45±0.08 for RWEP08).

5. Discussion

The combination of BIB milling for high quality surface preparationand SEM imaging gives access to much larger areas than FIB (Ambrose


et al., 2010; Curtis et al., 2012; Desbois et al., 2009; Heath et al., 2011;Keller et al., 2011; Sisk et al., 2010; Sondergeld et al., 2010) so that rep-resentative areas in shales can be studied to quantify themicrostructureand pore space. The comparison of BIB-SEM andMIP offers the possibil-ity to link pore morphologies to bulk properties. Considering that theBIB-polished surfaces are flat, and that the edges of pores intersectedby BIB are sharp with radii less than a nm, the BIB-polished surfacescan be considered as 2D sections through the material. Using the basicprinciples of stereology (Underwood, 1970), the 2D area fraction in aREA is a good estimate of the 3D bulk porosity. The differences betweenthe BIB-SEM results andMIPwill only be due to the unconnected poros-ity and differences in resolution of small pores.

5.1. Microstructural evaluation

Mineral content, vitrinite reflectance and TOC give results verysimilar to those of other studies of Posidonia Shale (Littke et al.,1988; Mann, 1987; Rullkötter et al., 1988). Our samples are typicalof the upper calcareous shale unit at low maturity. An area of140×140 μm2 is interpreted to be representative (up to at least mmscale) based on the BIB cross-section in terms of mineralogy and po-rosity (Fig. 4), when large organic particles (about 15×100 μm2 inFigs. 4, 6B) are not taken into account. Because of the low frequencyand low porosity of such organic particles, this does not significantlyaffect the overall porosity in these samples. Pore morphologies andsizes are similar for both samples and are clearly related to the min-eralogy (Fig. 9). At the scale of observations, the largest pore sizes (typ-ically several microns) are found in large calcite grains (Figs. 7–9), andthe smallest ones in the matrix. A significant part of the porosity visiblein SEM images is associated with carbonate fossils (about 20% of poros-ity down to the practical BIB-pore resolution, Table 4), which are em-bedded in the matrix. Organic matter exhibits only few intra-particle



Fig. 8. SEM analysis of cross-section RWEP08. (A) SE mosaic overview (20kx, 1.0×REA) with segment pores in red. (B) BSE mosaic of (A), (15kx). SE image of large pores partlysurrounded by low porous, at resolution used, matrix and porous fossil fragments (C). SE image of large pores partly surrounded by low porous matrix, porous fossil fragmentsand a large pore in organic matter (D). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


pores (b30 nmwhen visible) andmost of the pores (withwidth around200 nmand axial ratio about 3, Fig. 9A) related to organicmatter are lo-cated at interfaces with matrix.

Pore areas detected by BIB-SEM, both in carbonate fossils andnon-carbonate fossils, are interpreted to have power-law distributions(Fig. 10). This enables comparison of the pore size distributions. Apower-law distribution of pore throats is also interpreted from theMIP data (Fig. 11I, J).

The dual-power-law distribution of pore areas in carbonate fossils(Fig. 10A-C) implies that pores smaller than bin 24,600 nm2 contributeprogressively less to the porosity of carbonate fossils. As the break inslope occurs both in the range where pores are fully resolved and at thesame length scale in all three mosaics, the dual-power-law distributionof pore areas in carbonate fossils is interpreted to represent the porosityin the carbonate fossils. Because of the progressively lower contributionto porosity below bin 24,600 nm2, we interpret that the total porosity as-sociated with carbonate fossils is virtually resolved by BIB-SEM (Table 4).In Bour et al. (2007), rectangular-shaped, sub-micrometer-size poreshave been observed corroborating our interpretations. Pore areas of thenon-fossil phase follow a single power-law distribution with power-lawexponents D of 2.02±0.07 and 2.2±0.12 for the representative mosaicsof RWEP06 and RWEP08, respectively, which suggests that the smallpores contribute significantly to the visible porosity in non-fossil phase.Significant contribution of small pore throats to the overall connected po-rosity is also emphasized by theMIP datawhich show that pore throat di-ameters are power-law distributed with power-law exponents D of2.42±0.07 and D=2.45±0.08 for RWEP06 and RWEP08, respec-tively (Fig. 11I, J). Because the carbonate fossils are fully embedded inmatrix, we expected comparable power-law exponents from BIB-SEM


measurements for non-fossil pores and MIP data. However, these arequite different; the contribution of small pores being significantly higherfrom results given by MIP data. This discrepancy suggests that a signifi-cant part of porosity in the matrix is not connected.

