297
CBM and CO 2 -ECBM related sorption processes in coal Von der Fakultät für Georessourcen und Materialtechnik der Rheinisch -Westfälischen Technischen Hochschule Aachen zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften genehmigte Dissertation vorgelegt von Diplom Physiker Yves Gensterblum aus Würselen Berichter: Univ.-Prof. Dr.rer. nat. Ralf Littke Univ.-Prof. Dr.rer. nat. Guy de Weireld Tag der mündlichen Prüfung: 04. Juni. 2013 Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar.

CBM and CO2-ECBM related sorption processes in coal

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

CBM and CO2-ECBM related sorption processes in coal

Von der Fakultät für Georessourcen und Materialtechnik der

Rheinisch -Westfälischen Technischen Hochschule Aachen

zur Erlangung des akademischen Grades eines

Doktors der Naturwissenschaften

genehmigte Dissertation

vorgelegt von

Diplom Physiker Yves Gensterblum

aus Würselen

Berichter: Univ.-Prof. Dr.rer. nat. Ralf Littke

Univ.-Prof. Dr.rer. nat. Guy de Weireld

Tag der mündlichen Prüfung: 04. Juni. 2013

Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

2 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

3 | P a g e

Publications

D 82 (Diss. RWTH Aachen University, 04th June 2013)

1. CBM and CO2-ECBM related sorption processes in coal: A review

Busch A. and Gensterblum Y.

International Journal of Coal Geology, 87, (2011), 49–71

2. European inter-laboratory comparison of high-pressure CO2 sorption isotherms.

I: activated carbon

Gensterblum Y., P. van Hemert, P. Billemont, A. Busch, D. Charriere, D. Li. B.M. Krooss, G. de

Weireld, D. Prinz, K-H.A.A. Wolf

Carbon, 47, (2009), 2958 –2969

3. European inter-laboratory comparison of high pressure CO2 sorption isotherms.

II: natural coals

Gensterblum Y., P. van Hemert, P. Billemont, E. Battistutta, A. Busch, B.M. Krooss, G. de Weireld,

K-H.A.A. Wolf

International Journal of Coal Geology, 84, (2010), 115–124

4. Molecular concept and experimental evidence of H2O, CH4 & CO2 sorption on organic

material

Gensterblum Y., A. Busch, B.M. Krooss,

Fuel , 114 (2014), 581-588

5. High-pressure CH4 and CO2 sorption isotherms as a function of coal maturity and the

influence of moisture

Gensterblum Y., A. Merkel, A. Busch, B.M. Krooss

International Journal of Coal Geology 118, (2013), 45-57

6. Gas saturation and CO2 enhancement potential of coalbed methane reservoirs as a

function of depth

Gensterblum Y., A. Merkel, A. Busch, B.M. Krooss, R. Littke

AAPG Bulletin (2013) in press

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

4 | P a g e

Submission of the written thesis: 6. January 2013

Day of thesis defence: 04. June 2013

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

5 | P a g e

Dedicated

to my family.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

6 | P a g e

We have not succeeded in answering all our problems. The answers we have found only

serve to raise a whole set of new questions. In some ways we feel we are as confused as

ever, but we believe we are confused on a higher level and about more important things.

(Earl C. Kelley)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

7 | P a g e

Summary

Unconventional gas, such as shale gas, tight gas or coalbed methane is able to offer a new

and attractive perspective for low-carbon fuel supply. Furthermore, they may provide

possibilities for CO2-storage and coevally for enhanced gas recovery. This is suggested by

theoretical considerations of displacing the adsorbed methane by a more strongly

adsorbing gas such as CO2. Optimizing the development of these unconventional resources,

while minimizing their environmental impact, requires a more fundamental understanding

of the coupled physico-chemical processes involved. In order to better understand the

interaction of gas and water with coal of different maturity, a measuring plan has been

developed and executed.

The introduction chapter is a published literature review. The general observation is that

CO2 and CH4 sorption capacity of dry coals shows a parabolic trend with rank with a

minimum between 1.1 and 1.3% vitrinite reflectance. The CH4 and CO2 sorption capacity of

moist coals, in contrast, shows a linear increase with increasing maturity. The sorption

capacity ratio of CO2 and CH4 for dry coals has an approximately constant ratio of 2 over

the entire maturity range. For moist coals this ratio is higher (approximately 6) at low

maturity values and decreases to values around 2 with increasing maturity. The highest

CO2 to CH4 sorption capacity ratios are observed at low surface coverage.

Based on the literature review an extended measuring concept was developed to

investigate the competitive interaction between sorbed water and gases such as CH4 and

CO2.

The determination of the required high-precision CO2 sorption isotherms is challenging

because in the temperature and pressure range of interest (T=308-350K, P = 8 to 12 MPa)

the density of supercritical CO2 increases rapidly with increase pressure. To improve the

reliability of the sorption measurements, an inter-laboratory comparison between different

European research institutes has been conducted. The results and conclusions of this inter-

laboratory comparison are summarised in chapters 2 and 3. During the first inter-

laboratory comparison the comparability between the manometric and the gravimetric

sorption methods was demonstrated. In the second inter-laboratory test the properties of

natural coal sample and their influence on the accuracy of the measurements was

investigated. The observed excess sorption maxima for coals from the Velenje (SLO),

Brzeszcze (PL) and Selar Cornish (GB) mines are 1.77±0.07, 1.37±0.02 and 1.41±0.02 mol

kg-1 in the order of increasing vitrinite reflectance. A decrease of the pressure at maximum

excess sorption has been observed from lignite to bituminous coal to semi-anthracite. This

shift varies from roughly 6.89±0.5 MPa for the Velenje lignite, 6.68±0.4 MPa for the

Brzeszcze bituminous coal, to 5.89±0.6 MPa for Selar Cornish the semi-anthracite. These

could be related to an increase in micropore volume. Furthermore, these chapters contain

recommendations and best practice rules for CO2 sorption isotherm measurements on

coals.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

8 | P a g e

During the subsequent phase of the PhD project the competitive interaction between

sorbed water and gases such as CH4 or CO2 was investigated. The results are summarised in

chapter 4. A conceptual model was developed explaining CO2 and CH4 sorption on coal as a

function of rank and the interaction mechanism of adsorbed water molecules. The

adsorption behaviour of CH4 and CO2 was studied on a sequence of coal samples, covering

the maturity range from 0.5 up to 3.3% vitrinite reflectance. A deeper understanding of the

physical processes of the competitive sorption of CH4, CO2 and water has been achieved.

Based on the thermodynamic description of the experimental observations presented in

this thesis, the conceptual molecular model has been improved, by combining experimental

data and published molecular dynamic simulations. The proposed molecular concept

clarifies the thermodynamic process of CH4 and CO2 sorption on coals, the three

fundamental kerogen types and the effects of adsorbed water.

In Chapter 6 the experimental observations are discussed in detail. The main parameters

controlling ultimate gas recovery from coalbed methane (CBM) deposits are the natural

cleat permeability and the total gas content. Among the three gas storage mechanisms

(compressed gas in the open porosity, dissolved gas in the formation water and sorbed

gas), sorption is the dominant mechanism for CBM reservoirs. Sorption behaviour changes

with coal rank, but also with temperature and pressure and hence with burial history. The

fundamental understanding of the thermodynamics of CH4 sorption capacity as a function

of temperature and pressure is a prerequisite for assessing methane saturation in CBM

reservoirs. This concept is explained in Chapter 6. The results of this study lead to the

conclusion that the geothermal gradient is much more important than the hydrostatic

pressure gradient for the CH4 sorption capacity as a function of depth. For CBM as well as

CO2-ECBM activities a low geothermal gradient is favourable. Furthermore, a low-rank coal

deposit requires a lower pressure draw-down than high rank coal deposits to desorb an

equivalent amount of methane. The study of the thermodynamics of adsorption suggests

that it is more likely to observe under-saturated CBM reservoirs in high rank than in low

rank coal deposits.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

9 | P a g e

Zusammenfassung

Der weltweite Energieverbrauch steigt kontinuierlich. Erdgas stellt wegen seiner hohen

Energiedichte und des vergleichsweise geringen Kohlenstoffanteils eine ausgezeichnete

Alternative zu Erdöl und Kohle dar. Wegen des Rückgangs der Förderung aus

konventionellen Gaslagerstätten, ist die Erkundung und Nutzung unkonventioneller

Erdgaslagerstätten in den letzten Dekaden zunehmend in den Fokus gerückt. Die Injektion

von CO2 in unkonventionelle Lagerstätten wird aufgrund theoretischer Betrachtungen als

Möglichkeit zur Steigerung der Methangasausbeute gesehen. Zugleich könnte das

verwendete CO2 auf diese Weise durch Adsorption dauerhaft gespeichert werden. Die

Identifizierung und Erforschung der molekularen Prozesse bei der Sorption an

organischem Material verschiedener thermischer Reifestadien und der Einfluss von Wasser

auf die Sorption, am Beispiel der Flözgasförderung sind Gegenstand dieser Doktorarbeit.

Einleitend würde eine umfassende Literaturstudie zur Sorption von CO2 und CH4 an Kohlen

und dem Einfluss von sorbiertem Wasser durchgeführt und veröffentlicht.

Die CO2 und CH4 Sorptionskapazität von trockenen Kohlen hat einen parabolischen Verlauf

mit zunehmender Reife mit einem Minimum bei ca. 1% bis 1.2% Vitrinitreflektion. Bei

feuchten Kohlen beobachtet man eine lineare Abhängigkeit zwischen CH4 und CO2

Sorptionskapazität und der Reife der Kohle. Die detaillierten Trends werden im ersten

einleitenden Kapitel vorgestellt.

Auf der Basis dieser Literaturstudie wurde ein Messplan entwickelt, mit der Zielsetzung,

die CH4 und CO2 Sorptionsprozesse, den Einfluss von sorbiertem Wasser und die

dazugehörigen thermodynamischen Grundgrößen wie Enthalpie und Entropie zu

bestimmen. Aus diesen Messdaten und publizierten molekulardynamischen Simulationen

konnte ein molekulares Konzept entwickelt werden, das die experimentellen

Beobachtungen zu den Sorptionsprozessen sowie den Einfluss von sorbiertem Wasser auf

diese Prozesse erklärt.

Zur Bearbeitung dieser wissenschaftlichen Fragestellungen wurden CH4 und CO2

Sorptionsisothermen an verschiedenen thermisch-reifen Kohlen bestimmt. Die Nähe der

Messbedingungen zu den kritische Kenngrößen (Druck und Temperatur) von CO2 stellt

eine mess- und regelungstechnische Herausforderung dar. Um eine hohe Qualität und

reproduzierbarkeit der Sorptionsisothermen sicherzustellen wurde eine zwei-phasige

internationale Vergleichsstudien (Ringversuche) unter Beteiligung verschiedener

internationaler Arbeitsgruppen durchgeführt. In der ersten Phase konnten grundlegende

methodische Probleme geklärt und ausgeräumt werden. In der zweiten Phase standen die

Eigenschaften und experimentelle Schwierigkeiten von natürlichen Kohleproben im

Vordergrund. In mehreren Workshops wurden Empfehlungen und methodische

Optimierungsmöglichkeiten für die Bestimmung der CO2 Sorptionskapazität von Kohlen

erarbeitet. Die Ergebnisse der Studien sind in Kapitel 2 und 3 erläutert.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

10 | P a g e

Mit dieser verbesserten Methode wurde der Messplan, welcher auf der Basis der

einführenden Literaturstudie aus dem ersten Kapitel erarbeitet wurde, durchgeführt. Die

Ergebnisse und Schlussfolgerungen werden in Kapitel 4 erläutert. Zusammenfassend läßt

sich Schlussfolgern, die sauerstoffenthaltenden Funktionalen-Gruppen sind für die

sorbtiven Eigenschaften von unreifen Kohlen verantwortlich. Verschwinden diese durch

zunehmende Inkohlung mit zunehmender Kohlereife, verschwinden auch der Effekt das

durch bereits sorbiertes Wasser die CH4 und CO2 Sorptionskapazität reduziert wird.

Im fünften Kapitel werden die experimentellen Ergebnisse detailliert vorgestellt und

diskutiert.

In Kapitel 6 wird ein Konzept für die Anwendung der experimentellen Daten und ihren

thermodynamischen Grundgrößen im Rahmen der geologischen Entwicklung einer

unkonventionellen Gaslagerstätte gezeigt. Der Fokus hierbei, liegt auf der

thermodynamischen Abschätzung von Methangassättigung, Förderpotential sowie dem

CO2 Speicher- und Stimulationspotential. Aus diesen, ausschließlich auf

thermodynamischen Überlegungen basierenden Ergebnissen, lässt sich ableiten, dass

Kohleflöze in Sedimentbeckensysteme mit niedrigen thermischen Gradienten generell

besser für die Förderung von Methan geeignet sind. Im relevanten Temperatur- und

Druckbereiche ist der dominante Speichermechanismus die Sorption von Methan. In der

offenen Porosität komprimiertes sowie gelöstes Gas stellen dagegen nur geringe bzw.

vernachlässigbare Speicherkapazitäten für Methan zur Verfügung. Drittens, um die gleiche

Menge Methan zu fördern benötigt man bei Lagerstätten mit unreiferen Kohlen eine

geringere Absenkung des Porendrucks als bei höher reifen Kohlen. Abschließend lässt sich

aus den thermodynamischen Betrachtungen folgern, dass eine erhebliche Untersättigung

eines Kohleflözes mit Methangase allein durch die beckengeschichtliche Hebung in den

meisten Fällen nicht zu erwarten ist.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

11 | P a g e

Content

Publications ............................................................................................................................................ 3

Summary .................................................................................................................................................. 7

Zusammenfassung ................................................................................................................................ 9

Content .................................................................................................................................................... 11

1. Introduction: Motivation, concept and workflow ............................................................ 16

Why study sorption properties of coals? ........................................................................................ 16 Introduction: measuring concept and workflow ......................................................................... 17

Thematic Introduction: CBM and CO2-ECBM related sorption processes in coal ......... 20

Abstract ........................................................................................................................................................ 21 Introduction ............................................................................................................................................... 22 Adsorption methods................................................................................................................................ 25

Uncertainties in recording sorption isotherms ............................................................................ 31

Modeling of adsorption isotherms ..................................................................................................... 37

Excess versus absolute sorption ......................................................................................................... 37

Water sorption on coal ........................................................................................................................... 40 Coal-specific factors controlling water sorption .......................................................................... 42

CH4 and CO2 sorption of coals in the presence of water ............................................................ 47

Dependence of CO2, CH4 and water sorption on coal properties ........................................... 54 The influence of coal composition on gas sorption ..................................................................... 54

Coal Rank ..................................................................................................................................................... 56

CO2 and CH4 sorption kinetics on coal ............................................................................................. 62 Unipore sorption/diffusion models .................................................................................................. 67

Bidisperse sorption/diffusion models ............................................................................................. 68

Conclusions ................................................................................................................................................. 75 Acknowledgements ................................................................................................................................. 76 References ................................................................................................................................................... 77

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

12 | P a g e

2. European inter-laboratory comparison of high-pressure CO2 sorption isotherms.

I: activated carbon .............................................................................................................................. 93

Abstract ........................................................................................................................................................ 95 Introduction ............................................................................................................................................... 96

Previous inter-laboratory comparisons of carbon dioxide sorption on coal at high

pressures ..................................................................................................................................................... 97

Methods and materials ........................................................................................................................... 97 Sample and sample prepartion ........................................................................................................... 97

Experimental methods ......................................................................................................................... 101

Results ....................................................................................................................................................... 106 Parameterization of the experimental results ............................................................................ 109

Comparison with previously published results .......................................................................... 111

Discussion ................................................................................................................................................ 112 Effect of moisture content and variability of batch composition ........................................ 113

Using Helium as reference gas .......................................................................................................... 114

Conclusion ................................................................................................................................................ 115 Acknowledgements .............................................................................................................................. 116 Notation .................................................................................................................................................... 117 References ................................................................................................................................................ 118

3. European inter-laboratory compari-son of high pressure CO2 sorption isotherms.

II: natural coals ................................................................................................................................. 122

Abstract ..................................................................................................................................................... 123 Introduction ............................................................................................................................................ 124 Previous inter-laboratory comparisons of carbon dioxide sorption on coal at high

pressures............................................................................................................................................... 125 Methods and materials ........................................................................................................................ 126

Samples and sample preparation ..................................................................................................... 126

Experimental methods ......................................................................................................................... 128

Experimental Results ........................................................................................................................... 131 Discussion ................................................................................................................................................ 143

Comparison to previous inter-laboratory comparisons ......................................................... 145

Minimizing experimental errors for manometric methods ................................................... 150

Minimizing experimental errors for gravimetric methods .................................................... 150

Conclusion ................................................................................................................................................ 151 Acknowledgements ............................................................................................................................... 151

Notation ...................................................................................................................................................... 153

References ................................................................................................................................................ 154

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

13 | P a g e

4. Molecular concept of H2O, CH4 & CO2 adsorption on organic material ................. 159

Abstract ..................................................................................................................................................... 160 Article ......................................................................................................................................................... 161 Thermodynamics ................................................................................................................................... 161

Entropy ....................................................................................................................................................... 162

Results and Discussion ........................................................................................................................ 163 Conclusion ................................................................................................................................................ 176 Acknowledgments ................................................................................................................................. 177 References ................................................................................................................................................ 178

5. High-pressure CH4 and CO2 sorption isotherms as a function of coal maturity and

the influence of moisture .............................................................................................................. 182

Abstract ..................................................................................................................................................... 183 Introduction ............................................................................................................................................ 184

Competitive water/gas sorption ...................................................................................................... 184

Excess and absolute adsorption and adsorbed phase density.............................................. 185

Samples and sample preparation .................................................................................................... 187 Experimental procedure ..................................................................................................................... 189

Experimental conditions ..................................................................................................................... 189

Evaluation procedure ........................................................................................................................... 190

Data fitting procedure .......................................................................................................................... 190

Results and Discussion ........................................................................................................................ 191 Isotherm characteristics ..................................................................................................................... 192

Adsorbed phase density ....................................................................................................................... 192

Evolution of sorption properties with coal rank ....................................................................... 201 Evolution of sorption capacity with surface coverage ............................................................ 205

Influence of moisture on the sorption capacity .......................................................................... 205

CO2/CH4 sorption capacity ratio ....................................................................................................... 207

Influence of coal rank ............................................................................................................................ 207

Influence of moisture on the CO2/CH4 sorption capacity ratio ............................................ 208

Temperature ........................................................................................................................................... 210 Conclusion ................................................................................................................................................ 210 Appendix ................................................................................................................................................... 213 References ................................................................................................................................................ 215

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

14 | P a g e

6. Gas saturation and CO2 enhancement potential of coalbed methane reservoirs as

a function of depth ........................................................................................................................... 220

Abstract ..................................................................................................................................................... 221 Introduction ............................................................................................................................................ 222 Samples and sample preparation .................................................................................................... 230 Experimental and conceptual procedure ..................................................................................... 232

Parameterisation of experimental sorption data ....................................................................... 232

Temperature-dependence of gas sorption isotherms .............................................................. 234

In-situ gas sorption capacity of coal seams .................................................................................. 234

Gas recovery potential of coal seams .............................................................................................. 236

Results and geological implications ............................................................................................... 238 Discussion ................................................................................................................................................ 246

Influence of water on gas sorption capacity ................................................................................ 249

Conclusions .............................................................................................................................................. 250 Appendix ................................................................................................................................................... 252 References ................................................................................................................................................ 254

Main conclusions and discussion ............................................................................................... 271

Outlook ................................................................................................................................................ 276

Content: Figure and Table list ...................................................................................................... 278

Figures ....................................................................................................................................................... 278 Tables ......................................................................................................................................................... 285

Acknowledgements ......................................................................................................................... 289

Curriculum vitae (education) ...................................................................................................... 290

Publication list .................................................................................................................................. 293

Conference talks ............................................................................................................................... 297

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

15 | P a g e

True wisdom comes to each of us when we realize how little we understand about life,

ourselves, and the world around us.

(Sokrates)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

16 | P a g e

1. Introduction: Motivation, concept and

workflow

Why studying sorption properties of coals?

World gas consumption has steadily increased over the past decades leading to a depletion

of conventional gas reserves. The end of nuclear power generation in Germany in 2022 will

bring about an increase in gas consumption especially if CO2 reduction will be part of the

European energy strategy. Due to the fact that natural gas has the smallest carbon footprint

of the conventional energy sources (like oil, coal, etc.), its importance will increase. In the

future gas supply can only be maintained if additional gas resources are utilized. In this

context, gas production from unconventional reservoirs has substantially increased in

recent years, especially in the USA and Australia. For instance, the national economy of the

USA become probably in 2020 regardless of natural gas imports. In addition, the USA

shows bigger progress by the reduction CO2 than Europe already, by the substitute of coal,

with natural gas from unconventional reservoir for the primary energy production. In

Europe exploration of these unconventional reservoirs, including tight gas, coalbed

methane and gas-shales, has greatly expanded over the last decade. Unconventional gas,

such as coalbed methane offers an attractive low-carbon solution and furthermore

provides possibilities for CO2-storage and coevally for enhanced gas recovery. This is

justified in the theoretical considerations of displacing the adsorbed methane by a stronger

adsorbing gas such as CO2. Optimizing the development of these unconventional resources,

while minimizing their environmental impact, requires a more fundamental

thermodynamic understanding of the sorption processes involved. In order to better

understand gas and water interaction with coal of different maturity a measuring plan has

been developed and executed.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

17 | P a g e

Introduction: measuring concept and workflow

Firstly, the measuring method to determine CO2 sorption isotherms had to be improved.

The accurate prediction of equilibrium sorption capacities is important to estimate gas

saturation and recovery potential in basin modeling and/or production simulators. To

simulate reservoir conditions, sorption isotherm are generally collected at elevated

temperatures, usually between 298 and 350K, and elevated pressures (up to 20 MPa).

Although reflecting reservoir conditions, still insufficient data on the physical fundamentals

of gas adsorption of supercritical gases on coals or other geological samples in the presence

of water has been published. In addition, the influence of water on the sorption properties

has been investigated. To achieve this goal a differential approach has been used. The

isotherm difference between dried and moisturised coal samples taken from a maturity

range between 0.5% and 3.3% has been measured and evaluated.

Furthermore, we propose a fundamental molecular concept, which is able to explain the

experimental observation such as the influence of water on the CO2 and CH4 sorption

capacity as a function of maturity. This paper aims at shading light on these aspects based

on the datasets determined here.

In total, at least 45 isotherms have been measured. Isotherms of different temperatures are

required for thermodynamic treatment of the sorption processes with and without pre-

adsorbed water.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

18 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

19 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

20 | P a g e

Thematic Introduction: CBM and CO2-

ECBM related sorption processes in coal

International Journal of Coal Geology 87 (2011) 49–71

Andreas Busch,

Shell Global Solutions International, Kessler Park 1, 2288GS Rijswijk, The Netherlands

Yves Gensterblum,

RWTH Aachen University, Institute of Geology and Geochemistry of Petroleum and Coal,

Lochnerstr. 4-20, D-52056 Aachen, Germany

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

21 | P a g e

Abstract

This article reviews the state of research on sorption of gases (CO2, CH4) and water on coal

for primary recovery of coalbed methane (CBM), secondary recovery by an enhancement

with carbon dioxide injection (CO2-ECBM), and for permanent storage of CO2 in coal seams.

In the last decade a large amount of data has been published characterizing coals from

various coal basins world-wide for their gas sorption capacity. This research was either

related to commercial CBM production or to the usage of coal seams as a permanent sink

for anthropogenic CO2 emissions. Presently, producing methane from coal beds is an

attractive option and operations are under way or planned in many coal basins around the

globe. Gas-in-place determinations using canister desorption tests, CH4 isotherms are

performed routinely and have provided large datasets for correlating gas transport, and

sorption properties with coal characteristic parameters.

Publicly funded research projects have produced large datasets on the interaction of CO2

with coals. The determination of sorption isotherms, sorption capacities and rates has

meanwhile become a standard approach.

In this study we discuss and compare the manometric, volumetric and gravimetric methods

for recording sorption isotherms and provide an uncertainty analysis. Using published

datasets and theoretical considerations, water sorption is discussed in detail as an

important mechanisms controlling gas sorption on coal. Most sorption isotherms are still

recorded for dry coals, which do not represent in-seam conditions, and water present in

the coal has a significant control on CBM gas contents and CO2 storage potential. This

section is followed by considerations of the interdependence of sorption capacity and coal

properties like coal rank, maceral composition or ash content. For assessment of the most

suitable coal rank for CO2 storage data on the CO2/CH4 sorption ratio data have been

collected and compared with coal rank.

Finally, we discuss sorption rates and gas diffusion in the coal matrix as well as the

different unipore or bidisperse models used for describing these processes.

This review does not include information on low-pressure sorption measurements (BET

approach) to characterize pore sizes or pore volume since this would be a review of its

own. We also do not consider sorption of gas mixtures since the data base is still limited

and measuring methods are associated with large uncertainties.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

22 | P a g e

Introduction

The technical development of coalbed methane (CBM) and secondary CO2 enhanced CBM

(CO2-ECBM) production and CO2 storage in and from coal seams requires detailed and

reliable information on fluid transport and gas sorption as well as their inter-dependent

interaction. An improved understanding of these processes from the macroscopic to the

microscopic scale is important for the accurate prediction of gas and water production

rates as well as CO2 injection rates. The mechanisms of storage and transport of gas and

water in coal differ significantly from conventional gas reservoirs. Commonly, gas transport

in coal is considered to occur at two scales (Figure 1): (I) laminar flow through the cleat

system, and (II) diffusion and sorption in the coal matrix. Flow through the cleat system is

pressure-driven and may be described using Darcy’s law, whereas flow through the matrix

is assumed to be concentration-driven and is modeled using Fick’s law of diffusion. Gas

storage by physical sorption occurs mainly in the coal matrix (e.g. (Harpalani, 1997)).

Various articles address laboratory or field research performed in the context of primary

coalbed methane (CBM) or secondary CO2 enhanced CBM (CO2-ECBM) recovery and

reviews are provided by e.g. (Moore, 2012; White et al., 2005).

There are many factors that need to be addressed when attempting to understand the

complexity of the interaction of CH4, CO2 and water with coal in the cleat system and the

coal matrix, and research has diversified substantially, especially since CO2 storage in coal

is considered as a potential way of permanently immobilizing CO2 in the sub-surface.

