ShaleXEnvironmentDissemination Event TAMUQ –18 March 2018 · 2018. 3. 28. · Ronny Pini, [email protected] March 2018 High-pressure gas sorption studies on shales Ronny Pini

Ronny Pini, [email protected] March 2018

High-pressure gas sorption studies on shales

Ronny PiniDepartment of Chemical Engineering

Imperial College London

ShaleXEnvironment Dissemination EventTAMUQ – 18th March 2018

Ronny Pini – [email protected] - Imperial College London 2

Gas adsorption in shale

• Most of the porosity in shale is located in pores < 50 nm–What is the accessible pore space? How does it influence fluid behaviour?

–Gas adsorption contributes to Gas-In-Place (GIP); gas production is limited by the ability to desorb gas from the tight matrix

• Measurement of HP adsorption isotherms is a technical challenge– Inter-laboratory comparisons are difficult (lack of standards)

[e.g., Lancaster 1993 … Gasparik 2014]

– Interactions with shale constituents (e.g., clays and OM) are complex[e.g., Barrer 1954 … Loring 2012; Busch 2014]

–Fluid densification and/or depletion? [e.g., Rother 2012; Schaef 2014; Jeon 2014]

Of practical relevance, but quantification is still a challenge


= ρ Ahφ −V a⎡⎣ ⎤⎦+ na = ρAhφ + nex

GIP = free gas+ adsorbed gas

fluid density (P,T,z)

available pore space

adsorbed volume

amount adsorbed

• Understand the pore space• Understand excess adsorption

Excess sorption

Pini R 2014 Energy Procedia 63: 5556-61

!"##$%

$% $%

!"##

Gas adsorption in shale and Gas-In-PlaceMaterial balance analysis with sorption effects

&ex = !a − )*a


Gas adsorption in shale

• Most of the porosity in shale is located in pores < 50 nm–What is the accessible pore space? How does it influence fluid behaviour?

–Characterization of microporous solids relies on the adsorption and transport of confined fluids

• Measurement of HP adsorption isotherms is a technical challenge– Inter-laboratory comparisons are difficult (lack of standards)

[e.g., Lancaster 1993 … Gasparik 2014]

– Interactions with shale constituents (e.g., clays and OM) are complex[e.g., Barrer 1954 … Loring 2012; Busch 2014]

–Fluid densification and/or depletion? [e.g., Rother 2012; Schaef 2014; Jeon 2014]

Of practical relevance, but quantification is still a challenge


Sing K. S. W. et al. 1985 Pure Appl. Chem. 57Thommes M. et al. 2015 Pure Appl. Chem. 87:1051-69

Adsorption 101: Sub- vs. super-critical adsorption

Absolute: !aIUPAC


Sing K. S. W. et al. 1985 Pure Appl. Chem. 57Thommes M. et al. 2015 Pure Appl. Chem. 87:1051-69

Bulk Density [mol/L]0 5 10 15 20

Nor

m. E

xces

s Ad

sorp

tion

0

0.2

0.4

0.6

0.8

1

1.2

Silica Gel

13X ZeoliteActivated Carbon

CO2, 50°Cstrictly microporousstrictly mesoporous

20 MPa

Pini R et al. 2006 Adsorption 12:393-403; Pini R et. al 2008 Adsorption 14:133-41;

Pini R 2014 Micropor Mesopor Mat 187:40-52

Absolute: Surface excess:!a !ex = !a − '(aIUPAC

Adsorption 101: Sub- vs. super-critical adsorption


Lack of reproducibility particularly evident on natural materials

Supercritical gas adsorption: a technical challenge

M. Gasparik et al. 2014 Int J Coal Geology 132, 131–146

Supercritical CH4 adsorption on dry Posidonia shale

± 50 %


0 5 10 15 20 25Bulk density, b [mol/L]

-600

-400

-200

0

200

400

600

800

Exce

ss a

mou

nt, n

ex [

mol

/g]

Lack of reproducibility particularly evident on natural materials

Supercritical gas adsorption: a technical challenge

Supercritical CO2adsorption on dry Na-Montmorillonite

Schaef et al. 2014

Busch et al. 2008

Rother et al. 2012

This study

CO2, 45-50∘C

65 MPa

Schaef et al. 2014 Energy Procedia, 63, 7844-7851Rother et al. 2013, Environ Sci Technol, 47, 205-11Busch et al. 2008 Int J Greenhous Gas Control, 2, 297-308


Measuring supercritical gas adsorption

Gravimetric method – Rubotherm Magnetic Suspension balance

Pressure:

vacuum – 35 MPa

Temperature:

0 – 400°C

Gases:

He, CO2, HCs, N2,…

Sample:

< 15 g (± 50 μg)

Qatar Complex Porous Media Lab – May 2017


Excess adsorbed mass:

mex =ma − ρmVa

=M1(ρ,T )−M10 + ρmV

0

Apparent weight:

V 0 =V met +V s

buoyancy

M10

Measuring the adsorption of a dense gasOperating equations

Buoyant solid volume (Helium):

