Radioactive Waste ManagementNEA/RWM/CLAYCLUB(2013)1May 2013www.oecd-nea.org

NEA CLAY CLUB Workshop ProceedingsKarlsruhe, Germany6-8 September 2011

Clay characterisation from nanoscopic to microscopic resolution

NEA

109

Qualitative and Quantitative Characterization of Porosity in a Low Porous and Low

Permeable Organic Rich Shale by Combining Broad Ion Beam and Scanning Electron

Microscopy (BIB-SEM)

Jop Klaver1*, Guillaume Desbois 1, Janos L. Urai1

1 Structural Geology, Tectonics and Geomechanics, RWTH Aachen University (DE)

* Corresponding author: j.klaver@ged.rwth-aachen.de

Abstract

This contribution focuses on the characterization of porosity in low porous shale using a broad ion beam (BIB) polishing technique combined with a conventional scanning electron microscopy (SEM). Porosity was traced in certain representative elementary areas (REA) and pores detected are segmented from mosaics of secondary electron (SE) images. Traced pores could be classified into two major pore‐size classes. Relative large pores (> 0.5 μm2) were found in the organic matter and matrix. They contribute strongly to the overall porosity con-tent of the shale. Nevertheless the far majority of the pores traced have equivalent radius less than 400 nm. Including the latter pore class, the imaged porosity from both samples gives similar results in the order of < 1 %.

Introduction

This combined BIB-SEM technique of imaging and mapping of pores in shales enables to work on large areas and to actually see the pores and study the pore morphology. With the use of a BIB cross‐section polisher the area of the polished surface increased by a factor of sev-eral thousand compared to focused ion beam (FIB) techniques, enclosing an area of roughly 1-2 mm2 and allows working on REA. Conventional SEM imaging allows analysis of pores down to the nanometer scale. Two adjacent organic rich core samples from a shale interval (3.5 km depth) were investigated. From each sample one BIB polished cross-section was made and pores were segmented from several mosaics consisting of hundreds of images. Method and technique

Studied material was stored in foil at room temperature and atmospheric pressure and stored

in a tight plastic barrel. From the air-dried sample material, small pieces (~0.5 x 0.5 x 0.5

cm3) were cut off with a low speed diamond saw machine (Figure 1). After cutting, the sam-

110

ple was pre-polished using sandpaper reducing surface roughness down to 10 m. Pre-

polishing maximizes the quality of the cutting of the BIB cross-section. Eventual damage on

the surface of the shale sample created by the pre-polishing will be removed by the BIB. Sub-

samples were preserved in an exicator under low vacuum.

Figure 1. Picture of shale investigated (A). Electric low speed drill (B) used for sawing of the shale with diamond saw (pop-up). A polish tool (C) was used to create perpendicular edges with sandpaper (D). JEOL cross-section BIB polisher SM-09010 and sample setup and Atomic Force Microscopy (AFM) plot of a scanned BIB-polished surface (E).

The dry subsample was put in the BIB cross-section polisher for the argon polishing. On top

of the sample a mask is placed and argon ions mill into the sample that is not covered by the

mask. In 8 hours of milling under vacuum (1 x 10-3

-10-4

Pa), a 1 mm2 polished cross-section

was created with an amplitude less than 5 nanometers (Figure 1E). Here after the polished

cross-section was coated to prevent charging by the SEM. Imaging in the SEM was done un-

der high magnification with the back scatter electron (BSE) detector and the secondary elec-

tron detector (Inlens or SE2) with a pixel size of a few nanometers for the SE detectors. Minor

micro-scale cracks was interpreted as ex-situ damage and subtracted during analysis. REA

mosaics were created by stitching hundreds of images together with a bicubic interpolation

algorithm in Autopano 2 (http://www.kolor.com/image-stitching-software-autopano-

giga.html). Pores were traced manually in ArcGIS 10 from mosaics of SE images. After pore

segmentation, shapes were analysed in MATLAB with PolyLX (Lexa (2010)). Classification

and mapping of mineral phases were done using grey values from BSE images verified and

calibrated by a combination of energy-dispersive x-ray spectroscopy (EDX) and bulk X-ray

diffraction (XRD) analyses. REA was determined with the box counting method applied on

the classified BSE mosaic and turned out to be 80 x 80 m2.