Coalbed methane recovery projects are currently developed commercially all over the

world with a main focus on countries like Australia, China and the United States, and

exploration is ongoing in many further regions in e.g. Europe, Ukraine, or Indonesia. One of

the first CO2-ECBM micro-pilot field tests was set up in Alberta, Canada (Gunter, 2004)

which was similarly performed in China some years later (Wong et al., 2007). One

operation with a two-well test setting was carried out in Japan (Ohga, 2006; Yamaguchi,

2004). The only large-scale ECBM field pilot using CO2 injection was operated in the San

Juan Basin in New Mexico, United States (Erickson and Jensen, 2001; Reeves et al., 2003).

Finally a two-well demonstration test was performed in the Upper Silesian Basin, Poland

within the RECOPOL1 and MOVECBM2 projects funded by the European Union (van Bergen,

2006).

To predict a CBM production profiles either during primary or secondary production,

aspects like coal permeability and porosity, density, ash and moisture content, initial gas-

in-place (GIP) (from canister desorption tests), gas sorption capacity from laboratory

isotherms (to obtain gas saturations and desorption pressure), gas diffusivities, coal

volumetrics (thickness and areal extent) or thermodynamic properties (e.g. viscosity,

1 http://recopol.nitg.tno.nl/ 2 www.movecbm.eu

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

23 | P a g e

equation of state) of the fluids involved need to be understood as a minimum requirement.

When dealing with CO2-ECBM selective adsorption, counter diffusion in the coal matrix, or

coal shrinkage and swelling (from CH4 desorption and CO2 adsorption, respectively) need

to be investigated in addition to the parameters above.

During CO2-ECBM processes, the areal distribution of the CO2 injected is accomplished by

flow through the cleat network. When CO2 is entering the coal matrix by a combined

sorption/diffusion process it will sorb to the coal inner surface and at the same time

replace part of the CH4. This replacement occurs either by a reduction in the CH4 partial

pressure or by a higher selective sorption of CO2 over CH4. Because of a concentration

gradient between CH4 in the matrix compared to the cleat system, CH4 diffuses from the

coal matrix into the cleat system where, by pressure drawdown towards a production well,

it can be produced (Figure 1).

Figure 1: Illustration showing coal matrix blocks and cleat system of a coal.

In this context this review summarizes gas (CO2, CH4) and water sorption on coal and

specifically addresses the following topics:

The most common experimental setups for determining gas and water sorption on coal (or

other microporous materials), its differences and experimental uncertainties. Further the

modeling of sorption isotherms and the problems of calculating absolute from excess

sorption is addressed.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

24 | P a g e

CO2 and CH4 sorption on natural coals and its dependence on coal specific parameters like

coal rank, maceral composition or ash content.

Water sorption on coal, its dependence on coal properties like rank and coal chemistry and

gas sorption in the presence of water.

Findings related to sorption kinetic data obtained from the pressure decline versus time

plots of individual sorption steps. The different model assumptions are introduced,

sorption rates are compared for different moisture contents, coal rank and differences

between CO2 and CH4 sorption rates are defined.

Concluding remarks and recommendations for future research to better understand

gas/water/coal interaction.

In this review we do not cover the following issues due to insufficient data published and

because we aim at providing a focused and consistent paper addressing the issues above:

Sorption of any other gases than CO2 and CH4 since (i) we want this paper to specifically

address sorption related to CO2-ECBM or CBM recovery and (ii) only limited datasets are

available on the sorption of e.g. N2 or flue gas mixtures with additional co-contaminants like

e.g. H2, CO, O2, or SO2.

Gas sorption from gas mixtures. Although there are studies available dealing with binary

CO2/CH4 and CO2/N2 or ternary CO2/CH4/N2 gas mixtures, data is limited and the

experimental techniques still rather pre-mature. Some work has been published however

(e.g. (Busch et al., 2003b; Busch et al., 2006; Busch, 2007; Fitzgerald et al., 2005; Pini et al.,

2010)).

A detailed review of the various models available for fitting sorption isotherms. A summary

on this can be retrieved from textbooks dealing with gas sorption or from dedicated

journals dealing with microporous sorbents in a more general manner.

Although investigated in detail lately we also do not cover coal swelling in this study.

Authors are aware that coal swelling is an important mechanism affecting gas sorption on

coal and the interpretation of sorption isotherms. There are several studies dealing with

experimental (e.g. (Battistutta et al., 2010; Day et al., 2008b; Karacan, 2003; Karacan, 2007;

Kelemen, 2006; Mazumder, 2005; Mazumder and Wolf, 2008; Reucroft, 1986; van Bergen

et al., 2009)) and theoretical (e.g. (Palmer and Mansoori, 1998; Pan and Connell, 2007;

Vandamme et al., 2010)) aspects of this phenomenon.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

25 | P a g e

Adsorption methods

Sorption processes are subdivided, according to the energies of interaction involved, into

chemical (chemisorption) and physical sorption (physisorption). In chemisorption, the

sorbate is bound to specific sorption sites on the solid surface by electron transfer or

electron sharing which makes this process highly specific compared to physisorption which

is nonspecific. Chemisorption typically involves monolayer while physisorption can involve

either monolayer or multilayer surface coverage (Karge et al., 2008).

Physisorption occurs via van der Waals forces, i.e. attraction due to permanent and/or

induced dipoles between the sorbate molecules and the atoms composing the sorbent

surface. The enthalpy of physisorption is not much larger than the enthalpy of

condensation of the sorbate and considered to be 1.0 to 1.5 times the latent heat of

evaporation (compared to chemisorption where the enthalpy is >1.5 times the heat of

evaporation) (Karge et al., 2008).

Physisorption, in contrast to chemisorption, is distinguished by only minor changes in the

sorbate and sorbent. The sorbent remains unaltered until relaxation of the substrate lattice

occurs. Further the bondings in the sorbate are changed, which becomes apparent through

a modified oscillation frequency (measurable by FTIR- or Raman spectroscopy). Physical

adsorption is a reversible process, because there is no covalent bond between the sorbate

molecules and the solid surface. It is most likely that adsorption occurs as a monolayer at

low pressures and as multilayers at higher pressures, depending on the type of sorbent and

the sorbate investigated.

The investigation of sorption and desorption of microporous materials (in this case natural

coal) can be subdivided into two main areas: (a) measurements at low pressures that are

typically in the range of the vapour pressure of the sorptive gas and (b) measurements at

high pressures where the gas species is usually above its supercritical pressure. Low

pressure sorption measurements are used for determining structural parameters, like e.g.

pore size distribution or specific surface areas and micro- or mesopore volumes or for gas

purification. In high-pressure experiments the sorption capacity of gases on natural coals is

determined. Low-pressure sorption is not covered in this study; this topic would be subject

of an overview article of its own.

Methods to determine gas sorption

For the determination of high-pressure gas sorption isotherms of coals two different

techniques are commonly used. The techniques differ in terms of the physical parameters

used to determine the isotherms: (1) The manometric/volumetric method requires very

accurate determination of cell and void volumes. Here the amount of gas sorbed is

recorded by pressure readings (manometric method) or pressure and volume readings

(volumetric method). (2) In the gravimetric method, the amount of gas sorbed is measured

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

26 | P a g e

at constant pressure by means of a very accurate balance, with the sample either

suspended mechanically or by magnetic coupling across the wall of a high-pressure vessel.

Manometric Method

The manometric method is most widely used for determining gas sorption capacities on

coal (e.g. (Busch et al., 2003b; Busch et al., 2004; Busch, 2007; Bustin and Clarkson, 1998;

Chaback, 1996; DeGance, 1993; Harpalani et al., 2006; Joubert et al., 1973a; Krooss et al.,

2002; Laxminarayana and Crosdale, 1999b; Li et al., 2010; Mastalerz et al., 2004; Mavor,

1990; Nodzenski, 1998; Prinz, 2001; Siemons and Busch, 2007). The setups are either

custom made or designed in-house and consist typically of calibrated reference volume and

sample cells. Pressure and temperature transducers are either connected to the sample cell

only or to both, reference and sample cell. To analyze sorption from gas mixtures

(combination of usually N2, CH4 and CO2) the gas composition has to be determined

additionally by using a gas chromatograph (GC) equipped, for instance, with a Thermal

Conductivity Detector (TCD). Either sample or reference cell or both are connected to the

GC through a sampling valve. A basic schematic of this setup is provided in Figure 2.

Figure 2. Schematic setup for manometric sorption devices. V denotes valves and P denotes

pressure transducers.

In the manometric procedure, defined amounts of gas are successively transferred from a

calibrated reference volume into the sample cell containing the coal sample. Prior to the

sorption experiment the void volume (0

voidV ) of the sample cell is determined by expansion

of a “non-sorbing” gas, which is typically helium. Helium densities are calculated using the

equation of state (EOS) by (McCarty and Arp, 1990) or using the van der Waals equation

with the a and b parameters reported by (Michels and Wouters, 1941). This procedure also

provides the skeletal volume ( 0sampleV ) and the skeletal density ( 0

sample ) of the sample.

For gas sorption isotherms, the void volume multiplied by the density of the gas (or

supercritical) phase ( ),(20 pTVCO

void ), yields the “non-sorption” reference mass, i.e. the

amount of gas (supercritical fluid) that would be accommodated in the measuring cell if no

sorption takes place. Densities are calculated using the corresponding EOS for CO2, CH4, N2

or their binary and ternary mixtures (Kunz et al., 2007; Peng, 1976; Setzmann and Wagner,

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

27 | P a g e

1991; Span and Wagner, 1996). The excess sorption mass ( 2COexcessm ) is the difference

between the mass of gas that has been transferred into the measuring cell up to a given

pressure step and the “non-sorption” reference mass:

),(0 pTVmm gasvoid

gasdtransferre

gasexcess (1)

The mass transferred from the reference cell into the measuring cell during N successive

pressure steps is given by:

N

i

i

gas

ii

gas

iref

gas

dtransferre TpTpVm1

11 ),(),( (2)

Rather than in mass units, the excess amount of sorbed gas is usually expressed in amounts

of substance (mol) or volume of sorbed gas under standard temperature and pressure

(STP) conditions (m3; 288 K, 0.1 MPa). The excess sorption is usually normalized to the

initial mass of the sorbent.

Volumetric Method

The volumetric method (Fitzgerald et al., 2005; Gasem, 2002b; Hall, 1994; Ozdemir, 2004;

Ozdemir, 2003; Reeves, 2005; Sudibandriyo, 2002; Sudibandriyo et al., 2003).

A simplified scheme of the experimental apparatus is shown in Figure 3. The

pump/reference and sample cell sections of the apparatus are maintained in two constant

temperature air baths. Similar to the manometric method, the sample is placed in the

sample cell, the volume of which is determined using helium as non-sorbing gas. The

isotherm is determined by continuously decreasing the volume in the piston pump and

thus increasing the gas pressure. The amount of gas injected is determined by an accurate

reading of the volume changes in the pump. By adjusting the piston pump volume, sorption

quantities can be determined at defined pressures while for the manometric method (with

fixed reference and void volumes) the equilibrium pressure depends on the reference to

sample cell volume ratio and the pressure charge in the reference cell. The amount of gas

injected

can be determined from the pump position as it moves forward.

Therefore:

(3)

while

(4)

and

(5)

Here m denotes the mass of gas, p is pressure, T is temperature, M is the molar mass of the

gas species, Z is the compressibility coefficient of the pure gas species and R is the universal

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

28 | P a g e

gas constant. V is the volume change in the pump and Vvoid is the volume of the free gas in

the sample cell.

Figure 3: Schematic diagram of volumetric gas sorption device. Modified after (Sudibandriyo

et al., 2003).

Gravimetric Method

Most commercial gravimetric sorption devices used for determining gas sorption capacities

are commercial (Bae and Bhatia, 2006; Charrière et al., 2010; De Weireld et al., 1999;

Ottiger et al., 2006; Pini et al., 2010; Pini et al., 2006; Pini, 2006), either used as delivered or

with in-house modifications. In this instrument the sorbent is placed into the high-pressure

compartment of a magnetic suspension balance and exposed to the sorptive at constant

temperature (Figure 4). The excess sorption is determined from the weight change

(apparent mass change) of the sample ( ) recorded during

individual sorption steps, where is the original sample mass. The excess sorption is

derived from its apparent weight change by a buoyancy correction based on the skeletal

volume ) of the sample, corresponding to the same reference state as in the

manometric procedure. The determination of the skeletal density or volume is performed

with helium. The excess sorbed mass is then given by:

(6)

(7)

As in the manometric procedure, the volume and density of the sorbed phase are not

known and therefore their effect on the buoyancy term cannot be explicitly taken into

account. The excess sorption values are normalized to the original sample mass

The magnetic suspension balance is capable of measuring weight changes with an accuracy

of ±2 g. The system consists of an electromagnet linked to the balance and a permanent

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

29 | P a g e

magnet at the top of the suspension system for the crucible containing the sorbent. More

details are given in e.g. (De Weireld et al., 1999).

Figure 4: Suspension magnetic balance. EM=electro magnet; PM=permanent magnet;

TS=titanium sinker. Adapted from (Charrière et al., 2010).

Several studies use an in-house built gravimetric device (e.g. (Day et al., 2008c; Day et al.,

2005; Sakurovs et al., 2009; Sakurovs et al., 2008), Figure 5). This set-up measures the

weight change of the sample cell upon gas sorption/desorption. The free gas volume in the

sample cell is determined by He-pycnometry as in the manometric method. In addition an

empty reference cell is used to determine the mass of a defined volume of gas at the same

pressure as in the sample cell. Thus, the gas density at each reference pressure can be

determined without having to rely on an EOS. Especially for gas mixtures where EOS are

unavailable or not sufficiently accurate this method is practical.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

30 | P a g e

Figure 5. Gravimetric sorption device used at CSIRO Energy Technlology, Newcastle, AUS. RC

and SC are reference and sample cells respectively. Adapted from (Day et al., 2005).

Alternative methods for recording gas sorption on coal

Two additional test methods for sorption on coal have been reported recently. There are

many more methods used but not applied to coal/gas systems to the knowledge of the

authors:

Gas sorption (CO2, CH4) under controlled confining stress (6.9-13.8 MPa) on a cylindrical

coal sample plug (Pone et al., 2009) using a volumetric method. Results were compared to

powdered samples under unconfined conditions and showed a reduction in sorption

capacity by up to 91% for methane and up to 69% for CO2.

(Hol et al., 2011) recently reported a new, direct method of determining the uptake of CO2

by coal (i.e. by adsorption plus pore-filling). A cylindrical coal sample, 4 mm in diameter

and 4 mm in length, is jacketed in a tightly fitting, annealed gold capsule and exposed to

CO2 at constant pressure and temperature. Adsorption-induced swelling of the sample is

accommodated by ductile deformation of the capsule. Once the coal is saturated, the

capsule is sealed by mechanical loading and the external CO2 pressure removed. This

allows the CO2 to desorb from the coal and flow into an inflatable Al-foil bag attached to the

capsule. The volume of the bag, and hence the amount of CO2 stored in the coal sample, is

determined using the Archimedes method. While the capsule method is time-consuming

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

31 | P a g e

and not easily executed, the advantage is that it is independent of any Equation of State

(EOS), and that no volumetric effects or impurities distort the shape of the isotherm as with

the methods below. The method reported by (Hol et al., 2010) is currently being tested and

compared to the manometric method in collaboration with the present authors.

Uncertainties in recording sorption isotherms Potential sources of experimental errors have been discussed in numerous papers, both

within the coal research community and in relation to other microporous materials. Here

we only refer to those studies dealing with natural coals (e.g. (Gensterblum et al., 2010a;

Mavor, 2004; Mohammad et al., 2009; Pini et al., 2006; Sakurovs et al., 2009; van Hemert et

al., 2009; van Hemert et al., 2007; Yu et al., 2008a). Issues related to temperature control,

determination of sample cell volume, equations of state, gas impurities, water vapor

pressure and gas uptake in water (by dissolution) and mineral matter surfaces are

addressed her in some detail.

Helium as reference gas

Both, the gravimetric and the manometric/volumetric methods use helium (sometimes

argon) as reference gas to determine the buoyancy of the sample (gravimetric) or the void

volume of the sample cell (manometric/volumetric). By definition and experimental

practice, they represent differential methods with respect to helium.

Helium is commonly considered as ‘‘non-sorbing’’. A small degree of helium sorption

(which cannot be excluded but also not quantified) will lead to an underestimation of the

sample volume for both methods. (Gumma and Talu, 2003) proposed a method to correct

sorption data for helium sorption. Helium sorption in turn, would result in an error

(underestimation) of the excess sorption capacity. As noted by (Sakurovs et al., 2009),

helium adsorption, if present, will be in the mol g-1 range as compared to CH4 or CO2

sorption capacity on coal which is usually in the mmol g-1 range and hence can be

neglected.

Another potential problem in using helium as a reference gas is related to the accessibility

of the pore space. It is commonly assumed that the same pore volume is accessed by helium

as to CO2 or CH4. An overview on the state of discussion is provided by e.g. (Sakurovs et al.,

2009; Siemons and Busch, 2007; Walker Jr., 1988).

Temperature

Inaccuracies in temperature measurements and their effects on sorption measurements

have been reported and discussed frequently in the literature. Temperature errors are

commonly between 0.1 and 0.3 K for experimental temperatures well above room

temperature. This is because it is typically easier to heat a system than to cool it and to

keep it at a constant temperature (in time and space). Therefore, experimental

temperatures below room temperature are expected to have a larger uncertainty.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

32 | P a g e

Figure 6 shows CO2 and CH4 isotherms measured on dry Carboniferous coal from the

Silesia coal mine in Poland using a manometric sorption device. The measurements were

performed in the context of the RECOPOL project of the European Union. For both

isotherms the error in excess sorption capacity upon an increase in temperature of 0.1 K

has been calculated. Evidently this effect is almost negligible for CH4 (<0.1 %). For CO2

however it is quite significant and exhibits a maximum around the critical pressure of CO2

(pc=7.39 MPa) where small temperature changes cause large changes in CO2 density.

Figure 6: Error in calculating CH4 and CO2 sorption isotherm for a temperature increase or

decrease by 0.1 K. Carboniferous coal from Silesia mine (seam 315) of the Upper Silesian

Basin, Poland, measured in the dry state using a manometric method.

Equations of State (EOS)

For mass balance calculations an equation of state (EOS) is required to calculate the density

of the gases (CO2, CH4) at certain pressures and temperatures. There are numerous

different EOS available, however the most commonly used ones are (Span and Wagner,

1996) and (Setzmann and Wagner, 1991) for CO2 and CH4 respectively, and the more

universal EOS of Peng-Robinson (PR) and Soave-Redlich-Kwong (SRK). The two latter ones

can be used for a large suite of gas species, using different interaction parameters. (Mavor,

2004) performed a similar comparison of z-factor variations, including further EOS for CH4.

As a consequence, (Mavor, 2004) pointed out that these differences in EOS can lead to

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

33 | P a g e

variations of up to 20% in the calculated sorption capacities for methane. The comparison

in figure 6 shows for a temperature of 318 K a maximum standard deviation in z-factors of

2 % at 30 MPa. CO2 shows its highest deviation of ~6.5 % at about 12 MPa. More details on

errors related to EOS are provided by (van Hemert et al., 2010).

Figure 7. Comparison of compressibility factors (Z) for CO2 and CH4 at 318 K and pressures up

to 30 MPa. calculated with different equations of state (EOS): Soave-Redlich-Kwong

(SRK), Peng-Robinson (PR) and (Span and Wagner, 1996) (SpW, CO2) and (Setzmann and

Wagner, 1991) (SeW, CH4). Dashed lines indicate the largest difference between the

respective EOS for CO2 (between SpW and SRK; short dashed line) and CH4 (between SW

and PR; long dashed line)

Volume calculations

The accurate volume determination of the reference/pump volume and void volume in the

sample cell are indispensable for accurate prediction of the sorption capacity in

manometric/volumetric setups. Figure 8 and Figure 9 show potential deviations from the

initially calculated sorbed amounts when changing the reference cell volume and void

volume in the sample cell by 0.5 %. Since the reference cell volume is constant

(independent on sample volume) and impacts directly the calculation of the void volume in

the sample cell, the error remains constant over the entire pressure range (0.5 %) for both

gases, CO2 and CH4. Increasing or decreasing the void volume of the sample cell however

leads to an error that is about one order of magnitude larger. For CO2 this error increases

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

34 | P a g e

sharply around the critical pressure and temperature and has an error of 12 % for CO2

(Figure 8) while the error for CH4 increases linearly to 7 % (Figure 9) at the respective final

experimental pressure. These errors will depend on the volume ratio of sample cell to

reference cell (which also depends on the sample volume in the sample cell) and are valid

for the given example only.

Figure 8: Error in calculating CO2 sorption isotherm for an uncertainty in reference and

sample cell volume of 0.5 %. Silesia 315 coal, Upper Silesian Basin, Poland, measured in

the dry state.

Figure 9: Error in calculating CH4 sorption isotherm for an uncertainty in reference and

sample cell volume of 0.5 %. Silesia 315 coal, Upper Silesian Basin, Poland, measured in

the dry state.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

35 | P a g e

Impurities in the measurement gas

Small gas impurities in the void volume can further lead to inaccuracies in the

determination of the true gas density and hence to wrong sorption capacities. Impurities

can be introduced by residual gas in the coal when sample cell evacuation or sample

degassing was not properly performed, i.e. evacuation was too short or the end pressure

not low enough. Further reasons might be helium from void volume determination that

remained in the tubing system or impurities in the measurement gas itself.

(Gensterblum et al., 2010a) showed that by using CO2 with different purities (0.9999

opposed to 0.999999) the difference in CO2 density at 10 MPa and 45°C would be 0.1 %

using the gas-mixture EOS by (Kunz et al., 2007). The same authors analyzed the case when

helium is not accurately evacuated from the system prior to CO2 or CH4 measurement and

assumed a remaining helium partial pressure of 0.1 MPa. This would lead to much higher

errors of 6.1 % under the same p/T conditions as above. Small impurities in the CO2 stream

result in drastic changes in the shape of the isotherm and therefore are easy to detect.

Contamination of the measuring gas can also occur for measurements on samples

containing water. To represent sub-surface moisture conditions, coal samples are typically

moisture equilibrated using a standard test method (ASTM, D1412-93) which saturates the

sample at 30°C and a relative humidity of 96-97 % under vacuum. This will lead to a certain

partial water pressure in the free gas phase leading to uncertainties in the gas density but

also in sample mass.

Figure 10 shows the error in CO2 and CH4 densities at 45°C when assuming a water vapor

partial pressure (in this case 0.009595 MPa). The error is relatively high at pressures <2

MPa but negligible at higher pressure. This uncertainty is not considered to have an impact

on the sorption isotherm.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

36 | P a g e

Figure 10: Plot of pure CO2 versus CO2+H2O and CH4+H2O densities calculated at 45°C and

water vapor pressures from (Wagner and Pruss, 1993). Density calculations have been

performed following the equation of state by (Kunz et al., 2007).

Gas dissolution in water and gas sorption on mineral matter

Gas sorption capacities are typically reported on a dry, ash-free (daf) basis assuming no gas

uptake by dissolution in water or sorption on mineral surfaces. Methane dissolution in

water is known to be more or less negligible whereas CO2 has shown to dissolve in water in

relevant amounts depending on the sample moisture content ((Busch, 2007)).

Gas sorption on mineral surfaces (especially clays) has been found to be significant (Busch

et al., 2008); (Weniger et al., 2010a); (Wollenweber et al., 2010). Although methane sorbs

on clay minerals (RWTH Aachen University, Germany, unpublished data), the total amounts

are small. For CO2 however sorption capacities are in the same order as for coal. Research

on this topic is still in progress and mechanisms are not well understood. Therefore it

might be misleading to report CO2 sorption capacities on a daf basis since dissolution in

water and adsorption on clay mineral surfaces (depending on the respective contents) is

significant and should not be neglected.

Minor sources of errors related to gas sorption on coal

A number of minor sources of error have been reported in the literature (van Hemert et al.,

2007) which are listed below but will not be discussed in detail:

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

37 | P a g e

Gas leakage: Every setup has a certain leakage rate that cannot be avoided and if this

rate is significant compared to the amounts of gas sorbed the mass balance calculation

should account for this.

Thermal and mechanical expansion/contraction of cells in the manometric/volumetric

setups. This is mainly relevant if the volume calibrations are performed with large

pressure differences between reference cell loading and relaxation into the sample cell

or the volume calibrations are performed at a different temperature than the sorption

measurement.

Changes in cell volumes due to changes in pressure

Measuring gas contamination from the steel typically used for cells leading potentially

to changes in gas density.

Sample mass determination: Weight balances usually have an accuracy of 100 g. Since

sample masses of >1 g are typically used this will pose a negligible error of <0.01 %.

Further sources of error are attributed to the coal sample itself (moisture content,

volumetric effects etc.) and will be addressed at a later stage of this manuscript.

Modeling of adsorption isotherms Many methods have been discussed to model sorption isotherms on coal or other

microporous materials and an overview is provided by (Ozdemir, 2003) while other

methods have been proposed by e.g. (Fitzgerald, 2006) and (Sakurovs, 2007).

This paper does not intend to discuss the various isotherm modeling efforts since this

would again be an overview paper of its own. In the (E)CBM industry and related reservoir

simulations approaches the well-known Langmuir equation is used as a simple method and

provides a reasonable fit to most experimental data:

(8)

Here is the Langmuir volume, denoting the amount of gas sorbed at infinite pressure

and is the Langmuir pressure, corresponding to the pressure at which half of the

Langmuir volume is reached.

Excess versus absolute sorption For gases like CO2 or CH4, the excess sorption is defined as the amount gas sorbed

on the sorbent (coal), however without occupying a certain volume for the sorbed phase

( ). Sorption of gas molecules takes place in one or more layers on the

surface at a higher density than the gas phase ( ). (Figure 11,

(Mohammad et al., 2009); (Sircar, 1999)). However, excess sorption didn´t take into

account different accessible volumes for He and the measuring gas like CH4 or CO2

.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

38 | P a g e

Figure 11. Schematic diagram of sorbate density in Gibbsian sorbed phase (simplified and

modified after (Sircar, 1999)).