5.2. Estimation of bulk porosity from BIB-SEM microstructural data

The porosity resolvable with BIB-SEM is lower than porositiesfound in early mature Posidonia shale by Mann (1987) and measuredwith MIP. The reason for this is that MIP measures porosity connectedby pore throats down to 3 nm in diameter compared to 36 nmcorresponding to the minimum practical pore diameter imagedusing BIB-SEM (Table 1). Resolvable porosities in BIB-SEM of a REAare comparable for both samples (2.75% and 2.74% for RWEP06 andRWEP08, respectively) and significantly higher than the porosityaccessed by MIP at comparable pore throat diameter (Fig. 12A andB). In order to compare MIP porosity and BIB-SEM porosity at smallerpore sizes, the power-law distributions described above are used toextrapolate BIB-SEM pore data down to pore size of 3 nm in diameter.To do this, we assume that bulk porosity is given by ϕ06 or 08;T

BIB;3* =ϕ06 or 08;NFBIB;3* +ϕ06 or 08;F

BIB;3* (see nomenclature in Section 3.6 and Table 2)and that pores un-resolved by the BIB-SEM approach follow thesame power-law distributions defined in Section 4.5 and Fig. 10.This results in ϕ06;NF

BIB;3*=4.34% (3.41–6.01%) and ϕ08;NFBIB;3*=6.13%

(3.47–13.36%), the values between the parentheses are the extrapo-lated porosities of the outermost fits (Fig. 5). Therefore, adding poros-ity associated with the carbonate fossils (Table 4 and see Section 5.1),we obtained ϕ06;T

BIB;3*=4.80% (3.88–6.48%) and ϕ08;TBIB;3*=6.55% (3.89–

13.78%). The extrapolated BIB-SEM porosities are thus higher than



Fig. 9. SE and BSE images of the typical pore morphologies related to mineralogy in RWEP06 and RWEP08. (A) SE image of elongated pores at the interface between organic matterand matrix; a visible meso-pore inside the organic matter is indicated by arrows. (B) BSE image of (A). (C) SE image of typical polygonal pores in carbonate porous fossils with insetfor magnified details. (D) BSE image of (C). (E) SE image of equidimensional shaped pores in pyrite framboid. (F) BSE of a non-porous carbonate fossil. Dark spheres on the surfaceof the fossil is due to coating. (G) SE image of a large pore in calcite grain with euhedral faceted grain towards the center of the pore. (H) BSE image of (G). (I) SE image of a largepore in calcite grain with euhedral faceted grain towards the center of the pore and typical very elongated pore at interface with mica. (J) BSE image of (I). (K) SE image of anon-porous pyrite framboid in sample RWEP08. (L) BSE image of (K). For all pictures, pores are segmented in blue or red. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of this article.)


porosities measured by MIP (Fig. 12). These differences are discussedin the following section in order to link BIB-SEM microstructural ob-servation with MIP data.

5.3. Linkage of BIB-SEM data with MIP data resulting in a pore modelbased on microstructures

In other studies good agreement has been found between poredata acquired by imaging and mercury intrusion in other comparablestudies based on low permeable and low porous rocks (Curtis et al.,


2010, 2011a; Heath et al., 2011; Loucks et al., 2009) but can also beproblematic (Hildenbrand and Urai, 2003) if pore collapse occurs dur-ing mercury intrusion at high pressure. In the present contributionthe intrusion of mercury into the pore space is demonstrated by theincrease of sample weight after MIP experiment and hysteresis be-tween intrusion and extrusion (see Section 4.6).

