1
Comparison of pore fabric based on high-resolution X-ray computed tomography and magnetic pore fabric in sedimentary rocks Yi Zhou*, Michele Pugnetti*, Anneleen Foubert**, Christoph Neururer** & Andrea R. Biedermann* *Institute of Geological Sciences, University of Bern, Baltzerstrasse 1+3, CH-3012 Bern, Switzerland **Department of Geosciences, University of Fribourg, Chemin du Musée 6, CH-1700 Fribourg, Switzerland The pore size and geometry, as well as their connectivity determine the permeability and preferred permeability direction. X-ray micro-tomography (XRCT) is a widely used technique to generate digital rock models, visualize and quantify pore shape and size distributions, and the connectivity of pores with a spatial resolution on the order of 10 microns. Magnetic pore fabrics (MPF), obtained by impregnating ferrofluid prior to measuring anisotropy of magnetic susceptibility could be complementary to existing techniques and capture smaller pores. This study is aimed at quantitatively analyzing relationships between pore fabric or permeability anisotropy and MPF. Plio-Pleistocene calcarenite (Apulia, Italy) with ~50% porosity, ~300 μm average equivalent diameter of pores and complex pore structure. Upper Marine Molasse sandstone (OMM, Belpberg, Switzerland) with 5- 20% porosity, ~150 μm average equivalent diameter of pores and relatively homogeneous pore structure. 1. Introduction 2.1 Samples 2.3 Magnetic pore fabric method Impregnation by vacuum and fifteen-directions measurement scheme to determine the full susceptibility tensor. 2. Methods 2.2 XRCT XRCT scanning and full-tensor permeability simulations in 6 directions 4. Summary 3.2 Sample: molasse Avizo 2019.4 pore network modelling User Guide from www.fei.com 15-position measuring 3. Results 3.1 Sample: calcarenite MI-1-Z3 (calcarenite) Total pore shape ellipsoid Minimum pore dimension Maximum pore dimension Max Min Intermediate MI-1-Z3 (calcarenite) D1112Y (molasse) Pores ≥4×4×4voxels Orientation density functions of maximum and minimum pore axes are affected by resolution artefacts unless smaller pores are filtered out. Total shape ellipsoid reflects preferred orientation of pore shape, and is largely unaffected by artefacts related to unresolved pores. Permeabilities can be simulated from the XRCT data, but can be still improved in accuracy and need verification against measurements. The orientation of the MPF maximum axis correlates with the maximum dimension of the mean shape ellipsoid at 95% confidence. The MPF anisotropy degree is generally smaller than that of the total shape ellipsoid. simulations in 6 directions Impregnation Orientations and orientation density functions of pore axes All measured pores Pores ≥4×4×4voxels Pores ≥16×16×16voxels Minimum pore dimension Maximum pore dimension Orientations and orientation density functions of pore axes Multiples of random distribution Max Min Intermediate Upper Upper Upper Upper Upper Upper Upper Upper Upper Upper Upper Upper Upper Upper Upper Multiples of random distribution Min Intermediate Total pore shape ellipsoid Upper 0.5 Multiples of random distribution 90 70 50 30 10 4.5 3.5 2.5 1.5 Multiples of random distribution Upper 4.5 3.5 2.5 1.5 0.5 Upper Max Int Min 90 (Y) 0 (X) Max Int Min 90 (Z) Upper Mean Individual data 95% confidence 3.5 3 2.5 2 1.5 1 0.5 Max Max Min Intermediate Upper Simulated permeability Max Int Min 90 (Y) 0 (X) Max Int Min Upper 90 (Z) Samples drilled in different directions MI-1-Z3 Z X Y MI-2-Y3 0 (X) 90 (Y) Max Int Min 90 (Z) Upper Max Int Min MI-1-Z3 MI-2-Y3 MI-3-X15 Max Min Intermediate Max Min Intermediate Max Min Intermediate Comparison on total shape ellipsoid Max Int Min 0 (X) Max Int Min 90 (Z) Upper 90 (Y) k1=max eigenvalue; k2=intermediate eigenvalue; k3=min eigenvalue; P = k1/k3, anisotropy degree; U = (2*k2-k1-k3)/(k1-k3), shape Checking for consistency of pore shape in three perpendicular cores Simulated permeability MPF P-U Upper Bootstrapped total shape ellipsoid Max Int Min 90 (Y) 0 (X) Max Int Min Upper 90 (Z) Mean Individual data 95% confidence Bootstrapped total shape ellipsoid P-U Upper

Comparison of pore fabric based on high-resolution X-ray ... · Comparison of pore fabric based on high-resolution X-ray computed tomography and magnetic pore fabric in sedimentary

  • Upload
    others

  • View
    1

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Comparison of pore fabric based on high-resolution X-ray ... · Comparison of pore fabric based on high-resolution X-ray computed tomography and magnetic pore fabric in sedimentary

Comparison of pore fabric based on high-resolution X-ray computed tomography and magnetic pore

fabric in sedimentary rocks Yi Zhou*, Michele Pugnetti*, Anneleen Foubert**, Christoph Neururer** & Andrea R. Biedermann* *Institute of Geological Sciences, University of Bern, Baltzerstrasse 1+3, CH-3012 Bern, Switzerland

**Department of Geosciences, University of Fribourg, Chemin du Musée 6, CH-1700 Fribourg, Switzerland

• The pore size and geometry, as well as their connectivity determine the permeability and preferred permeability direction.