Gas molecules sorbed occupy a certain volume , hence increase the solid volume. The

density of the sorbed gas can be calculated knowing the sorbed phase volume and

its mass referred to as the absolute sorption:

(9)

Calculating absolute from excess sorption can be done for the manometric (volumetric) and

for the gravimetric methods. The different calculation paths are given below:

Gravimetric method

For the non-sorption case the calculated apparent mass change („weight change“) due to

buoyancy of the sample in the gas (sorptive) at pressure p and temperature T is calculated

according to:

sample

gassamplegas

sample

samplesamplegassamplesample

nsgrav

TPmTP

mmTPVmm

),(1),(),( (10)

(sample mass – (calculated) sample buoyancy in the gas phase (sorptive))

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

39 | P a g e

For the sorption case the measured apparent mass change („weight change”) due to the

combined effect of sorption (mass uptake), buoyancy of the sample and buoyancy

contribution of the sorbed volume can be calculated according to:

sorbate

gas

adsorbed

sample

gas

sample

gasadsorbedgassampleadsorbedsample

s

grav

TPm

TPm

TPVTPVmmm

),(1

),(1

),(),(

(11)

(sample mass + sorbed mass - sample buoyancy – buoyancy of sorbed phase)

Further the excess mass normalized to the initial sample mass is:

(12)

Manometric method

In the non-sorption case (void volume previously determined with He) the calculated

amount of (sorptive) gas accommodated in the void volume of the measuring cell at given

pressure and temperature (requiring determination of the void volume by He expansion

and the equation of state of the sorptive gas) is calculated according to:

),(0 TPVm gasvoidnsman (13)

In the sorption case the measured amount of (sorptive) gas actually transferred into the

measuring cell at given pressure and temperature )( 0adsvoid VV is the residual free gas

volume:

),()( 0 TPVVmm gasadsvoidadssman

(14)

),( TPVmmmm gasadsadsorbednsman

sman

excessman (15)

sorbate

gasadsorbed

excessman

TPmm

),(1

(16)

Both procedures (for manometric and gravimetric methods) yield equivalent expressions

for the excess sorption.

A major problem in determining absolute sorption is that cannot be determined

directly and certain assumptions have to be made. Most authors use the liquid gas density

at the boiling point at ambient pressure which is 1278 and 423 kg m-3 for CO2 and CH4

respectively (Dreisbach, 1999; Mavor, 2004; Yee, 1993). (Murata et al., 2001) provides an

overview of additional approaches to determine values for the sorbed density. Absolute

sorption can be calculated from excess sorption for manometric (volumetric) and

gravimetric methods:

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

40 | P a g e

Especially for CO2 several other attempts have been used to incorporate potential further

changes of the volume of the sorbent like coal swelling. A summary of such effects is

provided by several authors (Siemons and Busch, 2007); (Ozdemir, 2003); (Sircar, 1999);

(Sudibandriyo et al., 2003). (Ozdemir, 2003) used an additional correction factor

accounting for the sorbent volume increase caused by the sorbed phase volume. (Siemons

and Busch, 2007) used a fitting procedure to combine all volumetric effects.

A common procedure to estimate the density of the sorbed phase consists in plotting the

sorbed amount against the free gas density. This approach was used for instance by

(Humayun and Tomasko, 2000). The slope and intercept of the linear portion of this plot

(at high free gas densities) yield the density and the volume of the sorbed phase. For a

Filtrasorb-400 activated carbon the sorbed phase obtained by this method agrees well with

the van der Waals density of 23.45 mmol cm-3 (1032 kg m-3). Similar values were obtained

by (Gensterblum et al., 2009b) for the same activated carbon. For CO2 sorption on natural

coals, however, (Gensterblum et al., 2010a) found higher sorbed phase density values.

These might be due to an increase in the solid volume by a poroelastic expansion of the coal

(coal swelling), not caused by the volume of the sorbed phase alone.

Significant volumetric effects are not expected for CH4 and using a sorbed phase density of

423 kg m-3 to calculate absolute sorption seems to be a generally accepted number. For CO2

this is more complex since every coal sample swells to a different amount, has a specific

pore system with a more or less unique distribution of functional groups which makes the

conversion from excess to absolute sorption non-universal. The use of the graphical

method proposed by (Humayun and Tomasko, 2000) is only appropriate if the

experimental pressure is high enough such that the isotherm shows the typical decline with

increasing pressure or gas density. Further fundamental research is needed to obtain more

reliable information on the density of the sorbed phase and any additional factors that

contribute to an increase in sorbent (coal) volume.

Water sorption on coal

The interaction of carbon materials like natural coal with water is more complex than with

nonpolar gases like helium, argon, nitrogen, methane, or carbon dioxide. This complexity is

due to the weak dispersion interaction of water with coal, the tendency of water to form

hydrogen bonds with other sorbed water molecules and surface chemical species, and the

chemisorptive interaction with the coal mineral matter. Many studies address the issue of

water sorption on coal or gas sorption on moist coal either by describing the principles or

by determining sorption capacities of gases (CO2, CH4) on moist coal samples (Allardice et

al., 2003; Busch et al., 2003c; Busch, 2007; Clarkson and Bustin, 1999a; Clarkson, 2000; Day

et al., 2008c; Fei et al., 2006; Fei et al., 2007; Goodman et al., 2007a; Gutierrez-Rodriguez,

1984; Hartman, 2005; Joubert et al., 1973a; Joubert et al., 1974; Kelemen and Kwiatek,

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

41 | P a g e

2009; Krooss et al., 2002; Lynch and Webster, 1982; Moore and Crosdale, 2006; Nishino,

2001; Ozdemir and Schroeder, 2009; Suuberg, 1993; Unsworth, 1989).

Figure 12 shows the difference in the isotherm shapes that is the result of the combined

effects of carbon hydrophobicity and the presence of surface functional groups that act as

primary sorption sites. It is believed that water molecules are strongly sorbed on the

surface sites via hydrogen bonds. This is followed by further sorption that results in the

creation of water clusters and eventually pore filling (Brennan et al., 2001).

Figure 12: Adsorption of water vapor on oxygenated carbons: (I) heated in vacuum at 200°C;

(II) in vacuum at 950°C; (III) in vacuum at 1000°C; (IV) at 1100°C in a hydrogen stream;

(V) in hydrogen at 1150°C; (VI) in hydrogen at 1700°C; and (VII) at 3200°C ((Muller et al.,

1996)).

Additionally the pore size distribution has an important control on water sorption on coal

or carbon materials in general. It plays a significant role in the observed energies of the

water sorption process (Salame and Bandosz, 1998). Based on X-ray diffraction of sorbed

water molecules in molecular sieving carbons, (Hanzawa, 1998) and (Iiyama, 1997)

concluded that water forms an ordered assembly structure in the hydrophobic nanospace.

The estimated size of such a cluster is about 0.6 nm and is likely associated with the

microscopic anomalies of water confined in small pores (Brennan et al., 2001; Iiyama,

1997). (Alcaniz-Mongue, 2001) found that the process of water sorption is due to both

physical sorption and chemical interaction with surface groups. In accordance with this

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

42 | P a g e

process, micropore filling is progressive: the narrow micropores are filled first, and

subsequently, water is sorbed in the remaining greater range of microporosity. This study

confirms that water sorbed in the porosity of carbon has a solid-phase structure

throughout the whole range of micropore size.

Several studies investigate the structure of water (clustering) in the sorbate phase

(Allardice and Evans, 1971; Salame and Bandosz, 1998; Salame and Bandosz, 1999; Salame

and Bandosz, 2000). Klier and Zettlemoyer (1977) elucidate water cluster stabilities and

referred to quantum mechanical calculations performed mainly by (Del Bene and Pople,

1969). These calculations describe how water molecules form larger clusters, ranging from

dimers to cyclic tetramers. They become energetically more stable with energies per water

molecule changing from -12.8 to-43.9 kJ/mol, respectively.

Coal-specific factors controlling water sorption

Functional groups

The sorption of water vapour and methane on coal surfaces has been studied for some

time, and carboxylic and hydroxylic functional groups have been shown to be important for

this process. (Joubert et al., 1974) determined that methane sorption capacities of high

oxygen coals undergo a much greater reduction when saturated with moisture than do

their low-oxygen counterparts.

The mechanism of water sorption was originally proposed by Dubinin and co-workers

(Dubinin, 1980; Dubinin and Serpinsky, 1981). Their conclusions are based on the results

of experiments where the first points on the isotherm were measured at a relative pressure

of about 0.1. The authors proposed that water sorbs on primary adsorption centres first

(functional groups containing oxygen), followed by sorption on secondary centres (sorbed

water molecules) via hydrogen bonding (Given, 1986; Gutierrez-Rodriguez, 1984; Lynch

and Webster, 1982; Nishino, 2001).

Direct measurement by FTIR (Mu and Malhotra, 1991) and ionic thermal current (Suárez et

al., 1993) confirm the assumption that water vapour is mostly sorbed on carboxyl groups,

where strong interactions occur and the amount of water vapour sorption on hydroxyl

groups is limited.

Coal rank

Several studies report a correlation between coal rank and the extent of water uptake,

particularly at high relative pressures, indicating that water uptake decreases with an

increase in rank, showing a minimum at about 1.2 % VRr. Water uptake then increases

towards higher rank coals (Figure 13, e.g. (Prinz et al., 2004); (Bratek, 2002)). This

parabolic behaviour is similar to the rank dependence of the micropore volume. (Prinz and

Littke, 2005b) applied low pressure sorption isotherms to show that water molecules do

not seem to be able to penetrate the interlayer spacing of crystallite structures (<0.4 nm).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

43 | P a g e

Instead, water is present in the mesopores and larger micropores (~0.4–30 nm). The

results of this study are in general agreement with observations by (Lu, 2001).

The isosteric heat of water sorption on coal

For thermodynamic reasons (G<0 and S<0 because of a higher ordered adsorbed state)

adsorption is an exothermic process and therefore the sorption capacity decreases with

increasing temperature (Sircar, 1992). The temperature dependence of the sorption

capacity is controlled by the isosteric heat of adsorption which is a function of surface

coverage for heterogenous sample such as coals. In general the isosteric heat of sorption

depends on the surface chemistry and the pore structure.

Many published studies show that the isosteric heat of sorption Qst for gases decreases

slightly when adding water. This suggests that there are some higher energetic sorption

sites within the coal that have a greater affinity for these sorbates, but these sites are

preferentially occupied by water molecules (e.g. (Day et al., 2008c)). This is consistent with

the classification by Dubinin and co-workers of primary and secondary sorption sites (see

figure 14). The isosteric heat of water sorption was found to have values very close to the

latent heat of bulk water condensation (45 kJ mol−1 at a surface coverage up to 10%). It was

also found that when pores are too small to accommodate functional groups, the released

heat of sorption at very low relative pressure is equal to the heat of condensation. It has

even been observed to be a few kJ per mol above the heat of condensation (Salame and

Bandosz, 2000). Therefore we can conclude that the isosteric heat of sorption is higher

than the heat of condensation when functional groups are present.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

44 | P a g e

Figure 13: Equilibrium moisture content of coals of varying rank ((Beamish, 1998; Bratek, 2002; Busch, 2007; Clarkson and

Bustin, 2000; Crosdale et al., 1998; Day et al., 2008c; Day et al., 2005; Gasem, 2002a; Hildenbrand et al., 2006a;

Laxminarayana and Crosdale, 1999a; Laxminarayana and Crosdale, 2002; Mastalerz et al., 2004; Moore and Crosdale, 2006;

Ottiger et al., 2006; Ozdemir and Schroeder, 2009; Pan and Connell, 2007; Prinz et al., 2004; Saghafi et al., 2007)).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

45 | P a g e

The low isosteric heat of sorption Qst at high surface coverage suggests that interaction

between the carbon surface and water molecules is weak (Walker Jr., 1993). It was

calculated (Rubes et al., 2010; Rubes et al., 2009) and demonstrated (Avgul and Kiselev,

1970) that the energy of interaction of water molecules and a pure graphite surface is low.

Mahajan (1970) argued that hydrogen bonding should produce higher heats of sorption at

low surface coverage (Mahajan, 1970). They referred to (Darcey et al., 1958), who reported

results for water sorption on Saran charcoals with a heat of sorption of 63 kJ/mol for a

surface coverage of only 1%, whereas at about 5% surface coverage the heat of sorption

approaches the heat of water condensation (45 kJ mol−1) (Darcey et al., 1958; Mahajan,

1991; Walker Jr., 1993).

Figure 14: Net heat of water sorption on coal as a function of moisture content ((Day et al.,

2008c)). The net heat of sorption Qnet is equal to the isosteric heat of sorption Qst and

subtracting the heat of condensation. This relationship is only valid for inert sorbents.

(Huang et al., 1995) discussed changes in the hydrogen bonding patterns for water

molecules bound to high and low densities of oxygen functional groups on the coal surface.

For high oxygen content coals the oxygen functional groups are in close proximity to one

another, enabling the formation of bridged clusters of water molecules.

Salame and Bandosz (1998) observed a similar trend in the heat of sorption on various

activated carbons with different degrees of surface oxidation. Their results showed that the

isosteric heats of water sorption are affected by surface chemical heterogeneity only at low

surface coverage (Salame and Bandosz, 1998). This conclusion is also supported by

Gubbins and co-workers and based on a good agreement between molecular simulations

(Maddox et al., 1995; Muller et al., 1996; Ulberg and Gubbins, 1995) and experimental data.

It was shown that on the surface of activated carbons with functional groups the creation of

water clusters occurs before all primary sorption centres are occupied by water molecules

(Salame and Bandosz, 1998; Salame and Bandosz, 1999; Salame and Bandosz, 2000).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

46 | P a g e

Referring to the competitive sorption of gases, CO2 and CH4 are forced to lower energy sites

when water is present. Water appears to be sorbed on functional groups, whereas CO2 and

CH4, although showing some degree of preferential sorption, are generally able to access all

of the sorption sites throughout the coal (Day et al., 2008c).

Water sorption/desorption hysteresis

(Puri, 1966) reported sorption-desorption hysteresis for activated carbons with different

surface chemistry. With the presence of acidic oxygen surface groups the sorption-

desorption isotherm is not completely reversible at zero relative vapour pressure. A certain

amount of water cannot be desorbed under vacuum (10-4 mbar), only by additionally

elevating the temperature. This residual moisture is irreversibly sorbed water and is

presumably held tightly within the micropores (Mahajan, 1970). (McCutcheon, 2001) and

Charrière and Behra (2010) found that the extent of hysteresis for sorption/desorption

(integrated area difference) correlates linearly with the coal oxygen content. The data of

(Mahajan, 1970) supports this hypothesis. Contrary, Allardice and Evans (1970) suggest

that hysteresis occurs because the swelling is proportional to the water sorbed. However,

they suggest that the corresponding shrinkage of the structure during desorption is

pressure delayed by water strongly sorbed on primary sorption sites on the internal

surface of the coal, and this sorbed water appears to be confined to the monolayer region.

Figure 15 shows schematically a water isotherm. Monolayer sorption occurs in the lower

relative pressure range of the isotherm. A sharp increase in the isotherm, where the

relative water vapour pressure is still below 0.1 is attributed to water sorption on the

primary sorption sites. Multilayer condensation is responsible for the straight line region in

the middle section of the isotherm. Capillary condensation is the dominating factor in the

upper concave part of the isotherm. The inflection point is the pressure where capillary

condensation kicks-in. (Allardice and Evans, 1971) showed that this inflection point shifted

to higher relative pressures with increasing temperature.

(McCutcheon, 2001) and Charrierre and Behra (2010) examined two parts of the hysteresis

while studying water sorption on coals with different ranks. Low and high relative pressure

hysteresis have occurred especially for lower rank bituminous coals. McCutcheon 2001 and

McCutcheon and coworkers 2003 and Charrière and Behra 2010 have suggested that the

hysteresis at high relative pressure is due to the ink-bottle effect, which would require a

condensation (a liquid phase).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

47 | P a g e

Figure 15: Schematic of water uptake on coal.

Horikawa and his coworkers 2010 observe recently that the water cluster size in

mesopores is larger than in micropores, and the hysteresis loop of sorption–desorption in

mesopores is greater than in micropores (Horikawa et al., 2010).

Finally, we would like to underline that most of this hypothesis is based on only a few

experiments, and in some cases, on indirect observations. To verify this hypothesis, further

investigations are needed (Gauden, 2005).

CH4 and CO2 sorption of coals in the presence of water

Effects on sorption capacity

As early as 1936, Coppens reported the observed decrease in methane sorption capacity

due to moisture in several Belgian coals (Coppens, 1936). Polar sites such as hydroxyl

groups on the coal surface are preferentially occupied by water and in the process reduce

the capacity for CO2 and CH4. Therefore moist coals have a significantly lower maximum

sorption capacity for either gas compared to dry coal. However, the extent to which the

capacity is reduced depends on coal rank. Higher rank coals are less affected by the

presence of moisture than low rank coals (Day et al., 2008c).

Sorption capacity decreases with increasing moisture content until the equilibrium

moisture content is reached. Above this moisture content the gas sorption capacity remains

constant (Day et al., 2008c; Joubert et al., 1973a; Joubert et al., 1974). This limiting

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

48 | P a g e

moisture content depends on the rank of the coal and, for Australian low rank coals,

corresponds approximately to the equilibrium moisture content that would be attained by

exposing the coal to about 40–80% relative humidity. Allardice and Evans (1970) conclude,

based on entropy calculations, that water above this limit has a liquid structure. Therefore

they conclude water condensates on the surface and that gas will only dissolve in this

water (Allardice and Evans, 1971). The experimental results indicate that the loss of

sorption capacity by the coal in the presence of sorbed water can simply be explained by

“volumetric displacement of CO2 and CH4 by water” (Day et al., 2008c; Levy et al., 1997). An

interesting consequence of volumetric displacement would be that estimates of wet coal

sorption capacity could be calculated from dry coal isotherms. Dry samples are

experimentally easier to handle in comparison to moisturized samples. The assumption

behind this hypothesis is that the density of the sorbed phase is constant for comparable

coal types. For Australian and Chinese bituminous coals, approximately 0.3 molecules of

CO2 are displaced by each water molecule and each molecule of water displaces 0.2

molecules of CH4 up to the limiting moisture content (Day et al., 2008c). To verify a general

validity of this hypothesis, further investigations are necessary.

The fact that CO2 and CH4 continue to be sorbed on moisture-equilibrated coals indicates

that water and these two gases compete only for a portion (Figure 16, intersection α and β

or γ) of the total available sorption sites. Figure 16 shows schematically the total set of

sorption sites for CO2, CH4 and water on coal at constant temperature and pressure. Carbon

dioxide has the highest amount of sorption sites, followed by methane and water. The ratio

of sorption capacity or sorption sites of CO2 and CH4 is in the range of 1 to 9 (s. below) and

changes strongly to lower ratios when water content decreases or coal rank increases. The

intersection area of water and methane γ is smaller than the intersection area of water and

CO2 (β), as illustrated in Figure 16. This intersection area is based on the observation of

secondary sorption sites of water. Intersection δ represents the primary sorption sites of

water. If there is water present in the system, these sorption sites will be occupied by water

molecules, because of the higher heat of sorption for water on the primary sorption sites

compared to CO2 and CH4. However, the intersection areas and their respective ratios of

these will change when temperature and pressure are varied. For example at very low total

surface coverage the amount of sorbed water will be high in comparison to CH4 and CO2,

because of the high energetic primary sorption sites.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

49 | P a g e

Figure 16: Sorption sites of each gas at one given fixed surface coverage (fixed p,T) and the

intersection for multi-component sorption isotherms.

The coal surface can be envisaged to consist of sets of more water-prone (hydrophilic) and

more gas-prone sorption sites and there is a partial overlap of these sets of sorption sites.

For dried coals, sorption capacity shows a parabolic dependence on rank, whereas as-

received coals display a more linear dependence on fixed carbon content (Figure 18,

(Ozdemir and Schroeder, 2009)).

CH4 sorption capacity and the influence of moisture at elevated temperatures

High-pressure methane sorption isotherms are in general less sensitive to changes in

temperature than to variations in moisture content (Day et al., 2008c; Joubert et al., 1974;

Moore and Crosdale, 2006; Ozdemir and Schroeder, 2009). As discussed in the previous

chapter, CH4 sorption capacity decreases with increasing moisture content until moisture

saturation is reached. Above this moisture saturation limit, the gas sorption capacity

remains constant (Day et al., 2008c; Joubert et al., 1973a; Joubert et al., 1974).

However, an increase in temperature results in a decrease in moisture-saturation. Lower

equilibrium moisture contents will increase the gas sorption capacity; higher temperature

will decrease the sorption capacity (Day et al., 2008c; Joubert et al., 1974; Moore and

Crosdale, 2006; Ozdemir and Schroeder, 2009). This relationship is illustrated in Figure 17

and confirmed experimentally by (Gaschnitz, 2000) on coals from the Ruhr Basin,

Germany.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

50 | P a g e

Figure 17: Schematic diagram of sorption capacity of coals at different moisture contents and

as a function of temperature.

Which of these opposing effects has a larger influence on gas sorption capacity is property

specific for each coal and depends on the isosteric heat of sorption and the surface

coverage (amount of available sorption sites). Sorption of water on primary sorption sites

(Figure 16) results in higher sorption enthalpies compared to CH4 or CO2. Therefore, it is

unlikely to observe competition for the primary sorption sites of water. For sorption on the

secondary sorption sites, sorption enthalpies for water and each gas type are likely to

become more similar, especially when the moisture content of the sample is close to the

saturation content. Accordingly we observe a significant competition for this sorption site

between methane and water molecules. Up to a sample-specific limit additional water is

only present as free water on the surface coal without occupying sorption sites.

Crosdale and his coworkers investigated 2008 the influence of moisture and temperature

on the CH4 sorption capacity of low-rank coals. Their results show no significant

temperature dependence for CH4 sorption capacity at constant moisture content (Figure

19, Case 2). The common expectation for sorption isotherms as a function of temperature

at different moisture-levels would be that increasing temperature will cause a decreasing

sorption capacity (Figure 17, Figure 19 Case 1). However, due to the high displacement

ratios (see previous section) it is possible that the release of water sorption sites could be

able to compensate the decrease in CH4 total sorption sites due to the higher temperature.

Nevertheless, further studies are required to elucidate this observation.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

51 | P a g e

Figure 18: Sorption capacity at reference pressure of 5 MPa as a function of coal rank for moist and dry coals ((Bae and Bhatia,

2006; Battistutta et al., 2010; Busch et al., 2003b; Bustin, 2004; Chaback, 1996; Day et al., 2008c; Fitzgerald et al., 2005;

Harpalani et al., 2006; Kelemen and Kwiatek, 2009; Li et al., 2010; Mastalerz et al., 2004; Pini et al., 2010; Reeves, 2005; Ryan,

2002; Sakurovs, 2007; Weniger et al., 2010a; Yu et al., 2008b)).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

52 | P a g e

Figure 19: Schematic plot of the sorption capacity in the dry and moisturized state as a

function of temperature.

Water sorption kinetics on coal

A novel application of an empirical kinetic model was proposed by (McCutcheon, 2001) to

quantify water sorption kinetics in bituminous coals. They introduced a flow rate kinetic

parameter to quantify the rate of water transfer from outside the coal particle to the intra-

particle pore structure. Most of the bituminous coals investigated by a gravimetric method

show similar uptake rates at relative water vapour pressures in the range of 0.2-0.9.

However for two semi-anthracites the flow rate was significantly less than for all other

bituminous coals investigated. (McCutcheon, 2001) believe that this difference is caused by

differences in the coal structure. Variations between coals in the uptake rate at low relative

pressures (0-0.2) appear to be caused by variations in the primary sorption sites. Effective

diffusivities for water uptake on coal have been measured by two authors and are

summarised in Table 1.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

53 | P a g e

Table 1: Effective water diffusivities on coal ((Charrière and Behra, 2010; McCutcheon et al., 2003)).

Coal type T

[K]

relative

pressures

[/]

grain size

[mm] model

uptake rate

[cm³ min-1 g-1]

effective

diffusivity

[s-1]

Source

bituminous 299 0.05<0.9 <0.25 empirical

kinetic model 0.07-0.015 10-2 - 10-3 McCutcheon

high volatile

bituminous, lignite 298 0.1<0.9 0.04-0.25

unipore

model n.E. 10-4 - 10-5 Charriere

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

54 | P a g e

Dependence of CO2, CH4 and water sorption on coal properties

The influence of coal composition on gas sorption Most sorption analyses of coal are carried out on dried samples and the sorption capacity is

reported to this basis. These results can be modified to expressions having another basis,

as illustrated in Figure 20.

Figure 20: Relationship of different analytical bases to various coal components

The moisture content of moisture-equilibrated coals is defined as the maximum water

sorption capacity of the specific coal. The “as received” moisture content of coals is not a

fixed coal specific value, because the moisture content is in equilibrium with the

surrounding or environmental humidity. It depends on the oxygen content of the coal,

because the functional groups represent more energetic sorption sites for water molecules.

Therefore a coal with a higher oxygen content will have a higher “as received” moisture

content. Derived from the thermodynamic equilibrium between the chemical potential of

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

55 | P a g e

the water partial pressure in the free gas phase and the sorbed phase of water, it is obvious

that the amount of occupied sorption sites is higher for a coal with a higher density of

functional groups. If the water partial pressure is reduced using only drying agents without

elevating the temperature, an “air-dried“ moisture content of the coal is achieved. The “air-

dried” moisture content is the coal specific moisture content.

Furthermore, the ash and total organic carbon (TOC) contents are sensitive to sample

heterogeneity. Often an increase in minerals matter in sieve fractions (especially at particle

diameters less then 100µm) is observed.

Total Organic Carbon (TOC) and ash content

(Weniger et al., 2010a) confirmed recently that CH4 and CO2 sorption capacities on coal can

be correlated with the total organic carbon content (TOC) (Figure 21). For carbon dioxide

the linear regression showed a non-zero intercept, indicating a significant sorption capacity

of the mineral matter for this gas species. This was not the case for CH4 (Weniger et al.,

2010a). These findings support our conclusion that CH4, CO2 and water compete for a part

of all sorption sites (figure 16).

The sorption capacity of coal for CH4 and CO2 is negatively correlated with ash content

(Bustin and Clarkson, 1998; Faiz et al., 2007; Laxminarayana and Crosdale, 1999a;

Laxminarayana and Crosdale, 2002; Yee, 1993). From this observation we conclude that

organic matter controls the CH4 and CO2 storage capacity. However the mineral content

accommodates a small additional CO2 sorption capacity

Figure 21: Sorption capacity of CH4 and CO2 as function of total organic carbon (TOC)

(Weniger et al., 2010a).

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

56 | P a g e

Coal Rank Besides the effect of mineral matter or ash content on the sorption capacity, the rank of

coals and related coalification processes has an important impact on the CO2 and CH4

sorption capacity. In some rank ranges the maceral composition seems to be dominant.

The study by (Prinz, 2004) shows a parabolic shape for the CH4 sorption capacity of dry

coals as a function of rank for the Ruhr Basin, Germany. For moisture-equilibrated coals,

the CH4 sorption capacity shows a slightly linear increasing trend with rank.

(Laxminarayana and Crosdale, 2002) investigate the CH4 sorption capacity of Indian coals.