In Section 5.2, we noticed that ϕ06 or 08;TBIB;36−104>ϕ06 or 08;T

MIP;36−104 andϕ06 or 08;TBIB;3* >ϕ06 or 08;T

MIP;3 (Fig. 12). Because MIP measures only connectedporosity, these discrepancies means that a significant part of the porespace resolved inBIBmosaics is not connected at pressures corresponding



Table 4Porosity inferred from BIB-SEM analyses and MIP at the practical pore resolution inmosaics covering REA, for the carbonate fossils and in the large calcite pores (LCP).

Sample BIB-SEM MIP

Mosaic Fossil LCP

RWEP06 1.99% 0.42% 0.66% 0.4±0.18%RWEP08 2.15% 0.41% 0.40% 0.5±0.18%


to the practical pore resolution. This is also corroborated byMIP data thatillustrates thatmost of the pore space (67%) is filledwithmercury in porethroats smaller than 11 nm (Fig. 11A–D) for both samples, with maxi-mum intrusion volumes measured for pore diameter of 6 and 7.2 nm(Fig. 11C, D; for RWEP06 and RWEP08, respectively). Therefore thepore space imaged in BIB mosaics is not representative of pore throatsbecause they are not resolvable down to the practical pore resolution.In other words, this suggests that the pore space imaged in BIB-SEMmosaics represents the pore bodies connected by non-resolved porethroats with sizes below 10 nm. This is in good agreement with differ-ences in mercury intrusion and extrusion curves which point to porebody-pore throat ratios between 1:1 and 2000:1 for the 3 to 7.2 nmpore throat diameter interval (Fig. 11G, H). Hence, the BIB-SEM re-solved pores are filled with mercury only at very high pressures, indi-cating a very low permeability.

Fig. 10. Pore size distribution of pore area in the carbonate fossils and non-carbonate fossilsample RWEP08. Data are plotted as the logarithm of normalized frequency of pore as a funpore area fully resolved whereas white-filled markers are pore area still detectable but nottribution and middle vertical dashed lines indicate slope break observed for pore area distrline through data is the best fit and surrounding dashed lines represent the error of the fits


In regards to these interpretations, porous carbonate fossils andlarge pores in calcite grains which contribute the most to the porosityin BIB-SEM data, (Table 4), are interpreted to be embedded in a lowpermeable and low porous matrix. We propose also that the poreconnectivity of the matrix is controlled by the clay-rich matrix. Inthis case, the power-law exponent for pore throat distribution in-ferred from MIP data (D~2.4–2.5, Fig. 11I, J) may be representativefor the clay-rich matrix.

6. Conclusions

1. Porosity resolvable by BIB-SEM is 2.75 and 2.74% for the two sam-ples RWEP 06 and RWEP08, respectively. This method can resolvepores of around 10 nm diameter, and allows qualitative and quan-titative study of porosity.

2. Using a combination of BSE and SE detectors and semi-quantitativesegmentation of the gigapixel images, we found that at the mm-scalethe representative area of the samples is around 140×140 μm2.

3. Different pore morphologies are found in fossils, calcite, organicmatter and pyrite framboids.

4. Pore areas are power-law distributed with a power-law exponentof about 2 for the pore population that excludes the fossils. Poresin the fossils show dual-power-law distribution with power-law

s phases inferred from BIB-SEM methods in mosaics I and II of sample RWEP06 and inction of the logarithm of binned pore area (Eq. (2), Fig. 5). As in Fig. 5, filled marks arefully resolvable at BIB-SEM resolution. Vertical dashed lines bound the power-law dis-ibution in carbonate fossil, equivalent pore diameter is given in the boxes below. Solid. Resulting power-law exponent (D) and Log [C] are given in each plot.