M1(ρ,T ) = ms +mmet +ma − ρm(V

met +V s +V a )

weight under vacuum

El. magnet

Perm. magnet

Ti Sinker

Sample holder

main source of error [1]

[1] Pini R 2014 Micropor Mesopor Mat 187:40-52

protocol includes several (repeated) pressure points


Experimental approach

A combination of engineered and natural microporous solids

Mesoporous carbons Source claysBowland Shale (UK)

Images reproduced from:[1] Fauchille A. L. et al. 2017 Marine and Petroleum Geology, 86, 1374-1390[2] Liang C. et al. 2008 Angew. Chem. Int. Ed., 47, 3696 – 717[3] 'Images of Clay Archive' of the Mineralogical Society of Great Britain & Ireland and The Clay Minerals Society

- Commercially available

- Regular pore structure (10 nm)

- Large porosity

- Source clay mineral (Swy-2)

- Hydrous layer silicates (2:1)

- Micro- and meso-pores (< 50nm)- Large porosity

[1] [2] [3]

- Quartz + Carbonate + Clay + OM

- Complex pore structure

- Moderate porosity

- Microporosity (1-200 nm) (>10%)

- Microfracture porosity (< 5-8 %)


Reduced fluid density, ;/;c0 0.5 1 1.5 2

Exce

ss a

dsor

ptio

n [7

mol

/g]

0

50

100

150

200• Experimental conditions:vac. – 25 MPa, 10 – 80°C

CO2 and CH4 adsorption on Bowland Shale (UK)

10°C40°C

80°C CH4

25 MPa



Exce

ss a

dsor

ptio

n [7

mol

/g]

0

50

100

150



10°C

40°C

80°C

10°C40°C

80°C CH4

CO2

25 MPa



Exce

ss a

dsor

ptio

n [7

mol

/g]

0

50

100

150


• Adsorption Capacity:

• Selectivity:


10°C

40°C

80°C

10°C40°C

80°C CH4HCO2 HCH4 ≈ 5−8

m3STP / t CO2 CH4Bowland 3 – 5 1.5 – 2

Eagle Ford ~ 9*** ~ 4*Barnett** ~ 5 ~ 2

*Jang et al. 2016 Energy Sources,38(16), 2336-2342** Heller and Zoback 2014 Journal of Unconventional Oil and Gas Resources 8, 14–24***Carey JW, Pini R et al. 2017 Caprock Integrity in Geological storage – AGU Monograph.

CO2



GIP [m3STP/m3]

0 5 10 15 20

P/z

[MPa

]

0

10

20

30

40

50

Production of adsorbed gas

kicks in

Gas sorption most effective

volum

etric

reserv

oir

volumetric reservoir

CH4

CO280°C

Gas recovery Gas storage

Potential for enhanced gas recovery?



Exce

ss a

dsor

ptio

n [7

mol

/g]

0

50

100

150

200

Scaling of shale data: the role of organics

[1] Fauchille A. L. et al. 2017 Marine and Petroleum Geology, 86, 1374-1390

MC (10 nm)Shale

40°C

80°C

40°C80°C CH4

• Independent measurements on mesoporous carbon

• Scaling factor (5%) ≈ shale sample TOC (6%wt.[1])

• Observations–Pronounced features of

mesoporosity

CO2



Exce

ss a

dsor

ptio

n [7

mol

/g]

0

50

100

150

200

Bulk Density [mol/L]0 5 10 15 20

Nor

m. E

xces

s Ad

sorp

tion

0

0.2

0.4

0.6

0.8

1

1.2


Scaling of shale data: the role of organics

• Independent measurements on mesoporous carbon

• Scaling factor (5%) ≈ shale sample TOC (6%wt.[1])

• Observations–Pronounced features of

mesoporosity– Some indications of

microporosity–MC analogue for TOC (?)


100

101

102

0 0.2 0.4 0.6 0.8 1

Adsorbedamount,n[cm

3 STP/g]

Relative pressure, p/p0 [-]

MC

Pore space characterisation of shaleLow-pressure (< 1 bar) N2 adsorption isotherms at 77K

Micromeritics 3flex and Tristar

Bowland #6

Clays

0

0.2

0.4

0.6

0.8

1

0 - 2 nm 2 - 20 nm 20 - 50 nm

Normalizedcumulativeporevolum

e,V p[-]

Pore diameter, d [nm]

BowlandEagleFordMontmorilloniteKaoliniteMes.Carbon


100

101

102

0 0.2 0.4 0.6 0.8 1

Adsorbedamount,n[cm

3 STP/g]

Relative pressure, p/p0 [-]

MC

Pore space characterisation of shaleLow-pressure (< 1 bar) N2 adsorption isotherms at 77K

Micromeritics 3flex and Tristar

0

0.2

0.4

0.6

0.8

1

0 - 2 nm 2 - 20 nm 20 - 50 nm


e,V p[-]