Sorption isotherm analysis of dry samples showed that the sorption capacity follows a

second-order polynomial trend with rank (0.62% up to 1.46% R0 max). Moisture-

equilibrated samples showed a linear increase in sorption capacity with rank and a

significantly reduced sorption capacity compared to dry coals. The observed effective

diffusivity (

) decreased with rank. This was attributed to an increase in micro-

porosity during coalification. Bulk coals (i.e. a coal sample with seam specific mixture of

dull and bright layers) tested showed 2–3 times larger effective diffusivities than bright

coals, while dull coals have intermediate values.

(Levy et al., 1997) suggest that the minimum in sorption capacity with rank may

correspond to a coalification step, which occurs at around 87% fixed carbon. This step is

characterised by a reduction in the oxygen and water content of the coal and an increase in

methane release ((Stach et al., 1982)). More recent investigations (Prinz and Littke, 2005b;

Prinz, 2001)) have shown a minima in micropore volume of dry coals in the same rank

range. This might be due to plugging of the micropore structure by liquid hydrocarbons in

high-volatile bituminous coals, thus reducing its micropore volume At higher coal maturity

these volatiles will undergo thermal cracking hence liberating surface area which results in

an increased sorption capacity (Rice, 1993).

Further, (Saghafi et al., 2007) considered the effect of rank on CO2 sorption capacity for the

Sydney basin, but only up to pressures of 5 MPa.

Dull/bright coals and maceral composition

Besides coalification the primary composition of the coal has a significant influence on the

pore structure and therefore on the sorption capacity. Different coal lithotypes have

different sorption capacities, because the source material is different. Bright coal is

vitrinite-rich and has a lower inertinite content, while the contrary holds for dull coal. This

difference in source material results in different porosity, pore structure and therefore,

also in different sorption rates ((Laxminarayana and Crosdale, 2002)).

The effect of coal composition, particularly the organic fraction, on gas sorption has been

investigated for several worldwide coal basins (Table 2). The gas sorption data from

(Gamson et al., 1996) suggests that both, dull and bright coals, can be divided into two

categories: coals with rapid and coals with slow sorption behaviour.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

57 | P a g e

Figure 22: CH4 and CO2 sorption capacity of dry coals and the influence of rank. The two lines show the second-order polynomial

fit for CH4 and CO2. The regression factor for methane is in an acceptable range, however for CO2 it is not. Because of the lack

of sorption data of coals at higher coal rank, the reliability of Figure 22 and Figure 23 has to be proven. So far, the crossover

of the trend lines in these figures has no significant meaning.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

58 | P a g e

Figure 23: CH4 and CO2 sorption capacity on moisture-equilibrated and as received coals as a function of rank. The two lines show

the linear fit for CH4 and CO2. The regression factor for CH4 is in an acceptable range; however for CO2 the trend is weaker and

dominated by the data point at high vitrinite reflectance.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

59 | P a g e

Further research on the effect of petrographic composition on CH4 sorption capacity

showed that bright coals have a higher sorption capacity than dull coals of the same rank

(Crosdale and Beamish, 1993; Lamberson and Bustin, 1993; Bustin and Clarkson, 1998;

(Crosdale et al., 1998); Clarkson and Bustin, 1999; Laxminarayana and Crosdale, 1999;

Mastalerz et al., 2004). Other authors found no correlation between sorption capacity and

maceral composition (Carroll, 2003; Faiz et al., 2007; Laxminarayana and Crosdale, 2002).

(Chalmers and Marc Bustin, 2007) observed that for the CH4 sorption capacity the maceral

composition is more important in high rank coals. It is found that the CH4 sorption capacity

shows a moderate positive trend with microporosity and both increase with rank.

Microporosity positively correlates with vitrinite content (in particular the telovitrinite

(structured vitrinite) content). Furthermore, the same authors demonstrate that liptinite-

rich samples show high CH4 sorption capacities but low micropore volumes, which might

be attributed to the gas being held in solution. Due to a generally low liptinite content in

coal, the effect of liptinite on CH4 sorption capacity is superseded by the sorption capacity

of more dominant macerals and therefore, the effect of liptinite has not been fully

examined. For low-rank coals there is no significant difference between bright and dull

samples except for one coal with the dull sample having a greater sorption capacity than its

bright equivalent. For higher-rank coals, the bright samples have a higher CH4 sorption

capacity than the dull samples and the difference between sample pairs (bright and dull)

increases with coal rank (Chalmers and Marc Bustin, 2007).

From investigations of gas sorption uptake of activated carbon, metal organic frameworks

and zeolites, it is well known that the specific surface area is inherently linked to the pore

size distribution of a sample whereby the specific surface area progressively increases with

decreasing pore size for samples with similar porosity. Therefore, vitrinite with high

microporosity can store more CH4 than inertinite (Lamberson and Bustin, 1993); Beamish

and Crosdale, 1995) because inertinite contains more macroporosity and less

microporosity than their vitrinite equivalents (Unsworth, 1989). Similar studies with CO2

are not available, it can however be speculated that they will follow a similar trend as with

CH4 because of the higher surface area in microporous vitrinite compared to more meso-

/macroporous inertinite.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

60 | P a g e

Table 2: World-wide coal basins investigated for gas sorption capacity.

Basin Country Reference

Bowen and Sydney Australia (Laxminarayana and Crosdale, 1999a;

Laxminarayana and Crosdale, 1999b); (Levy et al.,

1997)

Hunter Valley and

Illawarra coal regions

Australia (Day et al., 2008a)

Sydney Australia (Saghafi et al., 2007)

- India (Laxminarayana and Crosdale, 2002)

Raniganj, Jharia,

South Karanpura

coalfield

India (Dutta et al., 2011)

Ruhr Basin Germany (Prinz et al., 2004; Prinz, 2001)

Campine Basin Belgium (Hildenbrand et al., 2006a)

Upper Silesian Coal

Basin

Poland (Busch et al., 2004; Busch et al., 2006; Hemza et al.,

2009; Kedzior, 2009); (Weishauptová and

Sýkorová, 2011)

Paraná Basin Brazil (Weniger et al., 2010a)

Argonne premium

coals

USA (Busch et al., 2003b; Goodman et al., 2007a;

Goodman et al., 2004; Kelemen and Kwiatek, 2009;

Mastalerz et al., 2004; Ozdemir, 2004; Ozdemir and

Schroeder, 2009)

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

61 | P a g e

Temperature

Sorptive surface coverage θ at a specific gas pressure decreases with increasing

temperature as derived from thermodynamics. At the same time the sorption capacity also

decreases with increasing temperature for the pressure range that can be realised in

laboratories (up to 50 MPa). This has been observed on sorption isotherms for dry coals

(e.g. (Lama, 1998; Levy et al., 1997; Sakurovs et al., 2008); (Crosdale et al., 2008). However

from thermodynamics the sorption capacity of a specific sample is similar for all

temperatures at infinite gas pressures, assuming no change of the sample surface area over

this temperature range.

The impact of varying temperature on equilibrium constants like the Langmuir pressure K

is given by the van 't Hoff equation:

(17)

or

(18)

The variation of the Langmuir pressure K must be isosteric at constant surface coverage θ.

The enthalpy ΔH is negative, because sorption is an exothermic reaction. R is the universal

gas constant.

The surface coverage at constant gas pressure decreases with increasing temperature,

however the magnitude of this decrease varies for all gases and depends on the sample/gas

specific enthalpy of sorption.

Isosteric heat of adsorption

Isosteric heats of adsorption can be determined experimentally by measuring sorption

isotherms at several temperatures. Here, isosteres are constructed by plotting the pressure

as a function of the reciprocal temperature at a constant surface coverage θ. From the

Clausius–Clapeyron equation, the isosteric heat of adsorption Qst is given by:

(19)

From a plot of ln p versus 1/T the sorption isostere is received, the slope can be used to

calculate Qst. If the relative pressure p/p0 is used instead of p, the net heat of sorption is

obtained according to:

(20)

while Qvaporisation is heat of vaporisation which is a fixed value for each gas. p0 is the

saturation vapour pressure and R the universal gas constant.

The heat of adsorption for both CO2 and CH4 decreases slightly with increasing moisture

content. A fundamental requirement of an equilibrium is that sorption sites are

progressively filled (Figure 16 α, β and δ). The available higher energy sites for CO2 and CH4

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

62 | P a g e

on dry coal are already occupied when water is present, gas molecules will compete with

water for the same sorption sites, namely the functional groups.

This interpretation or hypothesis is contrary to some previous studies that have indicated

no specific adsorption sites for CO2 (Amarasekera et al., 1995); (Goodman et al., 2005).

Other evidence supporting the view that CO2 may be preferentially sorbed at carboxyl

functional groups was reported by (Nishino, 2001). These sites would be among the first to

be occupied by water and would therefore lead to a reduction in the heat of adsorption for

both CO2 and CH4. Another hypothesis for the observed reduction in heat of adsorption is

related to the suggestion that it increases with decreasing pore size (Stoeckli, 1990). In this

situation, progressively filling (condensation) the smallest pores which are highest energy

pores by water molecules would reduce the heat of adsorption of CO2 and CH4.

CO2 versus CH4 sorption

Several studies report CO2 and CH4 sorption on the same coal sample under different

moisture conditions to obtain information on the selectivity of the coal for either gas

species. For all these datasets (Bae and Bhatia, 2006; Battistutta et al., 2010; Busch et al.,

2003b; Busch et al., 2004; Bustin, 2004; Chaback, 1996; Day et al., 2008c; Harpalani et al.,

2006; Kelemen and Kwiatek, 2009; Li et al., 2010; Mastalerz et al., 2004; Pini et al., 2010;

Reeves, 2005; Sakurovs, 2007; Weniger et al., 2010a; Yu et al., 2008b) higher CO2 compared

to CH4 sorption was observed while the CO2/CH4 sorption ratio varies from 1.1 to 9.1

(Figure 24). Interestingly a relative decrease in sorption ratio with coal rank can be

observed for moisturized coal samples, having a ratio of ~9 at low coal rank and decreases

to 1.2-1.5 for anthracite rank coals. For dry coal such a relationship with coal rank was not

observed. Data scatter between ~1 and 3 for different coals maturities.

For single gas measurements higher sorption of CO2 compared to CH4 was always observed

for the same coal sample under the same pressure and temperature conditions. However a

few mixed-gas sorption measurements reported preferential sorption of CH4 compared to

CO2 specifically at low pressures (e.g. (Busch et al., 2006; Busch et al., 2003d; Crosdale,

1999; Majewska et al., 2009). Other authors again report clear preferential sorption of CO2

from binary gas mixture experiments (e.g. (Fitzgerald et al., 2005; Ottiger et al., 2008; Pini

et al., 2010). The reason for the different observations is not explained in the literature and

further work on gas mixture sorption on coal is recommended.

CO2 and CH4 sorption kinetics on coal

Numerical modeling of the processes related to CBM and CO2-ECBM requires information

on the kinetics of the sorption and desorption processes. In principle diffusion determines

the rate of CH4 desorption for primary recovery and the rate of CO2 sorption for CO2

storage in coal. For CO2-ECBM both processes are of importance and the counter-diffusion

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

63 | P a g e

of CH4 diffusing out of the coal matrix and CO2 diffusing into the coal matrix is determined

by their respective sorption rates.

In order to better understand sorption rates in coal the pore size distribution should be

explained in more detail. Coal is a highly heterogeneous material that develops with burial

in sedimentary basins while coal maturity increases with temperature. In coal, four

different pore systems are commonly distinguished: cleat-porosity, macro-, meso- and

micro-porosity. In each of these porosities different transport processes take place that

need to be understood individually. Commonly, gas transport in coal is considered to occur

at two different scales: (I) laminar flow through the cleat system, and (II) diffusion through

the coal matrix. Flow through the cleat system is pressure-driven and may be described

using Darcy’s law, whereas flow through the matrix is assumed to be concentration-driven

and is modeled using Fick’s law of diffusion. Gas storage by physical adsorption occurs

mainly in the coal matrix (Harpalani, 1997); (Busch et al., 2004).

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

64 | P a g e

Figure 24: CO2/CH4 sorption ratio for different moist and dry coal sample at various temperatures as a function of coal rank.

Values were picked between 1 and 5 MPa. Data fit for moist samples only (Bae and Bhatia, 2006; Battistutta et al., 2010;

Busch et al., 2003b; Bustin, 2004; Chaback, 1996; Day et al., 2008c; Fitzgerald et al., 2005; Harpalani et al., 2006; Kelemen

and Kwiatek, 2009; Li et al., 2010; Mastalerz et al., 2004; Pini et al., 2010; Reeves, 2005; Ryan, 2002; Sakurovs, 2007; Weniger

et al., 2010a; Yu et al., 2008b)

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

65 | P a g e

In general, within the coal porous system three distinct diffusion mechanisms can be

distinguished: (1) Diffusion described by Fick’s law where inter-molecular collisions

between gas molecules (Brownian motion) is the dominant mechanism. This process is

significant for large pore sizes and/or high gas pressures; (2) Knudsen diffusion

(concentration-driven) by collisions between the gas molecules and pore walls. It is

important when the mean free path of the gas molecules is larger than the pore size. This

process can be significant for small pore sizes and low gas pressures; (3) surface diffusion

where sorbed molecular species move along the pore wall surface. This process is

important in micropores with strongly sorbed gas species.

(van Krevelen, 1993) defines the matrix pore system as follows:

Macropores: The pore size is larger than the mean free path of the molecules.

Mesopores: The pore size is smaller than the mean free path of the molecules.

Micropores (< 2 nm): Activated diffusion

It should be noted here that the mean free path length depends on temperature and

pressure, so this classification cannot be applied to specific pore radii.

This classification is further specified and classified by the International Union of Pure and

Applied Chemistry ((IUPAC, 2001)):

Macropores: pores with widths exceeding about 0.05 m (50 nm)

Micropores: pores with widths not exceeding about 2.0 nm (20 A)

Mesopores: pores of intermediate size (2.0 nm < width < 50 nm)

In addition some authors proposed ultramicropores, defined as the interlayer spacing of

the crystalline carbon structures (d002-spacing) which is <0.4 nm and decreases with rank

(Hirsch, 1954; Prinz and Littke, 2005b).

Using low pressure sorption techniques applying the BET method ((Brunauer, 1938)),

(Prinz and Littke, 2005b; Prinz et al., 2004) determined surface areas for

meso/macropores (using N2) and for micro/ultramicropores (using CO2) on a suite of

Carboniferous coals from the Ruhr Basin, Germany. This suite ranges in coal rank from high

volatile bituminous (h.v.b.) to semi-antracite coals (Table 3). In addition the same authors

(Prinz et al., 2004) compared low-pressure N2 sorption with small angle neutron scattering

(SANS) data and found similar results for the specific mesopore surface area of different

rank coals from the Ruhr Basin, Germany (Figure 25). Further data indicates that

significant amounts of closed porosity are only present in low-rank coals of the sample set

studied. Further, they found that the microporosity increases with coal rank while at the

same time meso/macroporosity decreases. This indicates that coal is dominated by a

bimodal pore size distribution up to medium volatile bituminous (m.v.b.) and above this

rank coals can rather be classified by a unimodal distribution (Figure 26).

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

66 | P a g e

These porosity trends with coal rank should have implications on the sorption rate. It is

indicated that a unipore approach is rather suitable for high rank coals while a bidisperse

approach is better suited for lower rank coals.

Figure 25: Specific surface areas of meso and macropores as obtained from small angle

neutron scattering (SANS, circles) and low-pressure N2 adsorption experiments (triangles

and crosses refer to different evaluation methods) (after (Prinz et al., 2004).

When evaluating the rates of gas sorption by manometric or gravimetric methods it should

be kept in mind that the recorded pressure or weight changes only provide a net transfer

rate within the closed system from the gas phase to the sorbed phase (gas uptake).

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

67 | P a g e

Figure 26: Specific surface areas for meso/macropores and micro/ultramicropores

determined on a suite of Carboniferous coals from the German Ruhr Basin ((Prinz and

Littke, 2005b; Prinz et al., 2004)). Open symbols represents the mesopore surface area

determined with N2 at 77K.

Unipore sorption/diffusion models The unipore model is based upon the solution to Fick’s second law for spherical symmetric

flow ((Clarkson and Bustin, 1999b)):

(21)

with the initial condition:

(22)

where r is the sphere radius (particle radius) and C the sorbate concentration within the

sphere, D the diffusion coefficient and t is time. The model assumes isothermal conditions,

a homogeneous pore structure and a non-dependence of the diffusion coefficient on

concentration and location in the coal particle.

There are several solutions to Fick’s Law used in the literature for gas diffusion in coal: For

representing gas desorption from coal or gas sorption on coal, (Clarkson and Bustin,

1999b) and (Pan et al., 2010) consider a sphere of radius r which is initially at a uniform

concentration C1 while the surface concentration is maintained constant at Co. Then the

total amount of substance entering or leaving the sphere is given by (Crank, 1975):

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

68 | P a g e

(23)

where Mt is the total amount of the diffusing species at time t and M is the total desorbed

mass.

It should be noted that the model assumes spherical, homogeneous coal particles of

identical radius with smooth surfaces. Although this model is often referred to as “unipore

model”, a pore system within the spherical particles and surface sorption on the pore walls

is not explicitly assumed.

Busch and his co-workers 2004 (and applied in a similar form by (Pone et al., 2009)) used a

diffusion model that assumes a sphere with radius r, placed in a limited volume where the

free volume (i.e. not occupied by the particles) is V. The concentration of sorptive gas in the

free volume is always uniform and is initially C0. The initial concentration of sorbate within

the spheres is zero. The total amount Mt of gas sorbed after time t is expressed as a fraction

of the corresponding quantity M after infinite time by the relationship (Crank, 1975):

(24)

here qn are the non-zero roots of:

(25)

and

(26)

For small times t a simplified model was used by (Charrière et al., 2010), based on (Carslaw

and Jaeger, 1959):

(27)

Here a plot of Mt/M versus t1/2 will have a slope of 6(D/)1/2. This approach is used by

(Smith, 1984a; Smith, 1984b) to obtain diffusion coefficients for CH4 desorption from coal

seams. However it has limitations for CO2 rather than for CH4 as already noted by

(Charrière et al., 2010): When CO2 is expanding during a gravimetric or manometric

sorption experiment into the measuring cell where the sample is placed (s. above) it will

cool due to the Joule-Thompson effect. Especially at small times this cooling may have a

large effect on the gas uptake curve, mass balance calculation or buoyancy correction (for

gravimetric methods) and might therefore have a significant influence on the calculation of

the diffusion coefficient. This effect is lower for CH4.

Bidisperse sorption/diffusion models

Bidisperse sorption/diffusion models can be used for coals that are characterized by two

distinct pore systems (s. above). (Ruckenstein, 1971) define these as an agglomeration of

many microporous spheres that are contained in macropores. They developed a model

which assumes a linear isotherm in both macro -and microspheres and a step change in the

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

69 | P a g e

boundary concentration at the start of a sorption step. Model equations are provided in the

original publication. A more detailed description of the bidisperse porous coal system was

developed by (Bhatia, 1987).

Figure 27: Bidisperse pore model as defined by (Ruckenstein, 1971).

The Ruckenstein model was later extended to model gas flow and transport in fixed beds

by (Smith and Williams, 1984) to investigate methane desorption rates in coalbeds, and

later by (Shi, 2003a; Shi, 2003b) to study the mechanisms of CO2-ECBM.

(Clarkson and Bustin, 1999b) state that the Ruckenstein model cannot be applied for high-

pressure sorption experiments. The reasoning is that the model assumes a step change in

external (to sorbent particles) concentration of the diffusing gas at time zero, and that this

concentration remains unchanged with time. This assumption is not true for constant

volume, variable pressure sorption rate experiments. In addition CO2 and CH4 sorption

isotherms are typically non-linear for natural coal which further limits the applicability of

the Ruckenstein model.

Further bidisperse gas uptake models were used by various authors (Busch et al., 2004;

Clarkson and Bustin, 1999b; Siemons et al., 2007; Yi et al., 2008). The basic principle of all

these approaches is similar although the complexity varies largely. (Busch et al., 2004) use

a linear combination of two first-order reactions and provide, in contrast to the unipore

model, a perfect fit of the experimental pressure decay curves for gas uptake for a low-

mature coal. They do not speculate on the properties of the two different pore systems for

the combined sorption/diffusion but rather classify the transport as “fast” and “slow”. For

volumetric gas sorption experiments the model proposed by (Clarkson and Bustin, 1999b)

accounts for nonlinear sorption in the micropores, as opposed to the Ruckenstein model

(Ruckenstein, 1971), and a time-varying gas pressure in the free gas volume in contact with

the coal particles. (Cui, 2004) used a bidisperse model based on previous work by

(Ruckenstein, 1971) and (Clarkson and Bustin, 1999b).

Results from CO2 and CH4 gas sorption kinetic experiments

Sorption kinetic data are obtained by monitoring the rate of pressure equilibration during

individual pressure steps of volumetric sorption tests or the rate of apparent weight

changes during gravimetric sorption experiments. These measurements can be readily

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

70 | P a g e

combined with the determination of sorption isotherms for the assessment of the gas

sorption capacity (e.g. (Clarkson and Bustin, 1999b)).

A summary of the various findings from sorption kinetic measurements are provided in

Table 3.

Ciembroniewicz and Marecka, 1993 performed CO2 sorption kinetic experiments on two

Polish hard coals at temperatures below 35°C and pressures up to 0.065 MPa. Although the

two coals were of similar rank, they exhibited different sorption kinetics. The sorption rate

for one coal was significantly faster than for the other one while no significant difference in

the rates was observed between sorption and desorption.

Marecka, 1998 studied CO2 and CH4 sorption rates on semi-anthracite coal using different

grain size fractions. They found decreasing sorption rates with increasing particle size (i.e.

lower specific surface area), slightly increasing rates for increasing temperatures and

higher rates for CO2 compared to CH4 (factor ~10-20). This latter observation is reported

by almost all authors under all temperature or pressure conditions and for almost all coal

samples and is typically attributed to the smaller kinetic diameter of CO2 compared to CH4

and the ability of CO2 to dissolve in the coal polymer structure (e.g. (Charrière et al., 2010)).

Clarkson and Bustin, 1999 performed CH4 (dry and moist) and CO2 (dry) sorption kinetic

measurements on different bituminous coals from the Gates formation in Canada. They

found that diffusivities are generally higher for CO2 than for CH4 (for dry coals). The CH4

diffusivities on dry coals were faster than on the corresponding moist coals. Authors used a

unipore and a bidisperse sorption kinetic model and found that for bright coals (vitrinite

rich) with a uniform pore size distribution the unipore model was sufficient. The bidisperse

model was found to better represent the observed gas uptake rates of dull or banded coals

(inertinite rich) with a more complicated pore structure. Further, authors found that the

applicability of either the unipore or the bidisperse model depends on gas pressure.

Busch and his co-workers 2004 performed CO2 and CH4 sorption kinetic experiments on

h.v.b. coal from the Upper Silesian Basin, Poland at 32 and 45°C. They reported faster

sorption rates for CO2 than for CH4 and decreasing sorption rates with increasing grain size

(<0.063 mm to 3 mm) while rates did not change significantly at grain sizes >100 m. This

indicates transport limiting grain sizes at a scale of ~100 m, i.e. above this size small

transport pathways in the form of cleats or cracks can trigger fast gas distribtution around

the matrix blocks. Further, increasing rates for increasing temperatures and decreasing

rates with increasing moisture contents have been reported. These findings are in

agreement with a more recent study by Gruszkiewicz et al., 2009. Here similar sorption

rates for the 5-10 mm compared to the 1-2 mm grain size fraction have been reported for

either gas on h.v.b. coal from the Black Warrior Basin, USA. As in the study by (Busch et al.,

2004) rates increased with temperature and decreased with moisture.

In another similar study, Li and his co-workers found 2010 for three dry Chinese coals,

ranging in rank from subbituminous to anthracite, faster sorption rates for CO2 compared

to CH4 for measurements on most samples. No distinct differences in rates were observed

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

71 | P a g e

for the three temperatures chosen (35, 45, 55°C). Rank was concluded to have a larger

effect on sorption rates. Authors noted that for both gases, CO2 and CH4, rank has a more

profound influence on sorption rates than temperature although no clear correlation was

found as a function of coal rank (i.e. (Li et al., 2010; Seewald, 1986)).

For a moisture-equilibrated h.v.b. coal, Cui, 2004 found decreasing sorption rates in the

order CO2>CH4>N2 at 30°C while the difference between CH4 and N2 is small. Authors relate

these changes to the differences in kinetic diameter of the gas species (CO2<N2≈CH4;

3.3Å<3.64Å<3.8Å, e.g. (Shieh, 1999)). The determination method of these kinetic diameters

however, could not be found in the literature and therefore these values should be used

with care.

Pone et al. 2009 studied sorption rates under confined and unconfined conditions using

powdered and solid coal for the unconfined measurements and different confining stresses

(6.9 and 13.8 MPa) for confined experiments on coal plugs. For both gases, CO2 and CH4

they found reduced diffusivities (factor 1.5) when increasing the sample size in the

unconfined experiments and an even higher reduction in diffusivity when applying stress

on the sample. The latter reduction was found to be proportional to the applied stress.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

72 | P a g e

Author Gas Coal Basin Coal

rank

Method Grain size

(mm)

dry/

moist

T

(°C)

P

(MPa)

Model D (m2/s)

CO2 CH4

Gruszkie-

wicz

CO2, CH4,

CO2/CH4

Black

Warrior

Basin, USA

h.v.b. Mano-

metric

0.045-0.15;

1-2; 5-10

dry,

moist

35,

40

1.4-6.9 graphical

(P vs t)

n.a. n.a.

Busch CO2, CH4 Upper

Silesian

Basin,

Poland

h.v.b. Mano-

metric

<0.063;

0.063-0.177;

0.177-0.354;

0.354-0.707;

0.707-2; ~3

dry,

moist

32,

45

CO2 <6;

CH4<10

unipore,

dual

matrix

porosity

10-11 n.a.

Charriere CO2, CH4 Loraine

Basin,

France

l.v.b. Gravi-

metric

0.5-1 dry 10-

60

<5 unipore 10-12 -

10-13

10-13

Li CO2, CH4 Inner

Mongolia,

Henan,

Shanxi,

China

subbit

umino

us,

m.v.b.,

anthra

cite

Mano-

metric

0.354-1 dry 35,

45,

55

<5 graphical

(P vs t)

n.a. n.a.

Pone CO2, CH4 Western

Kentucky

Coalfield,

USA

h.v.b. Mano-

metric

<0.25;

cylindrical: d

2.5, l 6.2

moist 20 3.1 unipore 10-18 10-17

Pan CO2, CH4 Sydney

Basin, Aus

m.v.b. flow

exp-

eriment

cylindrical: d

=25.4;

l =82.6

dry,

moist

26 <4 unipore n.a.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

73 | P a g e

D (m2/s)

Author Gas Coal

Basin

Coal rank Method Grain size

(mm)

dry/

moist

T (°C) P

(MPa)

Model CO2 CH4

Marecka CO2, CH4 n.a. semi-

anthracite

Vol-

umetric

<0.032;

0.30–0.10;

0.75–0.49;

1.50–1.00

n.a. 20, 30 n.a. unipore n.a. 10-12 –

10-13

Ciembro-

niewicz

CO2 Polish anthracite Mano-

metric

0.49-0.75 n.a. 16-35 <0.06

5

unipore 10-12 –

10-13

n.a.