Fig. 11. Results from MIP measurement performed in both samples RWEP06 (left) and RWEP08 (right). (A–B) Cumulative porosity as a function of pore throat diameter. Raw dataare indicated by filled markers and corrected data by hollow markers. Indicated errors bars are based on cut-offs used to correct the data, see text for more details. (C–D) Contri-bution to the total porosity as a function of pore throat diameter. Raw data are indicated by filled markers and corrected data by hollow markers. The vertical dashed line indicatethe pore throat which contributes for the most to the total porosity. (E–F) Normalized cumulative filled pore space versus pore throat diameter for the corrected intrusion and ex-trusion measurements (G–H) Pore body to pore throat ratio versus pore throat diameter. (I–J) Logarithm of the normalized frequency of pore throat diameter versus the logarithmof the pore throat diameter. The solid line represents the best fit of the data. Resulting power-law exponents (D) are given in each plot.


Please cite this article as: Klaver, J., et al., BIB-SEM study of the pore space morphology in early mature Posidonia Shale from the Hils area,Germany, Int. J. Coal Geol. (2012), doi:10.1016/j.coal.2012.06.012


Fig. 12. Comparison cumulative porosity inferred by MIP measurement with porosity from BIB-SEM approach at the practical pore resolution (filled marker) and extrapolated downto the pore resolved by MIP (hollow marker) as a function of equivalent pore (throat) diameters for the sample RWEP06 (A) and RWEP08 (B).


exponents of about 1 for pores smaller then bin middle24,600 nm2 (deq=177 nm) and 3 for pore areas above this size.

5. Extrapolation of power-law for pore area unresolved by BIB-SEMallows estimation of total porosity. The total porosity is estimatedto range from 3.9–6.5% and 3.9–13.8%, with mean values of 4.80%and 6.45%, which can be compared with 3.4–3.7 and 3.3–3.6% asmeasured by Mercury porosimetry for RWEP06 and RWEP08,respectively.

6. Difference between porosity inferred by MIP and BIB-SEM areinterpreted to reflect the unconnected part of the porosity resolv-able by MIP. Both microstructural investigations and MIP data sug-gest a significant ink-bottle effect.

7. The pore network consists of large pores embedded in theclay-rich matrix. In the clay-rich matrix pores are connected bypore throats smaller than 10 nm.

Acknowledgments

We gratefully acknowledge an anonymous reviewer and AndrewAplin for their constructive comments which much improved themanuscript. The authors would like to show their gratitude to U.Wollenberg for his help with the XRD analyses, BIB-milling and SEMimaging and to M. Houben and S. Hemes for the pleasant and fruitfuldiscussions during the project. We would like to thank K.A. Wilkinsfor his help with the pore segmentation, and thank Dr. A. Aretz forthe AFM analyses.

References

Ambrose, R.J., Hartman, R.C., Diaz-Campos, M., Akkutlu, I.Y., Sondergeld, C.H., 2010.New Pore-scale Considerations for Shale Gas in Place Calculations. SPE Unconven-tional Gas Conference. Society of Petroleum Engineers, Pittsburgh, Pennsylvania,USA, p. 17. SPE 131772.

Aplin, A.C., Matenaar, I.F., McCarty, D.K., van der Pluijm, B.A., 2006. Influence of me-chanical compaction and clay mineral diagenesis on the microfabric and pore-scale properties of deep-water Gulf of Mexico mudstones. Clays and Clay Minerals54, 500–514.

Barrett, E.P., Joyner, L.G., Halenda, P.P., 1951. The determination of pore volume andarea distributions in porous substances. I. Computations from nitrogen isotherms.Journal of the American Chemical Society 73, 373–380.

Bartenstein, H., Teichmüller, M., Teichmüller, R., 1971. Die Umwandlung derorganischen Substanz im Dach des Bramscher Massivs. Fortschritte in der Geologievon Rhein- land und Westfalen 18, 501–538.

Bernard, S., Horsfield, B., Schulz, H.-M., Wirth, R., Schreiber, A., Sherwood, N., 2012.Geochemical evolution of organic-rich shales with increasing maturity: a STXMand TEM study of the Posidonia Shale (Lower Toarcian, northern Germany). Ma-rine and Petroleum Geology 31, 70–89.

Bour, I., Mattioli, E., Pittet, B., 2007. Nannofacies analysis as a tool to reconstructpaleoenvironmental changes during the Early Toarcian anoxic event. Palaeogeography,Palaeoclimatology, Palaeoecology 249, 58–79.