Bowland #6


Clays

• Mesopores occupy majority of pore space in shale (> 80%)• Shale does contain microporosity (5 – 10%)• Clays contribute to microporosity (Bowland ~ 7%wt.[1])


Pore space characterization: Bowland shaleLow-pressure (< 1 bar) N2 adsorption isotherms at 77K


• Mesopores occupy majority of pore space in shale (> 80%)• Shale does contain microporosity (5 – 10%)• Bowland case study: apparent correlation between SSAN2 and TOC

0

0.2

0.4

0.6

0.8

1

0 - 2 nm 2 - 20 nm 20 - 50 nm

Normalizedcumulativepore

volum

e,V p[-] Bowland-8

Bowland-5Bowland-2Bowland-6KaoliniteMes.Carbon

0

0.2

0.4

0.6

0.8

1

0 - 2 nm 2 - 20 nm 20 - 50 nm


e,V p[-]



0

2

4

6

8

10

0 1 2 3 4 5 6 7

BETSSA,[m2 /g]

TOC, [wt%]

Bowland-2

Bowland-5

Bowland-6

Bowland-8

[1]


[1] Aranovich and Donohue J Colloid Interface Sci 1998, 200: 273-90 [2] Hocker et al, Langmuir 2003, 19: 1254-67[3] Ottiger et al. Langmuir 2010, 24: 9531-40[4] Pini et al. Adsorption 2010, 16: 37-46[5] Qajar et al. Fuel 2016, 163: 205-13

1 2 30

0.2

0.4

0.6

0.8

1

i

θ i

1 2 3 4 5 6 7 8 9 100

0.2

0.4

0.6

0.8

1

i

θ i

1 10 20 30 40 50 600

0.5

1

i

θ i

1nm 3.3 nm

20 nm

qB=0.1

qB=0.7

qB=0.1

qB=0.7

Density profiles in the poresCO2, T = 45°C

• Lattice DFT model[1]:– Discretization of the pore space

– Slit pores, hexagonal lattice– Nearest-neighbors interactions– Successfully applied to both

engineered materials [2] and rocks [3-5]

Modelling of gas adsorption in shale

21 3… J

j

i

jk(a) (b) (c)

How can we account for the complexity of shale’s pore space?


0.0 0.2 0.4 0.6 0.8 1.00.0

0.5

1.0

1.5

2.0

2.5

Ads

orbe

d am

ount

, n [m

mol

/g]

Relative pressure, p/po

×10-3

0.0 0.5 1.00.0

0.1

0.2

Application example: source clay (SWy-2)Sub- and super-critical data in a unique, consistent framework

LDFT

experiment

N2 adsorption at 77K

1 10 1000.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

11%

11%

9%8%7%

21%

18%

15%

ba

Pore

vol

ume,

∆V

[cc/

g]

Pore width, wp [nm]1 10 100

0.000

0.005

0.010

0.015

0.020

0.025

Pore

vol

ume

[cc/

g]

Pore width [nm]

Pore size distribution


Application example: source clay (SWy-2)

Identification of relevant pore-classes from N2 physisorption

Calibrated model applied to match supercritical isotherms

Solid-fluid interaction used as a fitting parameter (fluid-dependent)

0 5 10 15 200

100

200

300

400

Exce

ss a

mou

nt, nex

[m

ol/g

]

Bulk density, b [mol/L]

25°C50°C80°C115°C

CO2

CH4

Sub- and super-critical data in a unique, consistent framework

Supercritical adsorption

✓CO2

CH4


Application example: source clay (SWy-2)

0 5 10 15 200

100

200

300

400

Exce

ss a

mou

nt, nex

[m

ol/g

]

Bulk density, b [mol/L]

25°C50°C80°C115°C

CO2

CH4

CO2

CH4

Sub- and super-critical data in a unique, consistent framework

Supercritical adsorption

2.5 3.0 3.5

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.4 0.6 0.8 1.00.3

0.5

0.7

CH4 ×10-3

Satu

ratio

n fa

ctor

, csat

1/T [1/K]

CO2

Tc/T

Filling capacity

CO2

CH4


Concluding remarks• Understanding the shale’s pore space requires understanding

gas adsorption (and vice versa)

• Shale and clay-rich systems retains characteristics of a mesoporous material

• The presence of microporosity is confirmed from both sub-and supercritical adsorption studies

• Mesoporous carbon as analogue of OM?

• Significant adsorption selectivity of CO2 vs. CH4

• The complexity of the pore space requires rigorous modeling approaches that incorporate PSD and chemical heterogeneity


• Postdocs/students:

Dr. Lisa Joss, Humera Ansari, Junyoung Hwang

• Collaborators

Prof. Geoffrey Maitland, Imperial College LondonProf. Martin Trusler, Imperial College London

Prof. Alberto Striolo, University College London

• Funding

Acknowledgments

Documents

ShaleXEnvironmentDissemination Event TAMUQ –18 March 2018 · 2018. 3. 28. · Ronny Pini, [email protected] March 2018 High-pressure gas sorption studies on shales Ronny Pini