Siemons CO2 Great

Britain

h.v.b.;

semi-

anthracite

manome

tric

0.04-0.06;

0.06-0.18;

0.35-0.71;

0.71-2.0

dry,

moist

45 <12 Bidis-

perse

10-8 –

10-12

n.a.

Clarkson CO2, CH4 Lower

Cretaceo

us Gates

Formatio

n, Canada

m.v.b. Vol-

umetric

<0.25;

<4.76

CH4 dry,

moist;

CO2 dry

30 <2 unipore,

bidis-

perse

10-11 –

10-13

10-12 –

10-15

Cui CO2, CH4,

N2

- h.v.b. Mano-

metric

<0.25 moist 30 <7 Bidis-

perse

10-11 –

10-14

10-14 –

10-15

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

74 | P a g e

D (m2/s)

Author Gas Coal

Basin

Coal rank Method Grain size

(mm)

dry/

moist

T (°C) P

(MPa)

Model CO2 CH4

Seewald CH4 Ruhr

Basin,

Germany

h.v.b.;

m.v.b.;

anthracite

Vol-

umetric

0.04 to 1 in

9 fractions

dry,

moist

0, 15, 50 <0.1 unipore n.a. 10-12 –

10-14

Kelemen CO2, CH4 Argonne

Coals, US

h.v.b.,

l.v.b.

Gravi-

metric

~7mm3 dry 30, 75 1.8 unipore 10-13 –

10-16

10-14 –

10-15

Table 3. Sorption kinetic experiments performed by various authors ((Busch et al., 2004; Charrière et al., 2010; Ciembroniewicz

and Marecka, 1993; Clarkson and Bustin, 1999a; Clarkson and Bustin, 1999b; Cui, 2004; Gruszkiewicz et al., 2009; Kelemen

and Kwiatek, 2009; Li et al., 2010; Marecka, 1998; Pan et al., 2010; Pone et al., 2009; Seewald, 1986; Siemons et al., 2007)).

h.v.b, m.v.b. and l.v.b are high, medium and low volatile bituminous coals. Diffusion coefficients derived from unipore or

bidisperse models in previous subchapters.

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

75 | P a g e

Summary of sorption kinetic experiments with CO2 and CH4

Unipore and bidisperse gas transport models have been used to interpret and quantify

observed gas uptake rates. Some are rather simple and some others more sophisticated.

Although not generally confirmed, unipore models seem to better represent the sorption

kinetics of high rank coals (m.v.b. to anthracite) while bidisperse models better apply to

low mature coals. This is in line with the observation that micropores increase and

mesopores decrease with coal rank.

Diffusion coefficients for CO2 range from 10-11 to-16 m2 s-1 and for CH4 from 10-12 to

10-15 m2 s-1 (Table 3). Pone et al. 2009 found lower values of 10-17 and 10-18 m2 s-1 for

CO2 and CH4, respectively.

Gas sorption rates in dry coals are faster than in moist coals under all experimental

conditions.

Gas sorption rates increase with increasing temperature but seem to decrease with

increasing surface coverage.

In general, CO2 sorption rates are faster than for CH4 on the same sample. One

exception was found for a subbituminous coal (Li et al., 2010).

CO2 and CH4 sorption rates decrease with increasing grain size up to a certain limit.

Above this limit, transport through micro-cleats in the coal matrix seem to become

dominant.

Sorption rates under confining stress conditions are slower than under unconfined

conditions. Although this was only reported by one author (Pone et al., 2009) it seems

to be valid since it is a common finding for all kind of natural rocks that gas transport

decreases with increasing effective stress. Experience from coal mining supports these

observations: massive degassing of deep, stressed coal seams occurs only when the

stress field is strongly disturbed due to mining operations.

Conclusions

This review summarizes the present state of knowledge on the interaction of CO2, CH4 and

water with the coal matrix in the context of coalbed methane (CBM) and CO2-ECBM

recovery. We aimed at summarizing the various qualitative and quantitative findings

related to CO2, CH4 and water sorption on coal as well as related sorption kinetic data.

Further we provided insights into experimental details of recording sorption isotherms and

uncertainties related therewith.

In the following, various aspects that still need to be researched for a better understanding

of gas production from coal or gas storage into coal have been defined:

Gensterblum PhD-Thesis: CBM and CO2-ECBM related sorption processes in coal

76 | P a g e

Gas sorption under reservoir stress conditions. Usually, in conventional sorption

experiments no external stress is applied to the coal sample which only represents

subsurface conditions in terms of temperature and gas pressure. Experimental effective

stress acting on the coal sample changes sorption properties both in terms of sorption

amount and in terms of time to reach equilibrium between the sorbate and the sorbent.

Gas sorption from binary mixtures. Most literature sources report single gas CO2 and CH4

on one coal sample while the amount of either gas sorbed at a certain pressure is calculated

using the extended Langmuir approach. Indications from few authors exist that gas mixture

isotherms may provide different results.

Matrix CO2/CH4 counter diffusion. The timing of gas exchange in the coal matrix for CO2-

ECBM is determined by CO2 sorption rates and CH4 desorption rates. These processes

happen simultaneously and very limited data is available for recognizing the significance of

counter diffusion for CO2-ECBM.

Calculation of absolute from excess sorption isotherms. This calculation depends on the

sorbed gas density, which cannot be determined directly. Therefore, literature values like

the density at the liquid boiling point, the van der Waals density or a graphical method is

used. Although absolute CH4 sorption does not differ significantly from excess sorption

there are large discrepancies when applying this method to CO2 sorption isotherms using

different CO2 sorbed phase density values especially at elevated pressures (above e.g. 10

MPa).

Acknowledgements

Authors would like to thank Shell International Exploration and Production B.V. for getting

the opportunity to publish this article. Further we are grateful to Bernd Krooss and Rick

Wentinck for critical review and useful discussions on the content of this paper.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

77 | P a g e

References

Alcaniz-Mongue, J., Linares-Solano, A., Rand, B., 2001. Mechanism of Adsorption of Water

in Carbon Micropores As Revealed by a Study or Activated Carbon Fibers. J. Phys.

Chem., 106(No. 12): 3209 - 3216.

Allardice, D.J., Clemow, L.M., Favas, G., Jackson, W.R., Marshall, M. and Sakurovs, R., 2003.

The characterisation of different forms of water in low rank coals and some

hydrothermally dried products[small star, filled]. Fuel, 82(6): 661-667.

Allardice, D.J. and Evans, D.G., 1971. The-brown coal/water system: Part 2. Water

sorption isotherms on bed-moist Yallourn brown coal. Fuel, 50(3): 236-253.

Amarasekera, G., Scarlett, M.J. and Mainwaring, D.E., 1995. Micropore size distributions

and specific interactions in coals. Fuel, 74(1): 115-118.

ASTM, D1412-93. Standard Test Method of eqilibrium moisture of Coal at 96 to 97

percent relative Humidity and 30°C.

Avgul, N.N. and Kiselev, A.V., 1970. Chemistry and Physics of Carbon, 6, New York.

Bae, J.S. and Bhatia, S.K., 2006. High-Pressure Adsorption of Methane and Carbon

Dioxide on Coal. Energy & Fuels, 20(6): 2599-2607.

Battistutta, E., van Hemert, P., Lutynski, M., Bruining, H. and Wolf, K.-H., 2010. Swelling

and sorption experiments on methane, nitrogen and carbon dioxide on dry Selar

Cornish coal. International Journal of Coal Geology, In Press, Accepted Manuscript.

Beamish, B.B. and Crosdale, P.J., 1998. Instantaneous outbursts in underground coal

mines: An overview and association with coal type. International Journal of Coal

Geology, 35(27-55).

Bhatia, S.K., 1987. Modelling the pore structure of coal. AICHE Journal, 33(10): 1707-

1718.

Bratek, K., Bratek, W., Gerus-Piasecka, I., Jasienko, S. and Wilk, P., 2002. Properties and

structure of different rank anthracites. Fuel, 81: 97-108.

Brennan, J.K., Bandosz, T.J., Thomson, K.T. and Gubbins, K.E., 2001. Water in porous

carbons. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 187-

188: 539-568.

Brunauer, S., Emmett, P.H. and Teller, E., 1938. Adsorption of gases in multimolecular

layers. Journal of the American Chemical Society 60: 309-319.

Busch, A., Alles, S., Gensterblum, Y., Prinz, D., Dewhurst, D.N., Raven, M.D., Stanjek, H. and

Krooss, B.M., 2008. Carbon dioxide storage potential of shales. International

Journal of Greenhouse Gas Control, 2(3): 297-308.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

78 | P a g e

Busch, A., Gensterblum, Y. and Krooss, B.M., 2003a. Methane and CO2 sorption and

desorption measurements on dry Argonne Premium Coals: Pure components and

mixtures. International Journal of Coal Geology, 55: 205-224.

Busch, A., Gensterblum, Y. and Krooss, B.M., 2007. High-Pressure Sorption of Nitrogen,

Carbon Dioxide, and their Mixtures on Argonne Premium Coals. Energy and Fuels,

21: 1640-1645.

Busch, A., Gensterblum, Y., Krooss, B.M. and Littke, R., 2004. Methane and carbon dioxide

adsorption-diffusion experiments on coal: upscaling and modeling. International

Journal of Coal Geology, 60(2-4): 151-168.

Busch, A., Gensterblum, Y., Krooss, B.M. and Siemons, N., 2006. Investigation of high-

pressure selective adsorption/desorption behaviour of CO2 and CH4 on coals: An

experimental study. International Journal of Coal Geology, 66(1-2): 53-68.

Busch, A., Gensterblum, Y., Siemons, N., Krooss, B.M., van Bergen, F., Pagnier, H.J.M. and

David, P., 2003b. Investigation of preferential sorption behaviour of CO2 and CH4

on coals by high-pressure adsorption/desorption experiments with gas mixtures,

Proceedings of the 2003 Coalbed Methane Symposium, Tuscalosa, pp. 14.

Busch, A., Krooss, B.M., Gensterblum, Y., van Bergen, F. and Pagnier, H.J.M., 2003c. High-

pressure adsorption of methane, carbon dioxide and their mixtures on coals with a

special focus on the preferential sorption behaviour. Presented at Geofluids IV,

May 12-16, Utrecht., J. Geochem. Expl., 78-79: 671-674.

Bustin, R.M., 2004. Acid Gas Sorption by British Columbia Coals: Implications for

Permanent Disposal of Acid Gas in Deep Coal Seams and Possible Co-Production of

Methane, University of British Columbia, Vancouver.

Bustin, R.M. and Clarkson, C.R., 1998. Geological Controls on coalbed methane reservoir

capacity and gas content. International Journal of Coal Geology, 38: 3-26.

Carroll, R.E. and Pashin, J.C., 2003. Relationship of sorption capacity to coal quality: CO2

sequestration potential of coalbed methane reservoirs in the Black Warrior Basin,

International Coalbed Methane Symposium, Tuscaloosa, Al.

Carslaw, H.S. and Jaeger, J.C., 1959. Conduction of heat in solids Oxford University Press,

Oxford

Chaback, J.J., Morgan, W.D., Yee, D., 1996. Sorption of nitrogen, methane, carbon dioxide

and their mixtures on bituminous coals at in-situ conditions. Fluid Phase

Equilibria, 117: 289-296.

Chalmers, G.R.L. and Bustin, M.R., 2007. On the effects of petrographic composition on

coalbed methane sorption. International Journal of Coal Geology, 69(4): 288-304.

Charrière, D. and Behra, P., 2010. Water sorption on coals. Journal of Colloid and

Interface Science, 344(2): 460-467.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

79 | P a g e

Charrière, D., Pokryszka, Z. and Behra, P., 2010. Effect of pressure and temperature on

diffusion of CO2 and CH4 into coal from the Lorraine basin (France). International

Journal of Coal Geology, 81: 373–380.

Ciembroniewicz, A. and Marecka, A., 1993. Kinetics of CO2 sorption for two Polish hard

coals. Fuel, 72(3): 405-408.

Clarkson, C.R. and Bustin, R.M., 1999a. The effect of pore structure and gas pressure

upon the transport properties of coal: a laboratory and modeling study. 1.

Isotherms and pore volume distributions. Fuel, 78(11): 1333-1344.

Clarkson, C.R. and Bustin, R.M., 1999b. The effect of pore structure and gas pressure

upon the transport properties of coal: a laboratory and modeling study. 2.

Adsorption rate modeling. Fuel, 78(11): 1345-1362.

Clarkson, C.R. and Bustin, R.M., 2000a. Binary gas adsorption/desorption isotherms:

effect of moisture and coal composition upon carbon dioxide selectivity over

methane. International Journal of Coal Geology, 42: 241-271.

Clarkson, C.R. and Bustin, R.M., 2000b. Binary gas adsorption/desorption isotherms:

effect of moisture and coal composition upon carbon dioxide selectivity over

methane. International Journal of Coal Geology, 42(4): 241-271.

Coppens, L., 1936. Ann. Mines Belg., 37: 173.

Crank, J., 1975. The Mathematics of Diffusion. Oxford University Press, 414 pp.

Crosdale, P.J., 1999. Mixed Methane/carbon Dioxide sorption by coal: New evidence in

support of pore-filling models. Int. CBM Symposium, Univ. of Alabama, Tuscaloosa:

359-366.

Crosdale, P.J., Beamish, B.B. and Valix, M., 1998. Coalbed methane sorption related to

coal composition. International Journal of Coal Geology, 35(1-4): 147-158.

Crosdale, P.J., Moore, T.A. and Mares, T.E., 2008. Influence of moisture content and

temperature on methane adsorption isotherm analysis for coals from a low-rank,

biogenically-sourced gas reservoir. International Journal of Coal Geology, 76(1-2):

166-174.

Cui, X., Bustin, R.M. and Dipple, G., 2004. Selective transport of CO2, CH4, and N2 in coals:

insights from modeling of experimental gas adsorption data. Fuel, 83: 293-303.

Darcey, J.R., Clunie, J.C. and Thomas, D.G., 1958. The adsorption of water by Saran

charcoal. Transactions of the Faraday Society, 54: 250-256.

Day, S., Duffy, G., Sakurovs, R. and Weir, S., 2008a. Effect of coal properties on CO2

sorption capacity under supercritical conditions. International Journal of

Greenhouse Gas Control, 2(3): 342-352.

Day, S., Fry, R. and Sakurovs, R., 2008b. Swelling of Australian coals in supercritical CO2.

International Journal of Coal Geology, 74(1): 41-52.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

80 | P a g e

Day, S., Sakurovs, R. and Weir, S., 2008c. Supercritical gas sorption on moist coals.

International Journal of Coal Geology, 74(3-4): 203-214.

Day, S.J., Duffy, G.J., Saghafi, A., Sakurovs, R., Rubin, E.S., Keith, D.W., Gilboy, C.F., Wilson,

M., Morris, T., Gale, J. and Thambimuthu, K., 2005. Factors controlling CO2

absorption in Australian coals, Greenhouse Gas Control Technologies 7. Elsevier

Science Ltd, Oxford, pp. 2221-2224.

De Weireld, G., Frere, M. and Jadot, R., 1999. Automated determination of high-

temperature and high-pressure gas adsorption isotherms using a magnetic

suspension balance. Meas. Sci. Technol., 10: 117–126.

DeGance, A.E., Morgan, W. D., Yee, D., 1993. High Pressure Adsorption of Methane,

Nitrogen and Carbon Dioxide on Coal Substrates. Fluid Phase Equilibria, 82: 215-

224.

Del Bene, J. and Pople, J.A., 1969. Intermolecular energies of small water polymers.

Chemical Physics Letters, 4(7): 426-428.

Dreisbach, F., Staudt, R. & Keller, J.U., 1999. High pressure adsorption data of methane,

nitrogen, carbon dioxide and their binary and ternary mixtures on activated

carbon. Adsorption, 5: 215-227.

Dubinin, M.M., 1980. Water vapor adsorption and the microporous structures of

carbonaceous adsorbents. Carbon, 18(5): 355-364.

Dubinin, M.M. and Serpinsky, V.V., 1981. Isotherm equation for water vapor adsorption

by microporous carbonaceous adsorbents. Carbon, 19(5): 402-403.

Dutta, P., Bhowmik, S. and Das, S., 2011. Methane and carbon dioxide sorption on a set of

coals from India. International Journal of Coal Geology, 85(3-4): 289-299.

Erickson, D. and Jensen, J.R., 2001. CO2 sequestration in an unminable coalbed- San Juan

Basin, Colorado, U.S.A. In: D. Williams, B. Durie, P. McMullan, C. Paulson and A.

Smith (Editors), Proceedings of the Fifth International Conference on Greenhouse

Control Technologies, CSIRO Publishing, Cairns, Australia, pp. 589–592.

Faiz, M., Saghafi, A., Sherwood, N. and Wang, I., 2007. The influence of petrological

properties and burial history on coal seam methane reservoir characterisation,

Sydney Basin, Australia. International Journal of Coal Geology, 70(1-3): 193-208.

Fei, Y., Marshall, M., Jackson, W.R., Chaffee, A.L. and Allardice, D.J., 2006. A comparison of

adsorption isotherms using different techniques for a range of raw, water- and

acid-washed lignites. Fuel, 85(10-11): 1559-1565.

Fei, Y., Zhang, C., Marshall, M., Jackson, W.R., Chaffee, A.L. and Allardice, D.J., 2007. The

effect of cation content of some raw and ion-exchanged Victorian lignites on their

equilibrium moisture content and surface area. Fuel, 86(17-18): 2890-2897.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

81 | P a g e

Fitzgerald, J.E., Pan, Z., Sudibandriyo, M., Robinson, J.R.L., Gasem, K.A.M. and Reeves, S.,

2005. Adsorption of methane, nitrogen, carbon dioxide and their mixtures on wet

Tiffany coal. Fuel, 84(18): 2351-2363.

Fitzgerald, J.E., Robinson Jr., R.L. and Gasem, K.A.M., 2006. Modeling high-pressure

adsorption of gas mixtures on activated carbon and coal using a simplified local-

density model. Languir, 22: 9610-9618.

Gamson, P., Beamish, B. and Johnson, D., 1996. Coal microstructure and secondary

mineralization: their effect on methane recovery. Geological Society, London,

Special Publications, 109(1): 165-179.

Gaschnitz, R., 2000. Gasgenese und Gasspeicherung im flözführenden Oberkarbon des

Ruhr-Beckens (in German), RWTH-Aachen University, Aachen.

Gasem, K.A.M., Robinson, J., L. and Reeves, S.R., 2002a. Adsorption of pure methane,

nitroge and carbon dioxide and their mixtures on San Juan Basin coals. Topical

Report.

Gasem, K.A.M., Sudibandriyo, M., Fitzgerald, J.E., Pan, Z. and Robinson Jr., R.L., 2002b.

Measurement and modelling of gas adsorption on selected coals. Proceesings of the

AICHE Spring National Meeting, New Orleans, Lousiana, March 10-16, 2002.

Gauden, P.A., 2005. Does the Dubinin-Serpinsky theory adequately describe water

adsorption on adsorbents with high-energy centers? Journal of Colloid and

Interface Science, 282(2): 249-260.

Gensterblum, Y., van Hemert, P., Billemont, P., Battistutta, E., Busch, A., Krooss, B.M., De

Weireld, G. and Wolf, K.-H.A.A., 2010. European inter-laboratory comparison of

high pressure CO2 sorption isotherms. II: natural coals. International Journal of

Coal Geology, accepted.

Gensterblum, Y., van Hemert, P., Billemont, P., Busch, A., Charriére, D., Li, D., Krooss, B.M.,

de Weireld, G., Prinz, D. and Wolf, K.H.A.A., 2009. European inter-laboratory

comparison of high pressure CO2 sorption isotherms. I: Activated carbon. Carbon,

47: 2958-2969.

Given, P.H., Marzec, A., Barton, W.A., Lynch, L.J., Gerstein, B.C., 1986. The concept of a

mobile or molecular phase within the macromolecular network of coals: A debate.

Fuel, 65: 155-163.

Goodman, A.L., Busch, A., Bustin, R.M., Chikatamarla, L., Day, S., Duffy, G.J., Fitzgerald, J.E.,

Gasem, K.A.M., Gensterblum, Y., Hartman, C., Jing, C., Krooss, B.M., Mohammed, S.,

Pratt, T., Robinson Jr, R.L., Romanov, V., Sakurovs, R., Schroeder, K. and White, C.M.,

2007. Inter-laboratory comparison II: CO2 isotherms measured on moisture-

equilibrated Argonne premium coals at 55 [degree sign]C and up to 15 MPa.

International Journal of Coal Geology, 72(3-4): 153-164.

Goodman, A.L., Busch, A., Duffy, G., Fitzgerald, J.E., Gasem, K.A.M., Gensterblum, Y.,

Krooss, B.M., Levy, J., Ozdemir, E., Pan, Z., Robinson, J., R.L., Schroeder, K.,

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

82 | P a g e

Sudibandriyo, M. and White, C., 2004. An Inter-laboratory Comparison of CO2

Isotherms Measured on Argonne Premium Coal Samples. Energy and Fuel, 18:

1175-1182.

Goodman, A.L., Favors, R.N., Hill, M.M. and Larsen, J.W., 2005. Structure Changes in

Pittsburgh No. 8 Coal Caused by Sorption of CO2 Gas. Energy & Fuels, 19(4): 1759-

1760.

Gruszkiewicz, M.S., Naney, M.T., Blencoe, J.G., Cole, D.R., Pashin, J.C. and Carroll, R.E.,

2009. Adsorption kinetics of CO2, CH4, and their equimolar mixture on coal from

the Black Warrior Basin, West-Central Alabama. International Journal of Coal

Geology, 77(1-2): 23-33.

Gumma, S. and Talu, O., 2003. Gibbs Dividing Surface and Helium Adsorption.

Adsorption, 9(1): 17-28.

Gunter, W.D., Mavor, M.J. and Robinson, J.R., 2004. CO2 storage and enhanced methane

production: Field testing at Fenn-Big Valley, lberta, Canada, with application, 7th

International Conference on Greenhouse Gas Control Technology, Vancouver,

Canada.

Gutierrez-Rodriguez, J.A., Purcell, R.J. Jr.,Aplan F.F.,, 1984. Estimating the Hydrophobicity

of coal. Colloids and Surfaces, 12: 1-25.

Hall, F.E., Zhou, C., Gasem, K.A.M. and Robinson Jr., R.L., 1994. Adsorption of pure

methane, nitrogen and carbon dioxide and their binary mixtures on wet Fruitland

coals. SPE, 29194: 329-344.

Hanzawa, Y., Kaneko, K., Yoshizawa, N., Pekala, R.W., Dresselhaus, M.S., 1998. The pore

structure determination of carbon aerogels. Adsorption, 4: 187-195.

Harpalani, S. and Chen, G., 1997. Influence of gas production induced volumetric strain

on permeability of coal. Geotechnical and Geological Engineering, 15: 303-325.

Harpalani, S., Prusty, B.K. and Dutta, P., 2006. Methane/CO2 Sorption Modeling for

Coalbed Methane Production and CO2 Sequestration. Energy and Fuels, 20: 1591-

1599.

Hartman, C. and Pratt, T.J., 2005. A Preliminary Study of the Effect of Moisture on the

Carbon Dioxide Storage Capacity in Coal, paper 0533, Proceedings 2005

International Coalbed Methane Symposium, University of Alabama, Tuscaloosa, pp.

9.

Hemza, P., Sivek, M. and Jirásek, J., 2009. Factors influencing the methane content of coal

beds of the Czech part of the Upper Silesian Coal Basin, Czech Republic.

International Journal of Coal Geology, 79(1-2): 29-39.

Hildenbrand, A., Krooss, B.M., Busch, A. and Gaschnitz, R., 2006. Evolution of methane

sorption capacity of coal seams as a function of burial history -- a case study from

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

83 | P a g e

the Campine Basin, NE Belgium. International Journal of Coal Geology, 66(3): 179-

203.

Hirsch, P.B., 1954. X-Ray Scattering from Coals. Proceedings of the Royal Society of

London. Series A. Mathematical and Physical Sciences, 226(1165): 143-169.

Hol, S., Peach, C.J. and Spiers, C.J., 2010. A new experimental method to determine the

CO2 sorption capacity of coal, GHGT 10. Energy Procedia, Amsterdam, the

Netherlands.

Hol, S., Peach, C.J. and Spiers, C.J., 2011. A new experimental method to determine the

CO2 sorption capacity of coal. Energy Procedia, 4: 3125-3130.

Horikawa, T., Sekida, T., Hayashi, J.i., Katoh, M. and Do, D.D., 2010. A new adsorption-

desorption model for water adsorption in porous carbons. Carbon, 49(2): 416-424.

Huang, H., Wang, K., Bodily, D.M. and Hucka, V.J., 1995. Density Measurements of

Argonne Premium Coal Samples. Energy & Fuels, 9(1): 20-24.

Humayun, R. and Tomasko, D.L., 2000. High-Resolution Adsorption Isotherms of

Supercritical Carbon Dioxide on Activated Carbon. AlCHE Journal, 46(10): 2065-

2075.

Iiyama, T., Nishikawa, K., Suszuki, T., Kaneko, K., 1997. Study of the structure of a water

molecular assembly in a hydrophobic nanospace at low temperature with in situ X-

ray diffraction. Chemical Physics Letters, 274: 152-158.

IUPAC, 2001. Manual of symbols and terminology for physicochemical quantities and

units.

Joubert, J.I., Clifford, T. and Bienstock, D., 1973. Sorption of Methane in moist coal. Fuel,

52: 181-185.

Joubert, J.I., Grein, T. and Bienstock, D., 1974. Effect of moisture on the methane capacity

of American coals. Fuel, 53: 186-191.

Karacan, C.Ö., 2003. Heterogeneous Sorption and Swelling in a Confined and Stressed

Coal during CO2 Injection. Energy and Fuel, 17: 1595-1608.

Karacan, C.Ö., 2007. Swelling-induced volumetric strains internal to a stressed coal

associated with CO2 sorption. International Journal of Coal Geology.

Karge, H., Weitkamp, J. and Ruthven, D., 2008. Fundamentals of Adsorption Equilibrium

and Kinetics in Microporous Solids, Adsorption and Diffusion. Molecular Sieves -

Science and Technology. Springer Berlin / Heidelberg, pp. 1-43.