Broichhausen, H., Littke, R., Hantschel, T., 2005. Mudstone compaction and its influenceon overpressure generation, elucidated by a 3D case study in the North Sea. Inter-national Journal of Earth Sciences 94, 956–978.

Brunauer, S., Emmett, P.H., Teller, E., 1938. Adsorption of gases in multimolecularlayers. Journal of the American Chemical Society 60, 309–319.

Bustin, R.M., Bustin, A.M.M., Cui, X., Ross, D.J.K., Murthy Pathi, V.S., 2008. Impact ofShale Properties on Pore Structure and Storage Characteristics. SPE Shale Gas Pro-duction Conference. SPE, Fort Worth, Texas, USA, p. 28. SPE 119892.

Cerepi, A., Durand, C., Brosse, E., 2002. Pore microgeometry analysis in low-resistivitysandstone reservoirs. Journal of Petroleum Science and Engineering 35, 205–232.

Chalmers, G.R., Bustin, R.M., 2008. Lower Cretaceous gas shales in northeastern BritishColumbia, Part I: geological controls on methane sorption capacity. Bulletin ofCanadian Petroleum Geology 56, 1–21.

Chalmers, G., Bustin, R.M., Power, I., 2009. A Pore by Any Other Name Would Be asSmall: the Importance of Meso- and Microporosity in Shale Gas Capacity. AAPG An-nual Convention and Exhibition, Denver, Colorado, p. 1.

Chalmers, G.R., Bustin, R.M., Power, I.M., 2012. Characterization of gas shale pore sys-tems by porosimetry, pycnometry, surface area, and field emission scanning elec-tron microscopy/transmission electron microscopy image analyses: examplesfrom the Barnett, Woodford, Haynesville, Marcellus, and Doig units. AAPG Bulletin96, 1099–1119.

Curtis, M.E., Ambrose, R.J., Sondergeld, C.H., Rai, C.S., 2010. Structural Characterizationof Gas Shales on the Micro- and Nano-Scales. Canadian Unconventional Resourcesand International Petroleum Conference. Society of Petroleum Engineers, Calgary,Alberta, Canada, p. 15. SPE 137693.

Curtis, M.E., Ambrose, R.J., Sondergeld, C.H., Rai, C.S., 2011a. Transmission and ScanningElectron Microscopy Investigation of Pore Connectivity of Gas Shales on the Nano-scale. North American Unconventional Gas Conference and Exhibition. Society ofPetroleum Engineers, The Woodlands, Texas, USA, p. 10. SPE 144391.

Curtis, M.E., Ambrose, R.J., Sondergeld, C.H., Rai, C.S., 2011b. Investigation of the Rela-tionship Between Organic Porosity and Thermal Maturity in The Marcellus Shale.North American Unconventional Gas Conference and Exhibition, The Woodlands,Texas, USA, p. 4. SPE 144370.

Curtis, M.E., Sondergeld, C.H., Ambrose, R.J., Rai, C.S., 2012. Microstructural investiga-tion of gas shales in two and three dimensions using nanometer-scale resolutionimaging. AAPG Bulletin 96, 665–677.

Desbois, G., Urai, J.L., Kukla, P.A., 2009. Morphology of the pore space in claystones–evidence from BIB/FIB ion beam sectioning and cryo-SEM observations. eEarth Dis-cussions 4, 1–19.

Desbois, G., Urai, J.L., Kukla, P.A., Konstanty, J., Baerle, C., 2011. High-resolution 3D fab-ric and porosity model in a tight gas sandstone reservoir: a new approach to inves-tigate microstructures from mm- to nm-scale combining argon beam cross-sectioning and SEM imaging. Journal of Petroleum Science and Engineering 78,243–257.

Desbois, G., Urai, J., Kukla, P., Wollenberg, U., Pérez-Willard, F., Radí, Z., Riholm, S.,2012. Distribution of brine in grain boundaries during static recrystallization inwet, synthetic halite: insight from broad ion beam sectioning and SEM observa-tion at cryogenic temperature. Contributions to Mineralogy and Petrology 163,19–31.