BIB-SEM technique has been applied recently on salt and clays too (Desbois et al. (2009),

Desbois et al. (2011), Hemes et al. (2010) and Houben et al.(2011)). Challenges that needs to

be overcome are serial cross-sectioning to get a better understanding of the 3D pore structure.

Sample

Mask

Polished cross-section

Ion beam irridation area

Ion Beam

BIB

Z (n

m)

X (nm) Y (nm) 0 0

600

1200 600

1200

-5

5

0

Sample

Sample-holder

A B C D

E

111

An appropriate automated pore detection algorithm has to be designed to deal with the huge

amount of pores in images covering REA.

Results Quantitative analyses

Traced pores in both samples resulted in a very low visible porosity in all the different mosa-

ics down to the different SEM resolutions. 99% of the pores are in the range of 20-400nm

equivalent radius (req, Figure 2A), and porosities are the same in this range for both samples.

However most of the pore space, 82% in cross-section 1 and 42% in cross-section 2 are in the

1% largest (≥400nm req) pore size range. These largest pores are predominantly found in the

clay rich matrix, and organic matter. Most pores (>6600) were found in the clay rich matrix,

and results in an average relative porosity of 0.12% at the magnifications used. Pores together

have a power law distribution in all the phases over three orders of magnitude of pore sizes

(Figure 2B) and this infers a fractal dimension (Mandelbrot (1982)).

Figure 2. A. Equivalent pore radius distributions of both cross-sections (CS 1 and 2) at different magnifications. B. Power law exponent (gradient of the fitted line, given in the boxes) of seg-mented pores in the clay rich matrix traced at different magnifications.

Qualitative investigation

Pores traced in the clay rich matrix are isolated and therefore not connected at the resolutions

used (Figure 3A). However intragranular pores in pyrite and rutile are clustered and are con-

nected (Figure 3B and C). Pore orientation of the organic matter is sub-parallel with the bed-

ding and most occurs at the interface of the organic matter and the surrounding phase (Figure

3D). A pore throat is seen in the clay-rich matrix (Figure 3E).

A 10kx B 78kx C 10kx 10kx E 200kx

~5nm

4 m 400nm 4 m 8 m 200nm

D

CS1

CS2

CS1

CS2

A B

112

Figure 3. A. Isolated relative small pores (arrows) in the clay-rich matrix. B. Intragranular pores in rutile and pores in pyrite framboids (C). D. Elongated pores along organic matter (BSE image). E. Pore throat between pores in clay rich matrix.

Summary and Conclusions

Porosity in both samples is very low (<1%) at resolutions used and are similar in the smaller

pore range (req < 400nm). The pore size distributions show similar power law exponents for

each phase and at different magnifications. Each phase has its own typical pore characteristics,

regardless of the sample. BIB-SEM is a very suitable technique for the study of porosity in

shale.

Acknowledgement

We would like to acknowledge U. Wollenberg for his help with the XRD analyses, BIB mill-

ing and SEM imaging and Dr. rer. nat. Anke Aretz for the AFM analyses.

References

Desbois, G., Urai, J. L., and Kukla, P. A. (2009). Morphology of the pore space in claystones – evidence from BIB/FIB ion beam sectioning and cryo-SEM observations. eEarth, 4, 15-22.

Hemes, S., Desbois, G., Urai, J. L., De Craen, M. and Honty, M. (2011). Comparative study on porosity in fine- and coarse-grained Boom Clay samples (Mol-Dessel reference site, Belgium).- Mol, Belgium: SCK•CEN, 2011.- 46 p.- (External Report of the Belgian Nuclear Research Centre; ER-157.

Houben, M.E., Desbois, G., Urai, J.L. (submit.). Pore morphology and distribution in the shaly facies of Opalinus Clay (Mont Terri, Switzerland): insights from representative 2D BIB-SEM investigations on the mm to nm scale. Applied Clay Science.

Lexa, O. (2010). PolyLX - the MATLAB™ toolbox for quantitative analysis of microstructures. Ondrej Lexa Website.

Mandelbrot, B. (1982). The Fractal Geometry of Nature, W. H. Freeman.