Kedzior, S., 2009. Accumulation of coal-bed methane in the south-west part of the Upper

Silesian Coal Basin (southern Poland). International Journal of Coal Geology, 80(1):

20-34.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

84 | P a g e

Kelemen, S.R. and Kwiatek, L.M., 2009. Physical properties of selected block Argonne

Premium bituminous coal related to CO2, CH4, and N2 adsorption. International

Journal of Coal Geology, 77(1-2): 2-9.

Kelemen, S.R., Kwiatek, L.M., Lee, A.G.K., 2006. Swelling and Sorption Response of

Selected Argonne Premium Bituminous Coals to CO2, CH4, and N2, International

CBM Symposium, Tuscaloosa, Alabama.

Krooss, B.M., van Bergen, F., Gensterblum, Y., Siemons, N., Pagnier, H.J.M. and David, P.,

2002. High-pressure methane and carbon dioxide adsorption on dry and moisture-

equilibrated Pennsylvanian coals. International Journal of Coal Geology, 51: 69-92.

Kunz, O.K., R., Wagner, W. and Jaeschke, M., 2007. The GERG-2004 wide-range reference

equation of state for natural gases. GERG Technical Monograph 15 Fortschritts-

Berichte Verein Deutscher Ingenieue (VDI), 6. VDI-Verlag, Düsseldorf.

Lama, R.D., Bodziony, J., 1998. Management of outburst in underground coal mines. Int J.

Coal Geology, 35: 83-115.

Lamberson, M.N. and Bustin, R.M., 1993. Coalbed Methane characteristics of Gates

formation coals, Northeastern British Columbia: Effect of maceral composition.

AAPG Bulletin, 77(12): 2062-2076.

Laxminarayana, C. and Crosdale, P.J., 1999a. Role of coal type and rank on methane

sorption characteristics of Bowen Basin, Australia coals. International Journal of

Coal Geology, 40(4): 309-325.

Laxminarayana, C. and Crosdale, P.J., 1999b. Role of coal type and rank on methane

sorption charakteristics of Bowen Basin, Australia coals. Int. J. Coal Geology, 40:

309-325.

Laxminarayana, C. and Crosdale, P.J., 2002. Controls on methane sorption capacity of

Indian coals. AAPG Bulletin, 86(2): 201-212.

Levy, J.H., Day, S.J. and Killingley, J.S., 1997. Methane capacities of Bowen Basin coals

related to coal properties. Fuel, 76(9): 813-819.

Li, D., Liu, Q., Weniger, P., Gensterblum, Y., Busch, A. and Krooss, B.M., 2010. High-

pressure sorption isotherms and sorption kinetics of CH4 and CO2 on coals. Fuel,

89(3): 569-580.

Lu, L., Sahajwalla, V., Kong, C. and Harris, D., 2001. Quantitative X-ray diffraction analysis

and its application to various coals. Carbon, 39: 1821–1833.

Lynch, L.J. and Webster, D.S., 1982. Effect of thermal treatment on the interaction of

brown coal and water: a nuclear magnetic resonance study. Fuel, 61(3): 271-275.

Maddox, M., Ulberg, D. and Gubbins, K.E., 1995. Molecular simulation of simple fluids and

water in porous carbons. Fluid Phase Equilibria, 104: 145-158.

Mahajan, O.P., 1991. CO2 surface area of coals: A 25 year paradox. Carbon, 29: 735-742.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

85 | P a g e

Mahajan, O.P., Walker, P.L. Jr., 1970. Water adsorption on coals. Fuel, 50(3): 308-317.

Majewska, Z., Ceglarska-Stefanska, G., Majewski, S. and Zietek, J., 2009. Binary gas

sorption/desorption experiments on a bituminous coal: Simultaneous

measurements on sorption kinetics, volumetric strain and acoustic emission.

International Journal of Coal Geology, 77(1-2): 90-102.

Marecka, A. and Mianowski, A., 1998. Kinetics of CO2 and CH4 sorption on high rank coal

at ambient temperatures. Fuel, 77(14): 1691-1696.

Mastalerz, M., Gluskoter, H. and Rupp, J., 2004. Carbon dioxide and methane sorption in

high volatile bituminous coals from Indiana, USA. Int J. Coal Geology, 60: 43-55.

Mavor, M.J., Hartman, C. and Pratt, T.J., 2004. Uncertainty in Sorption Isotherm

Measurements, 2004 CBM Symposium, Tuscaloosa, Alabama, pp. 14.

Mavor, M.J.O., L.B. & Pratt, T.J., 1990. Measurement and evaluation of coal sorption

isotherm data. SPE, 20728: 157-170.

Mazumder, S., Siemons, N., Wolf, K.-H., 2005. Differential swelling and permeability

change of coal in response to CO2 injection for ECBM. SPE, 98475.

Mazumder, S. and Wolf, K.H., 2008. Differential swelling and permeability change of coal

in response to CO2 injection for ECBM. International Journal of Coal Geology,

74(2): 123-138.

McCarty, R.D. and Arp, V.D., 1990. A new wide range equation of state for helium.

Advances in Cryogenic Engineering, 35: 1465-1475.

McCutcheon, A.L., Barton, W.A. and Wilson, M.A., 2001. Kinetics of Water

Adsorption/Desorption on Bituminous coals. Energy and Fuel, 15: 1387-1395.

McCutcheon, A.L., Barton, W.A. and Wilson, M.A., 2003. Characterization of Water

Adsorbed on Bituminous Coals. Energy & Fuels, 17(1): 107-112.

Michels, G. and Wouters, H., 1941. Zahlenwerte und Funktionen aus Physik, Chemie,

Astromomie Geophysik und Technik, 2. Springer.

Mohammad, S., Fitzgerald, J., Robinson, R.L. and Gasem, K.A.M., 2009. Experimental

Uncertainties in Volumetric Methods for Measuring Equilibrium Adsorption.

Energy & Fuels, 23(5): 2810-2820.

Moore, T.A., submitted. Coalbed Methane: An introductory review. International Journal

of Coal Geology.

Moore, T.A. and Crosdale, P.J., 2006. The effect of moisture and temperature on

adsorption isotherms of a low-rank coal: Implications for reservoir modeling,

International CBM Symposium, Tuscaloosa, Al.

Mu, R. and Malhotra, V.M., 1991. A new approach to elucidate coal-water interactions by

an in situ transmission FT-i.r. technique. Fuel, 70(10): 1233-1235.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

86 | P a g e

Muller, E.A., Rull, L.F., Vega, L.F. and Gubbins, K.E., 1996. Adsorption of Water on

Activated Carbons: A Molecular Simulation Study. The Journal of Physical

Chemistry, 100(4): 1189-1196.

Murata, K., El-Merraoui, M. and Kaneko, K., 2001. A new determination method of

absolute adsorption isotherm of supercritical gases under high pressure with a

special relevance to density-functional theory study. The Journal of Chemical

Physics, 114(9): 4196-4205.

Nishino, J., 2001. Adsorption of water vapor and carbon dioxide at carboxylic functional

groups on the surface of coal. Fuel, 80: 757-764.

Nodzenski, A., 1998. Sorption and desorption of gases (CH4, CO2) on hard coal and

active carbon at elevated pressures. Fuel, 77(11): 1243-1246.

Ohga, K., Fujioka, M., Yamaguchi, S. and Nako, M., 2006. Field tests of CO2 sequestration

and ECBM in Japan International CBM Symposium, Tuscaloosa, Alabama, paper #

0635.

Ottiger, S., Pini, R., Storti, G. and Mazzotti, M., 2008. Competitive adsorption equilibria of

CO2 and CH4 on a dry coal. Adsorption, 14(4-5): 539-556.

Ottiger, S., Pini, R., Storti, G., Mazzotti, M., Bencini, R., Quattrocchi, F., Sardu, G. and

Deriue, G., 2006. Adsorption of Pure Carbon Dioxide and Methane on Dry Coal from

the Sulcis Coal Province (SW Sardinia, Italy). Environmental Progress, 25: 355-364.

Ozdemir, E., Morsi, B.I. and Schroeder, K., 2003. Importance of volume effects to

adsorption isotherms of carbon dioxide on coals. Langmuir, 19: 9764-9773.

Ozdemir, E., Morsi, B.I. and Schroeder, K., 2004. CO2 adsorption capacity of Argonne

Premium Coals. Fuel, 83: 1085-1094.

Ozdemir, E. and Schroeder, K., 2009. Effect of Moisture on Adsorption Isotherms and

Adsorption Capacities of CO2 on Coals. Energy & Fuels, 23(5): 2821–2831.

Palmer, I. and Mansoori, J., 1998. How Permeability Depends on Stress and Pore

Pressure in Coalbeds: A New Model. SPE Reservoir Evaluation & Engineering, SPE

52607: 539-544.

Pan, Z. and Connell, L.D., 2007. A theoretical model for gas adsorption-induced coal

swelling. International Journal of Coal Geology, 69(4): 243-252.

Pan, Z., Connell, L.D., Camilleri, M. and Connelly, L., 2010. Effects of matrix moisture on

gas diffusion and flow in coal. Fuel, 89(11): 3207-3217.

Peng, D.Y. and Robinson, D.B., 1976. A New Two-Constant Equation of State. Ind. Eng.

Chem. Fundam., 15(1): 59-64.

Pini, R., Ottiger, S., Burlini, L., Storti, G. and Mazzotti, M., 2010. Sorption of carbon

dioxide, methane and nitrogen in dry coals at high pressure and moderate

temperature. International Journal of Greenhouse Gas Control, 4(1): 90-101.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

87 | P a g e

Pini, R., Ottiger, S., Burlini, L., Storti, G., Mazzotti, M., Bencini, R. and Quattrocchi, F.,

2006a. Experimental study of CO2 sequestration by ECBM recovery: the case of

Sulcis coal, GHGT8, Trondheim, Norway.

Pini, R., Ottiger, S., Rajendran, A., Storti, G. and Mazzotti, M., 2006b. Reliable

measurement of near-critical adsorption by gravimetric method. Adsorption,

12(5): 393-403.

Pone, J.D.N., Halleck, P.M. and Mathews, J.P., 2009. Sorption Capacity and Sorption

Kinetic Measurements of CO2 and CH4 in Confined and Unconfined Bituminous

Coal. Energy & Fuels, 23(9): 4688-4695.

Prinz, D., 2004. Die Porenstruktur von Kohlen (in German), RWTH Aachen University,

Germany, Aachen, 184 pp.

Prinz, D. and Littke, R., 2005. Development of the micro- and ultramicroporous structure

of coals with rank as deduced from the accessibility to water. Fuel, 84(12-13):

1655-1662.

Prinz, D., Pyckhout-Hintzen, W., Krooss, B.M. and Littke, R., 2001. Investigations on the

Micro- and Mesoporous Structure of Coals of Varying Rank: A Combined Small

Angle Neutron Scattering (SANS) and Gas Adsorption Experiments Study. Int. CBM

Symposium, Tuscaloosa.

Prinz, D., Pyckhout-Hintzen, W. and Littke, R., 2004. Development of the meso- and

macroporous structure of coals with rank as analyzed with small angle neutron

scattering and adsorption experiments. Fuel, 83: 547-556.

Puri, B.R., 1966. Chemisorbed oxygen evolved as carbon dioxide and its influence on

surface reactivity of carbons. Carbon, 4(3): 391-400.

Reeves, S., Gonzalez, R., Gasem, K.A.M., Fitzgerald, J.E., Pan, Z., Sudibandriyo, M. and

Robinson Jr., R.L., 2005. Measurement and Prediction of Single- and Multi-

Component Methane, Carbon Dioxide and Nitrogen Isotherms for U.S. Coals, ICBM

Symposium Tuscaloosa, Tuscaloosa, Alabama.

Reeves, S., Taillefert, A., Pekot, L. and Clarkson, C., 2003. The Allison Unit CO2 - ECBM

Pilot: A Reservoir Modeling Study, Topical Report, U.S. Department of Energy (DE-

FC26-0NT40924).

Reucroft, P.J., Patel, H., 1986. Gas-induced swelling in coal. Fuel, 65: 816-820.

Rice, D.D., 1993. Composition and origins of coalbed gas. In: B.E. Law, Rice, D.D (Editor),

Hydrocarbons from coal. AAPG studies in Geology, pp. 159-184.

Rubes, M., Kysilka, J., Nachtigall, P. and Bludski, O., 2010. DFT/CC investigation of

physical adsorption on a graphite (0001) surface. Physical Chemistry Chemical

Physics, 12(24): 6438-6444.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

88 | P a g e

Rubes, M., Nachtigall, P., Vondrasek, J. and Bludsky, O., 2009. Structure and Stability of

the Water-Graphite Complexes. The Journal of Physical Chemistry C, 113(19):

8412-8419.

Ruckenstein, E., Vaidyanathan, S. and Youngquist, G.R., 1971. Sorption by solids with

bidisperse pore structure. Chem. Eng. Science, 26: 1305-1318.

Ryan, B. and Richardson, D., 2002. The potential for CO2 sequestration in British

Columbia coal seams, British Columbia Geological Survey.

Saghafi, A., Faiz, M. and Roberts, D., 2007. CO2 storage and gas diffusivity properties of

coals from Sydney Basin, Australia. International Journal of Coal Geology, 70(1-3):

240-254.

Sakurovs, R., Day, S. and Weir, S., 2009. Causes and consequences of errors in

determining sorption capacity of coals for carbon dioxide at high pressure.

International Journal of Coal Geology, 77(1-2): 16-22.

Sakurovs, R., Day, S., Weir, S. and Duffy, G., 2007. Application of a Modified Dubinin-

Radushkevich Equation to Adsorption of Gases by Coals under Supercritical

Conditions. Energy and Fuels, 21: 992-997.

Sakurovs, R., Day, S., Weir, S. and Duffy, G., 2008. Temperature dependence of sorption of

gases by coals and charcoals. International Journal of Coal Geology, 73(3-4): 250-

258.

Salame, I.I. and Bandosz, T.J., 1998. Experimental Study of Water Adsorption on

Activated Carbons. Langmuir, 15(2): 587-593.

Salame, I.I. and Bandosz, T.J., 1999. Study of Water Adsorption on Activated Carbons

with Different Degrees of Surface Oxidation. Journal of Colloid and Interface

Science, 210(2): 367-374.

Salame, I.I. and Bandosz, T.J., 2000. Adsorption of Water and Methanol on Micro- and

Mesoporous Wood-Based Activated Carbons. Langmuir, 16(12): 5435-5440.

Seewald, H. and Klein, J., 1986. Methansorption an Steinkohle und Kennzeichnung der

Porenstruktur. Glückauf-Forschungshefte, 47(3): 149-156.

Setzmann, U. and Wagner, W., 1991. A new equation of state and tables of

thermodynamic properties for methane covering the range from the melting line to

625 K at pressures up to 1000 MPa. Journal of Physical and Chemical Reference

Data, 20: 1061-1155.

Shi, J.-Q. and Durucan, S., 2003a. Gas Storage and Flow in Coalbed Reservoirs:

Implementation of a Bidisperse Pore Model for Gas Diffusion in Coal Matrix. SPE,

84342: 1-9.

Shi, J.Q. and Durucan, S., 2003b. A bidispere pre diffusion model for methane

displacement desorption in coal by CO2 injection. Fuel, 82: 1219-1229.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

89 | P a g e

Shieh, J.-J. and Chunh, T.S., 1999. Gas permeability, diffusivity, and solubility of Poly (4-

Vinylpyridine) film. Journal of Polymer Science, 37: 2851-2861.

Siemons, N. and Busch, A., 2007. Measurement and interpretation of supercritical CO2

sorption on various coals. International Journal of Coal Geology, 69(4): 229-242.

Siemons, N., Wolf, K.-H.A.A. and Bruining, J., 2007. Interpretation of carbon dioxide

diffusion behavior in coals. International Journal of Coal Geology, 72(3-4): 315-

324.

Sircar, S., 1992. Estimation of isosteric heats of adsorption of single gas and

multicomponent gas mixtures. Ind. Eng. Chem. Res., 31: 1813-1819.

Sircar, S., 1999. Gibbsian Surface Excess for Gas Adsorptions Revisited. Ind. Eng. Chem.

Res., 38: 3670-3682.

Smith, D.M. and Williams, F.L., 1984a. Diffusion models for gas production from coal:

Determination of diffusion parameters. Fuel, 63(2): 256-261.

Smith, D.M. and Williams, F.L., 1984b. Diffusion models for gas production from coals-

Application to Methane content determination. Fuel, 63: 251-255.

Smith, D.M. and Williams, F.L., 1984c. Diffusional effects in the recovery of methane from

coalbeds. SPE, 10821: 529-541.

Span, R. and Wagner, W., 1996. A new equation of state for carbon dioxide covering the

fluid region from the triple-point temperature to 1100 K at pressures up to 800

MPa. Journal of Physical and Chemical Reference Data, 25(6): 1509-1596.

Stach, E., Mackowsky, M.-T., Teichmuller, M., Taylor, G., Chandra, D. and Teichmuller, R.,

1982. Stach’s Textbook of Coal Petrology, 3rd edition. Gebruder Borntraeger,

Berlin.

Stoeckli, H.F., 1990. Microporous carbons and their characterization: The present state

of the art. Carbon, 28(1): 1-6.

Suárez, N., Laredo, E. and Nava, R., 1993. Characterization of four hydrophilic sites in

bituminous coal by ionic thermal current measurements. Fuel, 72(1): 13-18.

Sudibandriyo, M., Fitzgerald, J.E., Pan, Z., Robinson, J., R.L. and Gasem, K.A.M., 2002.

Measurements and modeling of gas adsorption on selected coals. Langmuir, 19:

5323-5331.

Sudibandriyo, M., Pan, Z., Fitzgerald, J.E., Robinson Jr., R.L. and Gasem, K.A.M., 2003.

Adsorption of Methane, Nitrogen, Carbon Dioxide, and Their Binary Mixtures on

Dry Activated Carbon at 318.2 K and Pressures up to 13.6 MPa. Langmuir, 19:

5323-5331.

Suuberg, E.M., Otake, Y, Yun, Y & Deevi, S.C., 1993. Role of moisture in coal structure and

the effect of drying upon the accessibility of coal structure. Energy & Fuels, 7: 384-

392.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

90 | P a g e

Ulberg, D.E. and Gubbins, K.E., 1995. Water adsorption in microporous graphitic

carbons. Molecular Physics: An International Journal at the Interface Between

Chemistry and Physics, 84(6): 1139 - 1153.

Unsworth, J.F., Fowler, C.S. and Jones, L.F., 1989. Moisture in coal - 2. Maceral effects on

pore structure. Fuel, 68: 18-26.

van Bergen, F., Pagnier, H. and Krzystolik, P., 2006. Field experiment of enhanced

coalbed methane-CO2 in the Upper Silesian basin of Poland. Environmental

Geoscience, 13: 201–224.

van Bergen, F., Spiers, C., Floor, G. and Bots, P., 2009. Strain development in unconfined

coals exposed to CO2, CH4 and Ar: Effect of moisture. International Journal of Coal

Geology, 77(1-2): 43-53.

van Hemert, P., Bruining, H., Rudolph, E.S.J., Wolf, K.-H.A.A. and Maas, J.G., 2009.

Improved manometric setup for the accurate determination of supercritical carbon

dioxide sorption. Review of Scientific Instruments, 80(3): 035103-11.

van Hemert, P., Rudolph-Floter, S., Wolf, K.-H., A. A. and Bruining, J., 2010. Estimate of

Equation of State Uncertainty for Manometric Sorption Experiments: Case Study

With Helium and Carbon Dioxide. SPE Journal, 15(1): pp. 146-151.

van Hemert, P., Wolf, K.-H.A.A. and Bruining, J., 2007. The Intrinsic Reliability of

Manometric Sorption Apparatus Using Supercritical Carbon Dioxide, SPE Annual

Technical Conference and Exhibition, Anaheim, CA, pp. 11.

van Krevelen, D.W., 1993. Coal: Typology-Physics-Chemistry-Constitution. Elsevier,

1002 pp.

Vandamme, M., Brochard, L., Lecampion, B. and Coussy, O., 2010. Adsorption and strain:

The CO2-induced swelling of coal. Journal of the Mechanics and Physics of Solids,

58(10): 1489-1505.

Wagner, W. and Pruss, A., 1993. International Equations for the Saturation Properties of

Ordinary Water Substance. Revised According to the International Temperature

Scale of 1990. Addendum to J. Phys. Chem. Ref. Data, 893 (1987). Journal of

Physical and Chemical Reference Data, 22(3): 783-787.

Walker Jr., P.L. and Mahajan, O.P., 1993. Pore structure in coals. Energy and Fuels, 7:

559-560.

Walker Jr., P.L., Verma, S.K., Rivera-Utrilla, J., Davis, A., 1988. Densities, porosities and

surface areas of coal macerals as measured by their interaction with gases, vapours

and liquids. Fuel, 67: 1615-1623.

Weishauptová, Z. and Sýkorová, I., 2011. Dependence of carbon dioxide sorption on the

petrographic composition of bituminous coals from the Czech part of the Upper

Silesian Basin, Czech Republic. Fuel, 90(1): 312-323.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

91 | P a g e

Weniger, P., Kalkreuth, W., Busch, A. and Krooss, B.M., 2010. High-pressure methane and

carbon dioxide sorption on coal and shale samples from the Paraná Basin, Brazil.

International Journal of Coal Geology, 84: 190-205.

White, C.M., Smith, D.H., Jones, K.L., Goodman, A.L., Jikich, S.A., LaCount, R.B., DuBose,

S.B., Ozdemir, E., Morsi, B.I. and Schroeder, K.T., 2005. Sequestration of Carbon

Dioxide in Coal with Enhanced Coalbed Methane RecoverysA Review. Energy and

Fuel, 19(3): 659-724.

Wollenweber, J., Alles, S., Busch, A., Krooss, B.M., Stanjek, H. and Littke, R., 2010.

Experimental investigation of the CO2 sealing efficiency of caprocks. International

Journal of Greenhouse Gas Control, 4(2): 231-241.

Wong, S., Law, D., Deng, X., Robinson, J., Kadatz, B., Gunter, W.D., Jianping, Y., Sanli, F. and

Zhiqiang, F., 2007. Enhanced coalbed methane and CO2 storage in anthracitic

coals--Micro-pilot test at South Qinshui, Shanxi, China. International Journal of

Greenhouse Gas Control, 1(2): 215-222.

Yamaguchi, S., Ohga, K., Fujioka, M. and Muto, S., 2004. Prospect of CO2 sequestration in

the Ishikara coal field, Japan, 7th International Conference on Greenhouse Gas

Control Technology, Vancouver, Canada.

Yee, D., Seidle, J.P. and Hanson, W.B., 1993. Gas Sorption on Coal and Measurements of

Gas Content. In: B.E. Law and D.D. Rice (Editors), Hydrocarbons from coal. AAPG

Studies in Geology, pp. 203-218.

Yi, J., Akkutlu, I.Y. and Deutsch, C.V., 2008. Gas Transport in Bidisperse Coal Particles:

Investigation for an Effective Diffusion Coefficient in Coalbeds. Journal of Canadian

Petroleum Technology, 47(10): 20-26.

Yu, H., Guo, W., Cheng, J. and Hu, Q., 2008a. Impact of experimental parameters for

manometric equipment on CO2 isotherms measured: Comment on "Inter-

laboratory comparison II: CO2 isotherms measured on moisture-equilibrated

Argonne premium coals at 55°C and up to 15 MPa" by Goodman et al. (2007).

International Journal of Coal Geology, 74(3-4): 250-258.

Yu, H., Yuan, J., Guo, W., Cheng, J. and Hu, Q., 2008b. A preliminary laboratory experiment

on coalbed methane displacement with carbon dioxide injection. International

Journal of Coal Geology, 73(2): 156-166.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

92 | P a g e

"Everything should be made as simple as possible, but not simpler."

(Albert Einstein)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

93 | P a g e

2. European inter-laboratory

comparison of high-pressure CO2

sorption isotherms.

I: activated carbon

CARBON 47 (2009) 2958 –2969

Y. Gensterblum, D. Prinz and B.M. Krooss

RWTH Aachen University, Institute of Geology and Geochemistry of Petroleum

and Coal, Lochnerstr. 4-20, D-52056 Aachen, Germany

P. van Hemert and K.-H.A.A. Wolf

TU Delft, Department of Geotechnology,Stevinweg 1, NL-2628 CN Delft, The

Netherlands

P. Billemont and G. de Weireld

Faculté Polytechnique de Mons, Service de Thermodynamique et Physique

Mathématique,Rue de Houdain, B-7000 Mons, Belgium

A. Busch

Shell International Exploration and Production B.V., Kessler Park 1, 2288 GS

Rijswijk, The Netherlands

D. Charriére

INERIS, 60550 Verneuil-en-Halatte, Fance

Université de Toulouse, INPT, LCA; ENSIACET, 118 route de Narbonne, F-31077

Toulouse Cedex 4

D. Li

Department of Resources and Earth Science, China University of Mining and

Technology, Beijing, P.R. China

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

94 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

95 | P a g e

Abstract

In order to assess and improve the quality of high-pressure sorption isotherms of

carbon dioxide (CO2) on coals, an inter-laboratory study (“Round Robin”) has been

conducted among four European research laboratories. In a first round of

measurements, excess sorption isotherms were determined on Filtrasorb 400 (F400)

activated carbon at 318 K using the manometric (TU Delft and RWTH Aachen

University) and the gravimetric (FP Mons and INERIS) method up to 16 MPa. The study

shows that CO2 sorption in the supercritical range can be determined accurately with

both gravimetric and manometric equipment but requires thorough optimization of

instrumentation and measuring as well as proper sample preparation procedures. For

the characterization of the activated carbon F400, which we used as benchmark, we

have determined a surface area of 1063 m2g-1, a DR micropore volume of 0.51 cm3g-1.

Additionally, we analysed the elementary near-surface composition by energy

dispersive X-ray spectroscopy (EDX). To characterise the bulk composition of the F400

activated carbon, a proximate and ultimate analysis was performed.

The observed excess sorption maxima around 5 MPa have values around

8.0 mol kg-1, which are consistently higher than the literature data so far (up to 0.8 mol

kg-1).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

96 | P a g e

Introduction

Among the various options considered for geological storage of carbon dioxide (CO2),

the injection of CO2 into deep, unminable coal seams in particular in

combination with the production of coalbed methane (CBM), is considered a niche

technology. The European RECOPOL* project has demonstrated the technical feasibility

of CO2 injection into typical European Carboniferous coal seams. The follow-up project

MOVECBM† started in 2006 to investigate in more detail the fate of injected CO2.