Espitalié, J., Laporte, J.L., Madec, M., Marquis, F., Leplat, P., Paulet, J., Boutefeu, A., 1977.Méthode rapide de caractérisation des roches mètres, de leur potentiel pétrolier etde leur degré d'évolution. Oil & Gas Science and Technology/ Revue de l'InstitutFrancais du Petrole 32, 23–42.

Heath, J.E., Dewers, T.A., McPherson, B.J.O.L., Petrusak, R., Chidsey, T.C., Rinehart, A.J.,Mozley, P.S., 2011. Pore networks in continental and marine mudstones: charac-teristics and controls on sealing behavior. Geosphere 7, 429–454.

Hemes, S., Desbois, G., Urai, J.L., Craen, M.d., Honty, M., 2011. Comparative study on po-rosity in fine- and coarse-grained Boom Clay samples (Mol-Dessel reference site,Belgium) SCK•CEN. , p. 46.

Hildenbrand, A., Urai, J.L., 2003. Investigation of the morphology of pore space in mud-stones—first results. Marine and Petroleum Geology 20, 1185–1200.




Hildenbrand, A., Krooss, B., Urai, J., 2005. Relationship Between Pore Structure andFluid Transport in Argillaceous Rocks. In: Gladwell, G., Huyghe, J.M., Raats, P.,Cowin, S. (Eds.), IUTAM Symposium on Physicochemical and Electromechanical In-teractions in Porous Media. : Solid Mechanics and Its Applications. Springer,Netherlands, pp. 231–237.

Holzer, L., Cantoni, M., 2011. Review of FIB-tomography. In: Utke, I., Moshkalev, S.,Russell, P. (Eds.), Nanofabrication with Focused Ion and Electron Beam: Principlesand Applications. Oxford University Press, NY, USA, p. 830.

Janssen, C., Wirth, R., Reinicke, A., Rybacki, E., Naumann, R., Wenk, H.-R., Dresen, G.,2011. Nanoscale porosity in SAFOD core samples (San Andreas Fault). Earth andPlanetary Science Letters 301, 179–189.

Jochum, J., Friedrich, G., Leythaeuser, D., Littke, R., Ropertz, B., 1995. Hydrocarbon-bearingfluid inclusions in calcite-filled horizontal fractures frommature Posidonia Shale (HilsSyncline, NW Germany). Ore Geology Reviews 9, 363–370.

Kahle, M., Kleber, M., Jahn, R., 2002. Review of XRD-based quantitative analyses of clayminerals in soils: the suitability of mineral intensity factors. Geoderma 109,191–205.

Kameda, A., Dvorkin, J., Keehm, Y., Nur, A., Bosl, W., 2006. Permeability–porosity trans-forms from small sandstone fragments. Geophysics 71, N11–N19.

Keller, L.M., Holzer, L., Wepf, R., Gasser, P., 2011. 3D geometry and topology of porepathways in Opalinus clay: implications for mass transport. Applied Clay Science52, 85–95.

Koch, G., Arnemann, H., 1975. Die Inkohlung in Gesteinen des Rhät und Lias imsüdlichen Nordwestdeutschland. Geologisches Jahrbuch A29, 45–55.

Lexa, O., 2010. PolyLX–the MATLAB™ toolbox for quantitative analysis of microstruc-tures. http://petrol.natur.cuni.cz/~ondro/polylx:home201012-12-2011.

Littke, R., Rullkötter, J., 1987. Mikroskopische und makroskopische Unterschiedezwischen Profilen unreifen und reifen Posidonienschiefers aus der Hilsmulde. Fa-cies 17, 171–179.

Littke, R., Baker, D.R., Leythaeuser, D., 1988. Microscopic and sedimentologic evidencefor the generation and migration of hydrocarbons in Toarcian source rocks of dif-ferent maturities. Organic Geochemistry 13, 549–559.