• X-ray micro-tomography (XRCT) is a widely used technique to generate digital rock models, visualize and quantify pore shape and size distributions, and the connectivity of pores with a spatial resolution on the order of 10 microns.

• Magnetic pore fabrics (MPF), obtained by impregnating ferrofluid prior to measuring anisotropy of magnetic susceptibility could be complementary to existing techniques and capture smaller pores.

• This study is aimed at quantitatively analyzing relationships between pore fabric or permeability anisotropy and MPF.

• Plio-Pleistocene calcarenite (Apulia, Italy) with ~50% porosity, ~300 μm average equivalent diameter of pores and complex pore structure.

• Upper Marine Molasse sandstone (OMM, Belpberg, Switzerland) with 5-20% porosity, ~150 μm average equivalent diameter of pores and relatively homogeneous pore structure.

1. Introduction

2.1 Samples

2.3 Magnetic pore fabric method

• Impregnation by vacuum and fifteen-directions measurement scheme to determine the full susceptibility tensor.

2. Methods

2.2 XRCT• XRCT scanning and full-tensor permeability simulations in 6 directions

4. Summary

3.2 Sample: molasseAvizo 2019.4 pore network modellingUser Guide from www.fei.com

15-position measuring

3. Results

3.1 Sample: calcarenite• MI-1-Z3 (calcarenite)

Total pore shape ellipsoidMinimum pore dimensionMaximum pore dimension

Max

Min

Intermediate

• MI-1-Z3 (calcarenite)

• D1112Y (molasse)

Pore

s ≥4

×4×4

voxe

ls

• Orientation density functions of maximum and minimum pore axes are affected by resolution artefacts unless smaller pores are filtered out.

• Total shape ellipsoid reflects preferred orientation of pore shape, and is largely unaffected by artefacts related to unresolved pores.

• Permeabilities can be simulated from the XRCT data, but can be still improved in accuracy and need verification against measurements.

• The orientation of the MPF maximum axis correlates with the maximum dimension of the mean shape ellipsoid at 95% confidence.

• The MPF anisotropy degree is generally smaller than that of the total shape ellipsoid.

simulations in 6 directions

Impregnation

Orientations and orientation density functions of pore axes

All

mea

sure

d p

ore

sPo

res

≥4×4

×4vo

xels

Pore

s ≥1

6×1

6×1

6vo

xels

Minimum pore dimensionMaximum pore dimensionOrientations and orientation density functions of pore axes

Mu

ltiples o

f rand

om

d

istribu

tion

Max

Min

Intermediate

Upper Upper

Upper

Upper Upper

Upper

Upper Upper Upper

UpperUpper

Upper Upper Upper Upper

Mu

ltiples o

f rand

om

d

istribu

tion

Min

Intermediate

Total pore shape ellipsoid

Upper

0.5

Mu

ltiples o

f rand

om

d

istribu

tion

90

70

50

30

10

4.5

3.5

2.5

1.5

Mu

ltiples o

f rand

om

d

istribu

tion

Upper4.5

3.5

2.5

1.5

0.5

Upper

Max

Int

Min

90 (Y)

0 (X)• Max• Int• Min

90 (Z)

Upper

MeanIndividual data95% confidence

3.5

3

2.5

2

1.5

1

0.5

Max

Max

Min

Intermediate

Upper

Simulated permeability

Max

Int

Min90 (Y)

0 (X)

• Max• Int• Min

Upper

90 (Z)

Samples drilled in different directions

MI-1-Z3

Z

X

Y

MI-2-Y3

0 (X)

90 (Y)

• Max• Int• Min

90 (Z)

Upper

Max

IntMin

• MI-1-Z3• MI-2-Y3• MI-3-X15

Max

Min

Intermediate

Max

Min

Intermediate

Max

Min

Intermediate

Comparison ontotal shape ellipsoid

Max

Int

Min

0 (X) • Max• Int• Min

90 (Z)

Upper

90 (Y)

k1=max eigenvalue; k2=intermediate eigenvalue; k3=min eigenvalue;P = k1/k3, anisotropy degree;U = (2*k2-k1-k3)/(k1-k3), shape

• Checking for consistency of pore shape in three perpendicular cores

Simulated permeabilityMPF

P-U

Upper

Bootstrapped total shape ellipsoid

Max

Int

Min

90 (Y)

0 (X)

• Max• Int• Min

Upper

90 (Z)

MeanIndividual data95% confidence

Bootstrapped total shape ellipsoid

P-U

Upper