Laboratory experiments conducted by the two groups at the Delft University of

Technology (The Netherlands) within the Dutch CATO‡ project and RWTH Aachen

University (Germany) within RECOPOL1 and the national CO2TRAP§ project provided

important fundamental information on the interaction of natural coals with carbon

dioxide and methane under in-situ conditions.

However, considerable problems in the reproducibility of supercritical CO2 sorption

measurements became evident. Similar problems were encountered by other groups

and have been addressed by two recent inter-laboratory studies [1,2]. The results of

these studies showed, in spite of considerable improvements in accuracy, the quality of

CO2 sorption isotherms does not yet meet the standards required for reliable modeling

and predictions.

In this context the four laboratory research groups at Delft University of Technology (TU

Delft), RWTH Aachen University (RWTH), INERIS and Faculté Polytechnique Mons

Belgium (FP Mons) decided to perform an inter-laboratory study to assess and improve

the quality of sorption data at high pressures for supercritical carbon dioxide. In

contrast to earlier inter-laboratory tests [1,2] this study was set up as an open project

with exchange of information and regular seminars. The main objective was to increase

the overall accuracy of CO2 excess sorption measurements, eliminate pitfalls and sources

of error and develop best practice standards and procedures.

A well-characterized activated carbon sample, Filtrasorb 400 (F400), was selected for

the first series of measurements. This material is homogeneous, readily available and its

chemical composition and micropore structure are similar to those of natural coal.

Furthermore, F400 is resistant to high temperatures. This facilitates the removal of

moisture and attainment of a defined initial condition, one of the main sources of

discrepancies in previous inter-laboratory comparisons. The F400 has been used in

pervious CO2-sorption studies [3-6] so that published reference data were available for

comparison.

Sorption experiments were performed at 318 K up to 16 MPa. These are typical

conditions for coalbeds suitable for CO2 storage with pressures ranging from 6 to 15

MPa temperatures between 300 and 330 K.

The most common procedures to determine excess sorption isotherms of gases are the

manometric [9-12] and the gravimetric method. [4,5,15,16,24,25] These are well

1 http://recopol.nitg.tno.nl/ † www.movecbm.eu ‡ www.co2-cato.nl § www.co2trap.org

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

97 | P a g e

established and known to provide accurate excess sorption isotherms for simple and

well-characterized sorption systems (e.g. methane sorption on activated carbon). Both

methods were used in this study.

Previous inter-laboratory comparisons of carbon dioxide sorption on coal at

high pressures

Two inter-laboratory comparison studies on high-pressure sorption of CO2 on Argonne

Premium coals [1,2] were initiated by the U.S. Department of Energy-National Energy

Technology Laboratory (DOE-NETL). Although experienced research groups were

involved in these inter-laboratory tests, large deviations were observed. The

discrepancies were attributed to varying moisture contents. Goodman et al. concluded

that further studies with well-defined procedures are required to improve

reproducibility. Therefore, in the present study time and temperature for evacuation

have been increased to ensure complete removal of remnant moisture.

The first inter-laboratory study [1] compared the sorption isotherms of CO2 on dry coals

at 295 and 328 K up to a pressure of 7 MPa measured by five laboratories. Five types of

coal, covering a maturity range from 0.25 to 1.68 % vitrinite reflectance, were used. The

preparation procedure involved drying of the samples for 36 hours at 353 K under

vacuum. It was found that excess sorption values for medium- to low-rank coals

deviated by more than 100 %. Sorption isotherms on high-rank coals were considered to

be sufficiently accurate. The discrepancies were attributed to varying residual moisture

contents after drying of the coal samples.

The second inter-laboratory study [2] compared the sorption isotherms of CO2 on

moisture-equilibrated coals at 328K and pressures up to 15 MPa measured by six

laboratories. Moisture-equilibration was achieved by a modified ASTM D 1412-99

procedure [7]. Three types of coal, covering a maturity range from lignite to high volatile

bituminous, were used. Sorption data showed good agreement for pressures up to 8

MPa. However, at higher pressures sorption diverged significantly for the different

laboratories. This deviation was attributed to substantial variations in moisture

contents.

Methods and materials

Sample and sample prepartion

FILTRASORB 400 (F 400) activated carbon of Calgon Carbon Corporation used in this

study was kindly supplied by Chemviron Carbon GmbH, Germany. Aliquots of the same

batch (Batch No.:FE 05707A) were distributed among the participants in order to

exclude heterogeneity effects. Analytical data for the F400 have been previously

published by Fitzgerald et al. [3]

The pore size characterization of F400 batch used in this study is determined by a low

pressure nitrogen adsorption isotherm at 77K (80 data points, 7x10-6 to 0.9965 p/p0),

evaluated using the non local density functional theory (NLDFT). Additionally, the pore

diameter distribution over the complete range and the important range between 0-5 nm

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

98 | P a g e

are within (Figure 28b,c). The prominent peaks are at 0.8 and 1.2 nm. The BET surface is

1063 m2/g, calculated over 10 data points in the relative range between 0.1 and 0.3. The

Dubinin-Radushkevich (DR) micropore volume is 0.51 cm3/g, determined in the relative

pressure range between 4x10-5 and 3x10-2 over 28 data points. Nearly the same pore-

size distribution (PSD) of a different batch of AC F400 was determined by Jagiello and

Thommes [8] using N2 and Argon as sorbents. All measurements in the present study

were performed on dry sample material. The activated carbon was dried at 473K for 24

hours (see below for drying procedures).

Table 4: EDX elemental analysis averaged over the total surface

atom % C O Al Si

EDX

(this study) 85 +/- 1 11.5 +/- 1 1.5 +/- 0.5 2 +/- 0.5

Table 5: F400 Proximate and Ultimate analysis parameters

% C O N S H Moist-

ure

Fix.

carbon

Vol.

matter ash

This

study

89.55

+/-

0.22

5.77

+/-

0

0.25

+/-

0.04

0.77

+/-

0.01

0.21

+/-

0.02

1.52

+/-

0.17

91.06

+/-

0.28

1.32

+/-

0.03

6.10

+/-

0.11

Fitzgerald

[3] 88.65 3.01 0.4 0.73 0.74 -/- 89.86 3.68 6.46

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

99 | P a g e

a)

b)

1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00

100

200

300

400

500

Relative pressure (p/p0)

Volu

me a

dsorb

ed (

cm

3 g

-1)

5 10 15 20 25 30

2E-03

4E-03

6E-03

8E-03

1E-02

Pore diameter (nm)

dV

(W

) (c

m3 n

m-1

g-1

)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

100 | P a g e

c)

d)

Figure 28: a)The N2 isotherm at 77K (80 data points, 7x10-6 to 0.9965 p/p0) of the F400 in

a log plot. b) pore diameter distribution over the complete range based on NLDFT and

c) the important range between 0-5 nm. d) SEM picture including EDX-Analysis (see

table1)

0 1 2 3 4 5

2E-03

4E-03

6E-03

8E-03

1E-02

Pore diameter (nm)

dV

(W

) (c

m3 n

m-1

g-1

)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

101 | P a g e

Sample preparation

The drying procedures differed slightly among the participating laboratories. In three

laboratories the sample was dried in the measuring cell to avoid any contact with

atmospheric air after drying.

At RWTH Aachen the sample (5-7g) was degassed at 473 K in situ within the sample cell

used for sorption measurements. The sample cell was placed into a heating sleeve and

heated to 473 K under vacuum (10-2 Pa) for 24h.

At FP Mons about 1.5 to 2 grams of original sample were degassed for 24 h at

10-2 Pa using a turbomolecular pump with a temperature ramp of 1 K min-1 up to 473K.

At the TU Delft laboratory the sample cell containing ~35 g of sample was detached from

the sorption set-up, placed in an electric oven and heated under vacuum to 473K for 24

h. The evacuation pressure of the pump was <100 Pa. To avoid air influx the sample cell

was filled with He for transfer to the sorption set-up.

At INERIS about 1.5 to 2 grams of original sample was dried in an oven at 473 K for 24

hours. Afterwards it was placed into the sample cell, which was then evacuated to 10-4Pa

for another 24 hours before the start of the experiment.

Experimental methods Three different techniques are commonly used to determine gas sorption isotherms on

coals: the manometric [9-12] or piezometric [13] method, the volumetric method [14]

and the gravimetric method [5,15,16]. Modern gravimetric sorption devices mostly

employ magnetic suspension balances that allow contactless weighing of samples across

the walls of closed high-pressure systems. The principles of the different techniques are

discussed by Goodman et al. [1,2]. Although these three methods make use of different

physical principles and parameters, they usually provide accurate and comparable

sorption isotherms for simple and well-characterized sorption systems (e.g. methane on

activated carbon or natural coals). The primary experimental parameter obtained from

all these procedures is the excess sorption (Gibbs sorption, Gibbs excess) [13].

Manometric set-up (RWTH Aachen, TU Delft)

Sorption experiments at RWTH Aachen University and TU Delft were performed using

the manometric technique with customized in-house experimental devices (see table 6

for details). Both set-ups have the same basic components such as reference volume,

measuring cell, valves, high-precision pressure gauges and temperature control units,

but differ in size.

In the manometric or piezometric procedure, defined amounts of gas are successively

transferred from a calibrated reference cell into the measuring cell containing the coal

sample. Prior to the sorption experiment the void volume of the measuring cell (0

voidV ) is

determined by expansion of a “non-adsorbing” gas - typically helium - using Boyle’s law

and McCarthy He-EoS. [27] This procedure also provides the skeletal volume ( 0

sampleV )

and the skeletal density ( 0

sample ) of the sample.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

102 | P a g e

The void volume, multiplied by the density of the gas (or supercritical) phase

( ),(20 pTV CO

void ), yields the “non-sorption” reference mass, i.e. the amount of gas

(supercritical fluid) that would be accommodated in the measuring cell if no sorption

took place. The excess sorption mass ( 2CO

excessm ) is the difference between the mass of gas

that has been actually transferred into the measuring cell up to a given pressure step

and the “non-sorption” reference mass:

),(222 0 pTVmm CO

void

CO

dtransferre

CO

excess (1)

The mass transferred from the reference cell into the measuring cell during N successive

pressure steps is given by:

N

i

COe

i

COf

iref

CO

dtransferre Vm1

,, )( 222 (2)

Thus, the following relationship is obtained for the excess sorption mass.

),()( 2222 0

1

,, pTVVm CO

void

N

i

COe

i

COf

iref

CO

excess (3)

The excess sorption mass is usually normalized to the initial mass of the sorbent. In the

present study it is expressed in units of amount of substance (mol/kg or mmol/g).

RWTH Aachen University

The manometric units at RWTH Aachen University have been described previously by

Krooss et al. [9] and Busch et al. [10]. They are equipped with Tecsis Series P3382

pressure sensors with internal diaphragm. Pressure ranges are from primary vacuum

(10-2 Pa rotary vane vacuum pump) up to 16 or 25 MPa with an accuracy of ±0.05 % of

the full-scale (FS) value and standard output is an RS 232-interface.

Two pneumatically actuated Valco 3-port switching valves with 1/16” connectors are

used to control the gas transfer through the calibrated reference volume and into the

sample cell. The reference volume consists of the void volume of the pressure sensor the

1/16” stainless-steel tubing between the switching valves. The reference volume is in

the range of 1.7 cm³ and is determined by helium expansion with an accuracy of ±

0.0003 cm³.

The entire manometric set-up (valves, pressure sensor, measuring cell) is kept in a

thermostated air bath. Thermostatic ovens of different suppliers (Heraeus, Binder,

Varian) are in use for the various sorption set-up presently operated in the RWTH

laboratory.

The experimental temperatures were monitored using type K (NiCr-Ni) thermocouples

(Roessel Messtechnik GmbH,) connected to a Keithley Model 2000 Multimeter equipped

with a 2001-TCSCAN Thermocouple Scanner Card with cold junction compensation

(CJC). Reference measurements with a high-precision (class A) Pt100 Resistive

Temperature Detector (RTD) revealed that the temperature readings taken via the

thermocouples were consistently lower than the true temperatures. The offset ranged

from 0.1 up to 0.4 °C. The a priori uncertainty of the excess sorption measurements,

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

103 | P a g e

determined by a blank isotherm is in the order of 0.005 mmol. Further specifications of

the experimental set-up are listed in table 6.

TU Delft

The volumes of the DUT sorption set-up and the sample amounts used in the

measurements are approximately six times larger than those used at RWTH Aachen. It is

equipped with a 9000 series Paroscientific pressure sensor with a piezo-electric

element. The entire set-up is immersed in a thermostated water bath (Lauda RP485)

that keeps temperature variations within 0.03 K. Temperature is measured with a Pt100

RTD with an accuracy of 0.02 K and recorded by an F200 reader (Automated System

Laboratories). The cells are connected by pneumatically actuated 2-position Valco

valves. The main contributions to the a priori uncertainty of 0.04 mmol/g are discussed

by van Hemert et al. [17].

Gravimetric method (FP Mons, INERIS)

In the gravimetric instruments the sorbent is placed into the high-pressure

compartment of a magnetic suspension balance (Rubotherm) and exposed to the

sorptive, CO2, at constant temperature and increasing pressure. The excess sorption is

determined from the mass change of the sample ( 0),( samplemeasured mpTmm )

recorded during this procedure, where 0

samplem is the original sample mass. This

apparent sample mass change is corrected by a buoyancy term based on the skeletal

volume ( 0

sampleV ) of the sample, i.e., the same reference state as in the manometric

procedure. The determination of the skeletal density or volume is performed with

helium, which is assumed to be non-adsorbing.[18] The excess sorbed mass is then

given by:

),(2

2 0 pTVmm COsample

CO

excess (4)

),(2

2

0

0

pTm

mm CO

sample

sampleCO

excess

(5)

As in the manometric procedure it is normalized to the original sample mass 0

samplem and

expressed in molar units (mol/kg) in this study.

The FP Mons set-up is discussed in detail by De Weireld et al. [19,20]. The most

important technical features are as follows:

The weight changes are measured with a 10 µg accurate Rubotherm magnetic

suspension balance. The magnetic system consists of an electromagnet linked to the

balance and a permanent magnet at the top of the suspension system for the crucible

containing the sorbent. The suspension system is housed in a high-pressure adsorption

chamber allowing for experiments under high temperature (243 to 393 K), high

pressure (vacuum - to 15 MPa) and corrosive conditions. Pressure is measured with

three different pressure sensors, an MKS Baratron 621B with a resolution of 1.3 Pa for

secondary vacuum to 133.3 kPa, an MKS Baratron 621B with a range from 32.5 Pa for

secondary vacuum up to 3.333 MPa and a Tecsis Series P3382 pressure sensor with

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

104 | P a g e

internal diaphragma for a maximum pressure of 16 MPa with an accuracy of 0.1% of the

full scale. Temperature measurements of the gas phase for the determination of the

density are performed with a high-precision (class A) Pt100 RTD. The installation is

placed in a thermostatic oven ensuring constant temperature during experiments. This

homogeneous temperature field avoids condensation of sub-critical gases [24,25]. The a

priori uncertainty of the excess sorption measurements is estimated at 5 % of the

maximum sorption.

The INERIS apparatus is very similar to the FP Mons set-up. At INERIS the pressure was

measured with one pressure sensor (GE Sensing PMP4010) with an accuracy of 0.08 %

of the full scale value. Temperature measurements of the gas phase for the

determination of the density were performed with a Pt100 RTD from Thermosensor

GmbH with an accuracy of ±0.05 K. The a priori uncertainty of the excess sorption

measurements is estimated at 5 % of the maximum sorption.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

105 | P a g e

Table 6: Specifications of experimental devices used in this inter-laboratory study

RWTH Aachen TU Delft FP Mons INERIS

Method manometric manometric gravimetric gravimetric

Max. CO2

pressure (MPa) 25 ± 0.0125 20.7 ± 0.001 16 ± 0.016 5 ± 0.01

Measuring

temperature (K)

318.6 ± 0.2

318.8 ± 0.2

318.11± 0.01

318.12± 0.01

318.5± 0.1

318.6± 0.1 318.2± 0.1

Sample mass (g) 5-7 ±0.0001 34.95±0.03

35.57±0.03

1.69804 ±

0.00002

1.68711 ±

0.00002

2.0151

±0.0002

Sample cell

Volume (cm³)

13.085 ±

0.001

78.33±0.06

75.9 ±1 - -

Reference cell

volume (cm³)

1.7785 ±

0.0003

12.152±0.009

3.524±0.004 - -

Average void

volume (cm³) 10.450

61.1 ±0.1

59.2 ±0.1 - 111.5±0.2

Volume of the

system (crucible

+ sample) (cm³)

- - 1.575 ±0.002

1.684 ±0.002 1.68±0.01

Equilibration

time (h) 1-2

30

3 1-3 24

CO2 purity 99.995% 99.990% 99.996% 99.998%

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

106 | P a g e

Results

The 318 K sorption isotherms for CO2 measured on the F 400 activated carbon

measured by the participating groups are plotted in Figure 29 on a linear (a) and a

logarithmic pressure axis (b). These plots of excess sorption vs. pressure show an

increase until a maximum value of ~8 mmol/g is reached at a pressure of ~5 MPa.

Subsequently, excess sorption decreases with an inflection point around 10 MPa.

Table 7 lists the maximum excess sorption values of the individual experiments and the

respective measuring temperature. The helium skeletal density values of the F400

activated carbon measured in these tests (Table 7, column 4) are essentially identical.

Table 7: Comparison of experimental results for CO2 excess sorption isotherms on F400

activated carbon.

nmax

[mmol/g]

T

[K]

skeletal density

[kg/m³]

FP Mons-1 7.95 318.5 2140

FP Mons-2 7.87 318.6 2200

RWTH-1 8.23 318.6 2110

RWTH-2 8.17 318.8 2113

TU Delft-1 7.97 318.2 2070

TU Delft-2 7.99 318.2 2100

INERIS 7.67 318.2 2280

average 8.0 2140

std. dev. 0.16 70

Duplicate measurements of the excess sorption isotherms within the individual

laboratories show excellent intra-laboratory repeatability with deviations of 0.4 , 0.3,

0.2 % for RWTH Aachen, FP Mons and TU Delft respectively.

The inter-laboratory comparison shows that the RWTH Aachen excess sorption values

are slightly higher than those of the other participants in the 1 to 8 MPa pressure range

with a maximum deviation of ~0.3 mmol/g. In the high pressure range from 12 to 16

MPa the excess sorption data of FP Mons are somewhat (~0.3 mmol/g) lower than those

of the two other laboratories. Generally, the level of accuracy for these types of

experiments is good to excellent and the variability in the sorption data of the different

laboratories is considered acceptable.

a)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

107 | P a g e

b)

Figure 29: Excess CO2 sorption isotherms on activated carbon Filtrasorb F400 plotted on a

linear (a) and a logarithmic (b) pressure scale.

0

1

2

3

4

5

6

7

8

9

0 2 4 6 8 10 12 14 16

Pressure / (MPa)

Excess

so

rpti

on

/ (

mm

ol/g

)

FP Mons-1 (318.5 K)

FP Mons-2 (318.6 K)

RWTH-1 (318.6 K)

RWTH-2 (318.8 K)

TU Delft-1 (318.2 K)

TU Delft-2 (318.2 K)

INERIS (318.0 K)

0

1

2

3

4

5

6

7

8

9

0.01 0.1 1 10 100

Pressure / (MPa)

Excess

so

rpti

on

/ (

mm

ol/g

)

FP Mons-1 (318.5 K)

FP Mons-2 (318.6 K)

RWTH-1 (318.6 K)

RWTH-2 (318.8 K)

TU Delft-1 (318.2 K)

TU Delft-2 (318.2 K)

INERIS (318.0 K)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

108 | P a g e

In figure 30 the excess sorption values of the laboratories FP Mons, TU Delft and RWTH

Aachen are plotted versus the density of the free CO2 phase at the corresponding

pressure and temperature conditions. These isotherm plots exhibit the typical shape

discussed by Menon [21]. Maximum excess sorption values of approximately 8 mmol/g

are reached by all isotherms at a CO2 density around 100 kg/m3. Beyond a CO2 density of

~250 kg/m3 the excess sorption isotherms decrease linearly. The intercept of the

extrapolated linear trend with the density axis provides an estimate of the density of the

sorbed phase (see Figure 30). At this point the densities of the free phase and the

adsorbed phase are identical, i.e. the two phases cannot be discriminated any longer.

The density of the adsorbed CO2 was estimated by extrapolation of the excess sorption

vs. density plots in the density range >250 kg/m³. The density values obtained by this

procedure ranged from 956 to 1014 kg/m³. The lowest density values arose from the FP

Mons data while the highest values were obtained from the TU Delft results.

Figure 30: Excess sorption versus density of free gas phase (using the Span & Wagner

equation) for activated carbon Filtrasorb F400 at 318 K

0

1

2

3

4

5

6

7

8

9

0 200 400 600 800 1000 1200CO2 density / (kg/m³)

Excess

so

rpti

on

/ (

mm

ol/g

)

FP Mons-1 (318.5 K)

FP Mons-2 (318.6 K)

RWTH-1 (318.6 K)

RWTH-2 (318.8 K)

TU Delft-1 (318.2 K)

TU Delft-2 (318.2 K)

sorbed phase

density range:

956 - 1014 kg/m³

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

109 | P a g e

Parameterization of the experimental results

The experimental high-pressure CO2 excess sorption isotherms obtained in this study

were approximated by the following excess sorption function:

sorbed

freeabsoluteexcess

pTpTspTs

),(1),(),(

[mmol/g] (6a)

Here the absolute amount of adsorbed substance, ),( pTsabsoute , was expressed by the

Langmuir function:

pK

pspTs

VL

absoluteabsoute

,

),(

(6b)

This function was chosen because it is a simple, steady function increasing

monotonously with pressure that can be derived from the concept of a dynamic

equilibrium between free and adsorbed molecules. The Langmuir parameters KL,V and

sorbedabsolutes and the density of the adsorbed phase ( sorbed ) in equation (6) were

adjusted simultaneously by a least-square regression. The Langmuir coefficient KL,V is

the controlling factor of the excess sorption function (6a) in the low-pressure region

where the volume of the adsorbed phase is negligible. The density ratio

sorbed

free pT

),( of

the free (“gas”) vs. the sorbed gas phase controls the shape of the isotherm above the

critical pressure when the density of the supercritical CO2 phase increases rapidly.

The results of the regressions for the individual experimental excess sorption isotherms

are listed in table 8. The quality of the fits is expressed by the parameter n according to

equation (7). Here N is the number of data points of the isotherm, and n and nfit are the

measured and the fitted excess sorption values of data point n.

2

1

1 N

fitnnN

n

(7)

Also listed in Table 8 is the parameter set that provided the best fit of equation (2) to the

seven excess sorption isotherms measured by TU Delft, FP Mons, INERIS and RWTH

Aachen during this inter-laboratory study. Prior to this regression calculation the data

densities of the experimental curves were homogenized in order to avoid a bias that

might result from different data densities of individual isotherms in certain pressure

intervals. Figure 31 shows the experimental sorption isotherms with the regression

function based on the best fit parameters in table 8.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

110 | P a g e

Table 8: Regression parameters of CO2 excess sorption isotherms on F400 activated

carbon obtained as best 3-parameter fits of equation (2) to experimental results of this

study. The density of the sorbed phase could not be determined from the INERIS

isotherm because the isotherm was only measured up to 5 MPa.

nsorbed()

[mmol g-1]

KL,V

[MPa]

sorbed

[kg m3]

n

(quality of fit)

FP Mons-1 11.19 1.235 963 0.030

FP Mons-2 11.03 1.225 964 0.036

RWTH-1 11.1 1.036 976 0.033

RWTH-2 11.2 1.099 993 0.030

TU Delft-1 11.0 1.126 992 0.013

TU Delft-2 10.99 1.283 995 0.010

INERIS (10.0) (0.90) (981) 0.042

Average 11.09 1.167 981

Std. dev. 0.085 0.087 13.5

Tel. std. dev. 0.8% 7.4% 1.4%

Best fit for all experimental data of this study:

EU-Round Robin 10.97 1.082 997 0.019

In Figure 32 the normalized deviations (sregression-smeasured)/sregression) of the experimental

data from the best fit function are plotted vs. pressure. It is evident from this diagram

that the majority of experimental data points in the pressure range >1 MPa fall within

+/- 5% of the regression function. This relative deviation is of the same order as the

overall variability of the experimental data from the participating laboratories. Thus, the

Langmuir-based regression function can be considered to represent, at present, with

sufficient accuracy high-pressure CO2 excess sorption isotherms measured in different

laboratories.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

111 | P a g e

Figure 31: Experimental excess sorption isotherms (318 K) for Filtrasorb F400 activated

carbon measured in this study with best fit of excess sorption function according to

equation (2).

Figure 32: Normalized deviation of experimental CO2 excess sorption isotherms (318 K) for

F400 from the best fit vs. pressure.

Comparison with previously published results Figure 33 shows a set of previously published CO2 isotherms for F400 activated carbon

((Humayun and Tomasko, 2000) Pini et al. [5], Sudibandriyo et al. [6]). Also included in

0

1

2

3

4

5

6

7

8

9

0 2 4 6 8 10 12 14 16

Pressure / (MPa)

Excess

so

rpti

on

/ (

mm

ol/g

)

FP Mons-1

FP Mons-2

RWTH-1

RWTH-2

TU Delft-1

TU Delft-2

INERIS

Best fit

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

0 2 4 6 8 10 12 14 16

Pressure / (MPa)

No

rma

lised

devia

tio

n

(sregression-smeasured)/sregression)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

112 | P a g e

this diagram is the best fit excess sorption isotherm obtained in the present inter-

laboratory study. The locations of the excess sorption maxima and the extrapolated

sorbed phase densities (cf. (Humayun and Tomasko, 2000) [4]) coincide with those

obtained in the present study, but the maximum excess sorption values of the present

set of experiments are consistently higher by approximately 10%. This slightly higher

excess sorption capacity is most probably due to differences in sample preparation or

due to difference in the sample batch. While the activation temperature in this study is

200°C over 24h and Humayun and Tomasko [4] dried the sample at 110°C over 12h.

Additonal the procedure of the EU round robin explicitly required either in-situ

activation of the F400 or transfer to the measuring cell without exposure to air. The

differences in the sample batch are hardly to measure. But Humayun and Tomasko [4]

specified her batch with a BET surface area of 850m2g-1 and micropore volume of

0.37cm3g-1 determined by N2 sorption at 77 K. The batch used in this study shows a BET

surface area of 1063 m2g-1 also determined by N2 sorption at 77K performed by RWTH -

Aachen. The measured N2 BET surface is 1063 m2g-1, calculated over 10 data points in

the relative range between 0.1 and 0.3. The DR micropore volume is 0.51 cm3g-1,

determined in the relative pressure range between 4x10-5 and 3x10-2 over 28 data

points.