Littke, R., Klussmann, U., Krooss, B., Leythaeuser, D., 1991a. Quantification of loss ofcalcite, pyrite, and organic matter due to weathering of Toarcian black shales andeffects on kerogen and bitumen characteristics. Geochimica et CosmochimicaActa 55, 3369–3378.

Littke, R., Leythaeuser, D., Rullkötter, J., Baker, D.R., 1991b. Keys to the depositionalhistory of the Posidonia Shale (Toarcian) in the Hils Syncline, northern Germany.Geological Society, London, Special Publications 58, 311–333.

Littke, R., Urai, J.L., Uffmann, A.K., Risvanis, F., 2012. Reflectance of dispersed vitrinite inPalaeozoic rocks with and without cleavage: Implications for burial and thermalhistory modeling in the Devonian of Rursee area, northern Rhenish Massif, Germany.International Journal of Coal Geology 89, 41–50.

Loucks, R.G., Reed, R.M., Ruppel, S.C., Jarvie, D.M., 2009. Morphology, genesis, and dis-tribution of nanometer-scale pores in siliceous mudstones of the MississippianBarnett shale. Journal of Sedimentary Research 79, 848–861.

Loucks, R.G., Reed, R.M., Ruppel, S.C., Hammes, U., 2012. Spectrum of pore types andnetworks in mudrocks and a descriptive classification for matrix-related mudrockpores. AAPG Bulletin 96, 1071–1098.

Mann, U., 1987. Veränderung vonMineralmatrix und Porosität eines Erdölmuttergesteinsdurch einen Intrusivkörper (Lias epsilon 2–3: Hilsmulde, NW-Deutschland). Facies17, 181–188.

Mann, U., Müller, P.J., 1988. Source rock evaluation by well log analysis (LowerToarcian, Hils syncline). Organic Geochemistry 13, 109–119.

Milner, M., McLin, R., Petriello, J., 2010. Imaging Texture and Porosity in Mudstones andShales: Comparison of Secondary and Ion-Milled Backscatter SEM Methods.Canadian Unconventional Resources and International Petroleum Conference. So-ciety of Petroleum Engineers, Calgary, Alberta, Canada, p. 5. SPE 138975.

Moro, F., Böhni, H., 2002. Ink-bottle effect in mercury intrusion porosimetry of cement-based materials. Journal of Colloid and Interface Science 246, 135–149.

Münch, B., Holzer, L., 2008. Contradicting geometrical concepts in pore size analysisattained with electron microscopy and mercury intrusion. Journal of the AmericanCeramic Society 91, 4059–4067.

Nelson, P.H., 2009. Pore-throat sizes in sandstones, tight sandstones, and shales. AAPGBulletin 93, 329–340.


Passey, Q.R., Bohacs, K.M., Esch, W.L., Klimentidis, R., Sinha, S., 2010. From oil-pronerock to gas-producing shale reservoir—geological and petrophysical characteriza-tion of unconventional shale-gas reservoirs. International Oil and Gas Conferenceand Exhibition in China. Society of Petroleum Engineers, Beijing, China, p. 29. SPE131350.

Pearson, F.J., 1999. What is the porosity of a mudrock? In: Aplin, A.C., Fleet, A.J.,Macquaker, J.H.S. (Eds.), Muds and Mudstones: Physical and Fluid Flow Properties:Geological society, London, Special publication, 158, pp. 9–21.

Petmecky, S., Meier, L., Reiser, H., Littke, R., 1999. High thermal maturity in the LowerSaxony Basin: intrusion or deep burial? Tectonophysics 304, 317–344.

Prinz, D., Littke, R., 2005. Development of the micro- and ultramicroporous structure ofcoals with rank as deduced from the accessibility to water. Fuel 84, 1645–1652.

Ross, D.J.K., Bustin, R.M., 2009. The importance of shale composition and pore structureupon gas storage potential of shale gas reservoirs. Marine and Petroleum Geology26, 916–927.

Rouquerol, J., Avnir, D., Fairbridge, C.W., Everett, D.H., Haynes, J.M., Pernicone, N.,Ramsay, J.D.F., Sing, K.S.W., Unger, K.K., 1994. Recommendations for the character-ization of porous solids (Technical Report). Pure and Applied Chemistry 66, 1739.

Rullkötter, J., Leythaeuser, D., Horsfield, B., Littke, R., Mann, U., Müller, P.J., Radke, M.,Schaefer, R.G., Schenk, H.J., Schwochau, K., Witte, E.G., Welte, D.H., 1988. Organicmatter maturation under the influence of a deep intrusive heat source: a naturalexperiment for quantitation of hydrocarbon generation and expulsion from a pe-troleum source rock (Toarcian shale, northern Germany). Organic Geochemistry13, 847–856.

Sachse, V.F., Littke, R., Heim, S., Kluth, O., Schober, J., Boutib, L., Jabour, H., Perssen, F.,Sindern, S., 2011. Petroleum source rocks of the Tarfaya Basin and adjacent areas,Morocco. Organic Geochemistry 42, 209–227.

Schieber, J., 2010. Common Themes in the Formation and Preservation of Intrinsic Po-rosity in Shales and Mudstones–Illustrated with Examples Across the Phanerozoic.SPE Unconventional Gas Conference. Society of Petroleum Engineers, Pittsburgh,Pennsylvania, USA, p. 10. SPE 132370.

Schneider, J., Flemings, P.B., Day-Stirrat, R.J., Germaine, J.T., 2011. Insights into pore-scale controls on mudstone permeability through resedimentation experiments.Geology 39, 1011–1014.

Schull, C.G., 1948. The determination of pore size distribution from gas adsorption data.Journal of the American Chemical Society 70, 1405–1410.

Sigal, R.F., 2009. A methodology for blank and conformance corrections for high pres-sure mercury porosimetry. Measurement Science and Technology 20, 12.

Sisk, C., Diaz, E., Walls, J., Grader, A., Suhrer, M., 2010. 3D Visualization and Classifica-tion of Pore Structure and Pore Filling in Gas Shales. SPE Annual Technical Confer-ence and Exhibition. Society of Petroleum Engineers, Florence, Italy, p. 4. SPE134582.

Slatt, R.M., O'Brien, N.R., 2011. Pore types in the Barnett and Woodford gas shales: con-tribution to understanding gas storage and migration pathways in fine-grainedrocks. AAPG Bulletin 95, 2017–2030.

Sondergeld, C.H., Ambrose, R.J., Rai, C.S., Moncrieff, J., 2010. Micro-Structural Studies ofGas Shales. SPE Unconventional Gas Conference. Society of Petroleum Engineers,Pittsburgh, Pennsylvania, USA, p. 17. SPE 131771.

Taylor, G.H., Teichmüller, M., Davis, A., Diessel, C.F.K., Littke, R., Robert, P., 1998. OrganicPetrology. Gebr. Borntraeger, Berlin-Stuttgart, p. 704 S.

Underwood, E.E., 1970. Quantitative stereology. Addison-Wesley Pub., Co.Veiga de Oliveira, L.C., Perilli, N., Duarte, L.V., 2007. Calcareous nannofossil assemblages

around the Pliensbachian/Toarcian boundary in the reference section of Peniche(Portugal). Ciências da Terra (UNL) 16, 45–50.

Wang, F.P., Reed, R.M., 2009. Pore Networks and Fluid Flow in Gas Shales. SPE AnnualTechnical Conference and Exhibition. Society of Petroleum Engineers, NewOrleans, Louisiana, p. 8. SPE 124253.

Washburn, E.W., 1921. The dynamics of capillary flow. Physical Review 17, 273.Webb, P.A., 2001. An introduction to the physical characterization of materials by mer-

cury intrusion porosimetry with emphasis on reduction and presentation of exper-imental data. Micromeritics Instrument Corp., Norcross, Georgia.


http://petrol.natur.cuni.cz/~ondro/polylx:home


Documents

International Journal of Coal Geology · Posidonia Shale samples from the Hils Syncline, Germany, using BIB-SEM. In addition, the bulk porosity is estimated and compared to MIP. Here