Figure 33: Comparison of CO2 excess sorption results for F400 activated carbon of the

present study (dashed line) with literature data.

Discussion

The inter-laboratory studies on CO2 sorption on natural (Argonne Premium) coals by

Goodman et al. (2004, 2007) [1,2] have revealed substantial discrepancies among the

results of the participating laboratories. The European inter-laboratory study on high-

pressure CO2 sorption, initiated in a joint attempt to overcome these experimental

0

1

2

3

4

5

6

7

8

9

0 5 10 15 20

Pressure / (MPa)

Excess

so

rpti

on

/ (

mm

ol/g

)

Pini et al. (2006); 318.4 K

Sudibandriyo et al. (2003); 318.2 K

Sudibandriyo et al. (2003); 318.2 K

Humayun&Tomasko (2000); 318.2 K

this study

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

113 | P a g e

problems, has yielded very promising results. The isotherms determined by the

participating laboratories on an F400 activated carbon sample showed an excellent

agreement with inter-laboratory deviations <5% and a very good intra-laboratory

reproducibility (variations <1%). Workshops and exchange of technical information

among the member groups during the first phase of this initiative have substantially

contributed to an improvement of sample preparation and measuring procedures and

the identification of potential errors and pitfalls in the determination of high-pressure

CO2 sorption isotherms. The direct comparison of manometric and gravimetric

techniques indicated excellent agreement. Selected issues concerning the accuracy of the

experimental measurements are discussed briefly in the following sections and

appendix.

Effect of moisture content and variability of batch composition

To study the effects of even short-term exposure to atmospheric air/moisture, one set of

sorption measurements was performed at RWTH Aachen with F400 aliquots that were

exposed to atmospheric air for several minutes after the activation procedure. The two

resulting isotherms denoted as RWTH-3 and RWTH-4 are plotted in Figure 34 together

with published literature data (see above). The two RWTH isotherms show a very good

agreement with those published by Humayun & Tomasko [4] and Sudibandriyo et al. [6]

reaching a maximum excess sorption capacity of ~7 mmol/g. The CO2 excess sorption

isotherm for F400 by Pini et al. [5] has a slightly higher maximum excess sorption

capacity (~7.4 mmol/g), ranging between the values of Humayun & Tomasko [4] and

Sudibandriyo et al. [6] on the one hand and those of the present study on the other hand.

This comparison shows that failure to rigorously adhere to well-defined identical

sample treatment procedures can lead to significant discrepancies in CO2 excess

sorption measurements on activated carbons. An alternative, though less likely

explanation for such discrepancies could be variability of sorption properties of

different batches of activated carbon (table 9).

Table 9: characteristic surface parameters determined low pressure isotherm of N2 at 77K

BET surface area

(m2 g-1)

DR Micropore

volume

(cm3 g-1)

NLDFT Micropore

volume

(cm3 g-1)

F400 batch of

this study 1063 0.51 0.64

Humayun [4] 850 0.37

We acknowledge that the surface area available to CO2 may be different from that

measured by N2 adsorption. However, this variability, in particular with respect to CO2

sorption could explain the difference to the published [4-6] isotherm.

In the present study the same batch of F400 was used by all participating laboratories.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

114 | P a g e

Figure 34: Comparison of previously published excess sorption isotherms for CO2 on F400

activated carbon with two measurements performed at RWTH Aachen on samples

exposed briefly to air after activation.

The experimental sorption isotherms from the literature and the RWTH-3 and RWTH-4

isotherms were fitted by the three-parameter excess sorption function (equation 2). The

parameters resulting from this procedure and information on the quality of the fits are

summarized in table 10.

Using Helium as reference gas Both methods gravimetric as well as manometric use helium as an reverence gas to

determine the dead volume or the sample volume buoyancy correction. Effects of a

possible Helium sorption in the experiment lead with both methods of a possible

underestimation of the sample volume. This results in a mistake of the absolute sorption

capacity, nevertheless, not in methodical differences. Hence, we recommend to

understand the sorptions experiments manometric as well as gravimetric as a

differential measuring methodology. It compares approximately not adsorbing gas,

helium with much stronger adsorbing gas, CO2. To the authors knowledge there is no gas

known which shows less sorptive interaction forces than helium.

Table 10: Fit parameters of CO2 excess sorption isotherms on F400 activated carbon from

the literature and RWTH Aachen measurements 3 and 4 (modified drying procedure).

Sorbed phase density (sorbed), maximum absolute sorption capacity nsorbed() and

Langmuir coefficient (KL,V) were fitted by regression of equation (2).

0

1

2

3

4

5

6

7

8

9

0 5 10 15 20 25

Pressure / (MPa)

Ex

ces

s s

orp

tio

n / (

mm

ol/g

)Pini et al. 2006 (318.4K)

Sudibandriyo et al. 2003 (318.2K)

Sudibandriyo et al. 2003 (318.2K)

Humayun&Tomasko 2000 (318.2K)

RWTH - 3 (318.2K)

RWTH - 4 (318.7K)

RR-Isotherm (best fit)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

115 | P a g e

Excess

sorption

at 5 MPa

(mmol/g)

T

(K)

sorbed

(kg/m³)

nsorbed()

(mmol/g)

KL,V

(MPa)

No. of

data

points

Qual-

ity of

fit

n

RWTH – 3 7.2 318.2 1014 9.5 0.97 21 0.20

RWTH – 4 7.1 318.2 995 9.5 0.98 17 0.17

Pini et al. [5] 7.4 318.4 1043 10.4 1.20 19 0.12

Sudibandriyo et al.

[6] 7.0 318.2 1019 9.3 0.94 13 0.12

Sudibandriyo et al.

[6] 6.9 318.2 1013 9.2 0.90 13 0.08

Humayun &

Tomasko [4] 7.0 318.2 1008 9.5 1.05 4779

Conclusion

The results of the first phase of the European inter-laboratory test on high-pressure CO2

sorption show an excellent agreement of the results obtained by the four participating

laboratories. The deviations of the sorption isotherms are less than 5% and the intra-

laboratory reproducibility is better than 1% for each laboratory. One result of the study

is the high comparability of isotherms obtained from different methods, i.e. manometric

and gravimetric, taking into account a proper sample preparation and starting

conditions.

Excess sorption values measured in this study are consistently higher than those

reported in the literature for the same sorbent/sorptive system (F400/CO2). This offset

is most likely due to difference in the initial conditions (activation and drying of the

samples) but could also reflect variability in the batches of F400 used.

A three-parameter excess sorption function based on a Langmuir-type absolute sorption

function was found to be adequate to represent the experimental data of this study with

sufficient accuracy within the limits of present-day experimental uncertainties. The “EU-

Round Robin” parameter set reported in table 5 is considered to define the presently

highest quality excess sorption isotherm for CO2 on F400 at 318 K.

The determination of accurate high-pressure sorption isotherms for CO2 still represents

a challenge. However, the increasing number of published high-quality data indicates

that significant progress has been made during recent years.

Inter-laboratory studies can help to identify and avoid pitfalls and to formulate standard

procedures that improve overall data quality. The present study has revealed a number

of potential experimental problems in particular of the manometric procedure and has

identified strategies to avoid or minimize their impact on data quality. Thus, apart from

using high performance pressure and temperature sensors, careful adjustment of

pressure steps during the measuring procedure is recommended. Finally the importance

of well-defined procedures for sample preparation is emphasized.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

116 | P a g e

The results presented here provide a benchmark for future studies on the sorption of

supercritical CO2 on natural coals and help to improve and optimize the experimental

standards. Meanwhile several other laboratories world-wide have joined this initiative

and have been supplied with standards and instructions. Work is presently in progress

to determine high-pressure excess sorption isotherms of CO2 on selected natural coals

with the same level of accuracy as for the F400 activated carbon.

Acknowledgements

This research was conducted in the context of the MOVECBM (Monitoring and

Verification of Enhanced Coalbed Methane) Project supported by the European

Commission under the 6th Framework Programme.

The raw data for this comparison were kindly provided by the authors of these

publications [4-6].

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

117 | P a g e

Notation

Symbols Units Physical

KL,V [MPa] Langmuir parameter

mtransferred [g] Mass transferred from reverence to sample cell

mcoal,msample [g] mass of coal

mexcess [g] Excess sorption mass

m [g] Incremental sample weight in gravimetrical experiment

n [mmol] amount of substance of gas

p [MPa] Gas pressure

pi [MPa]

i = equil. p in sample cell of previous sorption step

i+1 = equil. pressure of filling the reference cell

i+2 = equil. pressure in sample cell of sorption step

pri: [MPa] initial reference cell pressure

prf: [MPa] final reference cell pressure

psi: [MPa] initial sample cell pressure

psf: [MPa] final sample cell pressure

R = 8.31451 [J mol-1 K-1] gas constant

T [K] Temperature

tequil [h] Time taken for equilibration

sexcess [mmol g-1] Excess sorption

sabsolute [mmol g-1] Absolute sorption

VR , Vref [cm³] reference cell volume

VV, Vvoid [cm³] void volume of sample cell

Vsample [cm³] Skeletal volume

zri: initial reference real gas compressibility factor

zrf: final reference real gas compressibility factor

zsi: initial sample real gas compressibility factor

zsf: final sample real gas compressibility factor

i: inaccuracy of measurement of parameter i

i,CO2: [g cm-3] Free gas density of CO2 to the time i , e=equilibrium

sample: [g cm-3] Skeletal density of the sample

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

118 | P a g e

References

[1] Goodman AL, Busch A, Duffy G, Fitzgerald JE, Gasem KAM, Gensterblum Y, et al. An

Inter-laboratory Comparison of CO2 Isotherms Measured on Argonne Premium Coal

Samples. Energy and Fuel 2004; 18: 1175-1182.

[2] Goodman AL, A Busch, M Bustin, L Chikatamarla, S Day, GJ Duffy, JE Fitzgerald, et al.

Inter-laboratory comparison II: CO2 isotherms measured on moisture-equilibrated

Argonne premium coals at 55°C and up to 15 MPa. International Journal of Coal

Geology 2007; 72: 153-164.

[3] Fitzgerald JE, Robinson RL Jr, Gasem KAM, Modeling High-Pressure Adsorption of

Gas Mixtures on Activated Carbon and Coal Using a Simplified Local-Density Model.

Langmuir 2006; 22: 9610-9618.

[4] Humayun R, Tomasko DL, High-resolution adsorption isotherms of supercritical

carbon dioxide on activated carbon. AIChE Journal 2000; 46: 2065-2075.

[5] Pini R, Ottiger S, Rajendran A, Storti G, Mazzotti M, Reliable measurement of near-

critical adsorption by gravimetric method. Adsorption 2006; 12: 393-403.

[6] Sudibandriyo M, Pan Z, Fitzgerald JE, Robinson Jr RL, Gasem KAM, Adsorption of

methane, nitrogen, carbon dioxide, and their binary mixtures on dry activated

carbon at 318.2 K and pressures up to 13.6 MPa. Langmuir 2003;19: 5323-5331.

[7] ASTM D1412-99: Standard Test Method for Equilibrium Moisture of Coal at 96 to 97

Percent Relative Humidity.

[8] Jagiello J, Thommes M, Comparison of DFT characterization methods based on N2, Ar,

CO2, and H2 adsorption applied to carbons with various pore size distributions.

Carbon 2004; 42: 1227-1232.

[9] Krooss BM, van Bergen F, Gensterblum Y, Siemons N, Pagnier HJM, David P, High-

pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated

Pennsylvanian coals. International Journal of Coal Geology 2002; 51: 69-92.

[10] Busch A, Gensterblum Y, Krooss BM, Methane and CO2 sorption and desorption

measurements on dry Argonne Premium Coals: Pure components and mixtures.

International Journal of Coal Geology, 2003; 55: 205-224.

[11] van Hemert P, Wolf K-H A A, Bruining J, The intrinsic reliability of manometric

sorption apparatus using supercritical carbon dioxide, SPE Annual Technical

Conference and Exhibition, Arnheim, CA, 2007;pp. 11.

[12] Siemons N, Busch A, Measurement and interpretation of supercritical CO2 sorption

on various coals. International Journal of Coal Geology 2007; 69(4): 229-242.

[13] Sircar S, Gibbsian surface excess for gas adsorption - revisited. Industrial and

Engineering Chemistry Research 1999; 38: 3670-3682.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

119 | P a g e

[14] Fitzgerald JE, Z Pan, M Sudibandriyo, RL Robinson Jr, KAM Gasem, S Reeves,

Adsorption of methane, nitrogen, carbon dioxide and their mixtures on wet Tiffany

coal. Fuel 2005; 84: 2351-2363.

[15] Bae JS, Bhatia SK, High-Pressure Adsorption of Methane and Carbon Dioxide on

Coal. Energy & Fuels 2006; 20: 2599-2607.

[16] De Weireld G, Frère M, Jadot R, Automated determination of high-temperature and

high-pressure gas adsorption isotherms using a magnetic suspension balance.

Measurement Science and Technology 1999; 10: 117-126.

[17] van Hemert P, Bruining J, Rudolph E, Wolf K-H AA, Maas JG, Review of Scientific

Instruments 2009 (submitted)

[18] Rouquerol F, Rouquerol J, Sing K, Adsorption by powders and porous solids;

Academic Press London: 1999.

[19] Belmabkhout Y, Frère M, De Weireld G, High-pressure adsorption measurements. A

comparative study of the volumetric and gravimetric methods. Measurement Science

and Technology 2004; 15: 848-858.

[20] Frère M, De Weireld G, Jadot R, In Fundamentals of Adsorption, Proceedings of the

6th International Conference on Fundamentals of Adsorption, Presqu’île de Giens,

France, 1998; Meunier F, Elsevier Paris, France, 279-284.

[21] Menon PG, Adsorption at high pressures. Chemical Reviews 1967; 68: 277-294.

[22] Mavor MJ, Hartman C, Pratt TJ, Uncertainty in Sorption Isotherm Measurements,

paper 411, Proceedings 2004 International Coalbed Methane Symposium,

University of Alabama, Tuscaloosa (May 2004)

[23] Span R, Wagner W, A new equation of state for carbon dioxide covering the fluid

region from the triple-point temperature to 1100 K at pressures up to 800 MPa.

Journal of Physical and Chemical Reference Data 1996; 25: 1509-1596.

[24] Dreisbach F, Staudt R, Tomalla M, Keller JU, In Fundamentals of Adsorption,

Proceedings of the 5th International Conference on Fundamentals of Adsorption,

Asilomar, CA, USA, may 1995; Le Van, M.D.; Kluwer Academic Publishers1996,

Boston, MA, USA, 259-268.

[25] Dreisbach F, Seif R, Losch HW, Adsorption equilibria of CO/H2 with a magnetic

suspension balance: Purely gravimetric measurement. Journal of Thermal Analysis

and Calorimetry 2003; 71: 73-82.

[26] McCarty RD, Arp VD, A new wide range equation of state for helium. Advances in

Cryogenic Engineering 1990; 35: 1465-1475.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

120 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

121 | P a g e

“Science is organized common sense

where many a beautiful theory

was killed by an ugly fact.”

(Thomas Henry Huxley)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

122 | P a g e

3. European inter-laboratory compari-

son of high pressure CO2 sorption

isotherms. II: natural coals

International Journal of Coal Geology 84 (2010) 115–124

Y. Gensterblum and B.M. Krooss

RWTH Aachen University, Institute of Geology and Geochemistry of Petroleum

and Coal, Lochnerstr. 4-20, 52056 Aachen, Germany

P. van Hemert, E. Battistutta and K.-H.A.A. Wolf

TU Delft, Department of Geotechnology, Stevinweg 1, NL-2628 CN Delft, The

Netherlands

P. Billemont and G. De Weireld

U MONS University of Mons, Faculté Polytechnique, Service de

Thermodynamique et Physique Mathématique, 20 Place du Parc, 7000 Mons,

Belgium

A. Busch

Shell International Exploration and Production, Kessler Park 1, 2288 GS Rijswijk,

The Netherlands

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

123 | P a g e

Abstract

In order to assess and improve the quality of high-pressure sorption isotherms of

carbon dioxide (CO2) on coals, an inter-laboratory study (Round Robin) has been

conducted among three European research laboratories. Excess sorption isotherms,

determined in a first round of measurements, on Filtrasorb 400 (F400) activated carbon

showed excellent agreement. In the second round of this study, excess sorption

isotherms were determined on three coals at 318 K using the manometric (TU Delft,

Netherlands and RWTH Aachen University, Germany) and the gravimetric (University

Mons, Belgium) methods up to 16 MPa. The CO2 excess sorption isotherms for the three

coal samples, a lignite, a bituminous coal and a semi-anthracite, exhibited maximum

values of 1.77±0.07, 1.37±0.01 and 1.37±0.05 mol kg-1 respectively. The pressure ranges

for the observed maximum excess sorption capacities decreased with increasing

maturities from 6.89 ± 0.5 MPa for the lignite, to 6.68 ± 0.4 MPa for the bituminous coal

and to 5.89 ± 0.6 MPa for the semi-anthracite. The results show that high-pressure CO2

excess sorption isotherms on natural coals in the supercritical range can be determined

accurately with both gravimetric and manometric equipment.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

124 | P a g e

Introduction

Among the various options considered for geological storage of carbon dioxide (CO2),

the injection of CO2 into deep, unminable coal seams in particular in combination with

the production of coalbed methane (CBM), is considered a niche technology. The

European RECOPOL* project has demonstrated the technical feasibility of CO2 injection

into typical European Carboniferous coal seams. The follow-up project MOVECBM†

started in 2006 to investigate in more detail the fate of injected CO2. Laboratory

experiments conducted by the groups at the Delft University of Technology (The

Netherlands) within the Dutch CATO‡ project and the RWTH Aachen University

(Germany) within RECOPOL and the national CO2TRAP§ project provided important

fundamental information on the interaction of natural coals with carbon dioxide and

methane under in-situ conditions (Busch et al. 2006; Busch et al. 2007).

However, considerable problems in the reproducibility of supercritical CO2 sorption

measurements are evident. Similar problems were encountered by other groups and

have been addressed by two recent inter-laboratory studies (Goodman et al., 2007;

Goodman et al., 2004). The results of these studies showed, in spite of considerable

improvements in accuracy, the quality of CO2 sorption isotherms does not yet meet the

standards required for reliable modelling and predictions.

In this context the three laboratory research groups, from Delft University of Technology

(TU Delft), RWTH Aachen University (RWTH) and University of Mons Belgium (U Mons),

decided to perform an inter-laboratory study to assess and improve the quality of

sorption data at high pressures for supercritical carbon dioxide. In contrast to earlier

inter-laboratory tests (Goodman et al., 2007a; Goodman et al., 2004), this study was set

up as an open project with exchange of information and regular seminars. The main

objective was to increase the overall accuracy of CO2 excess sorption measurements,

eliminate pitfalls and sources of error and develop best practice standards and

procedures.

A well-characterized activated carbon sample, Filtrasorb 400 (F400), was selected for

the first series of CO2 isotherm measurements (Gensterblum et al., 2009b). In the second

series we selected three different coals of different rank from Slovenia (lignite VRr

=0.41 ± 0.04), Poland (bituminous coal VRr =0.8) and Wales (semi-anthracite VRr =2.41).

Measurements were performed on dry samples in order to eliminate the effect of water,

which was identified to be one major source of disagreement in capacity in an earlier

inter-laboratory comparison (Goodman et al., 2007a). Sorption experiments were

performed at 318 K up to 15 MPa.

The most common procedures to determine excess sorption isotherms of gases are the

manometric (TU Delft, RWTH Aachen) (Busch et al., 2003a; Busch et al., 2003d), van

Hemert et al. 2007) and the gravimetric (U Mons) methods (Bae and Bhatia 2006; De

* http://recopol.nitg.tno.nl/ † www.movecbm.eu ‡ www.co2-cato.nl § www.co2trap.org

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

125 | P a g e

Weireld et al., 1999; Fitzgerald et al., 2006; Ottiger et al., 2008). Both methods are well

established and known to provide accurate excess sorption isotherms for simple and

well-characterized sorption systems (e.g. methane sorption on activated carbon)

(Belmabkhout et al., 2004).

Previous inter-laboratory comparisons of carbon dioxide sorption on

coal at high pressures

In 2004 and 2007 two inter-laboratory comparison studies on high-pressure sorption of

CO2 on Argonne Premium coals (Goodman et al. 2004, 2007) were initiated by the U.S.

Department of Energy - National Energy Technology Laboratory (DOE-NETL). Although

experienced research groups were involved, large deviations were observed. The

discrepancies were attributed to varying moisture contents. (Goodman et al., 2004)

concluded that further studies with well-defined procedures are required to improve

reproducibility.

The first inter-laboratory study by Goodman et al. (2004) compared the sorption

isotherms of CO2 on dry coals at 295 and 328 K up to a pressure of 7 MPa measured by

five laboratories. Five coals from the Argonne Premium Coal Sample Program (Vorres

1990), covering a maturity range from 0.25 to 1.68 % vitrinite reflectance, were used.

The preparation procedure involved drying of the samples for 36 hours at 353 K under

vacuum. It was found that excess sorption values for medium- to low-rank coals

deviated by more than 100 %. Sorption isotherms on high-rank coals were considered to

be sufficiently accurate. The discrepancies were attributed to varying residual moisture

contents after drying of the coal samples.

In the second inter-laboratory study, Goodman et al. (2007) compared the sorption

isotherms of CO2 on moisture-equilibrated Argonne coals at 328K and pressures up to

15 MPa measured by six laboratories. Moisture-equilibration was achieved by a

modified ASTM D 1412-99 procedure. Three types of coal, covering a maturity range

from lignite to low volatile bituminous, were used. Sorption data showed good

agreement for pressures up to 8 MPa. However, at higher pressures sorption diverged

significantly for the different laboratories. This deviation was attributed to substantial

variations in equilibrium moisture contents.

A well-characterized activated carbon sample, Filtrasorb 400 (F400), was selected for a

more recent inter-lab comparison conducted by the same authors as in this study

(Gensterblum et al., 2009b). An activated carbon was chosen because it was considered

a good starting point to understand differences in sorption capacities since effects like

sample heterogeneity and residual moisture could be avoided. Furthermore, this

material was readily available and its chemical composition and micropore structure are

similar to those of natural coal. It is also resistant to high temperatures, which facilitate

the removal of moisture and attainment of a defined initial condition, one of the main

sources of discrepancies in previous inter-laboratory comparisons. Sorption

experiments were performed at 318 K up to 16 MPa. It was shown that an excellent

agreement in sorption values between the four different laboratories was achieved, with

deviations of less than 5%.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

126 | P a g e

Methods and materials

Samples and sample preparation The coal samples were supplied as lumps or cores from coal mines in Great Britain,

Slovenia and Poland. A summary of the ultimate and proximate analyses and vitrinite

reflectance and maceral compositions is provided in table 11. The Selar Cornish

(Westphalian B) semi-anthracite (VRr=2.41%) was mined in South Wales and has been

used in earlier sorption studies (Siemons et al., 2006; Siemons and Busch, 2007). The

high-volatile bituminous coal sample Brzeszcze 364 (VRr=0.78 %) originates from the

Upper Silesian Basin in Poland and was investigated in detail for sorption (Busch et al.,

2006; Mazumder et al., 2006) and coal swelling (Siemons and Busch, 2007) in the

context of the EU RECOPOL Project. The lignite sample was taken from the Velenje coal

mine in Slovenia and is of Tertiary age. The huminite reflectance of the Velenje sample is

0.41 ±0.04. The pore space characterisation was performed by low-pressure nitrogen

adsorption isotherms at 77K.

Sample preparation

For measurements on powdered coal, the crushed samples were divided and ground to

pass a sieve mesh size of 2 mm. All coal samples were dried at 378K (105°C) for 24h.

The drying procedure differed slightly among the participating laboratories.

At RWTH Aachen the sample (5-7g) was degassed within the measuring cell. The cell

was placed into a heating sleeve and heated to 378 K under vacuum (10-2 Pa) for 24h.

At U Mons about 1.5 to 2 g of coal were degassed for 24 h in a vacuum of 10-2 Pa

maintained by a turbomolecular pump with a temperature ramp of 1 K min-1 up to 378

K.

At TU Delft the measuring cell containing ~35 g of sample was detached from the

sorption set-up, placed in an electric oven and heated under vacuum to 378 K for 24 h.

The evacuation pressure of the pump was <100 Pa. To avoid exposure to air and

moisture, the measuring cell was filled with He for transfer to the sorption set-up.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

127 | P a g e

Table 11: Results from ultimate and proximate analyses and petrographic data for the

three coals studied

Velenje

(Slovenia)

Brzeszcze 364

(Poland)

Selar Cornish

(UK)

VRr (%) 0.41 ± 0.04 0.78±0.2 2.41

Liptinite (%) n.a. 8±1 0

Vitrinite (%) n.a. 63±3 74.5

Inertinite (%) n.a. 28±2 25.5

TOC 45 69.6±0.6 83.2

Pore space

characterisation

SSA by N2 (m2/g)§ 2.48 2.3* 1.79

DA aver. PS (nm) § 1.82 1.72* 1.68

Micro pore volume by

CO2 (cm3/g)# 0.033 0.03 0.044

Ultimate analysis

%C 49.7 ±0.2 74.8 ±0.7 85.4 ±0.2

%H 3.7 ±0.08 5.0 ±0.2 3.3±0.05

%N 1.1 ±0.02 1.5 ±0.02 1.05±0.4

%S 2.8 ±0.05 0.64 ±0.05 0.8±0.1

%Odiff 5.8 ±0.01 18.0 ±0.4 8.7±0.1

Proximate analysis

Moisture (%) 12.7±0.3 2.63±0.2 0.64±0.03

Volatile Matter** (%)

dry, ash-free 52.9±0.2 25.5±0.3 9.61±0.01

Fixed Carbon (%)

dry, ash-free 17.5±0.2 65.6±0.1 85.4±0.1

Ash yield (%) 16.9±0.1 8.0±0.1 4.4±0.04

* analysed after the CO2 sorption test

** heating the coal sample to 900 ± 5 °C for 7 minutes in a muffle furnace # at 273K with CO2 using the Quantachrome AS1 MP § at 77K with N2 using the Quantachrome AS1 MP

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

128 | P a g e

Experimental methods Three different techniques are commonly used to determine gas sorption isotherms on

coals: the manometric (Gensterblum et al., 2009b; Li et al., 2010; van Hemert et al., 2007)

or piezometric (Sircar, 1999) method (TU Delft and RWTH Aachen), the volumetric method

(Fitzgerald et al., 2005; Kelemen and Kwiatek, 2009) and the gravimetric method (Bae and

Bhatia, 2006; Day et al., 2008a; De Weireld et al., 1999; Ottiger et al., 2008; Pini et al., 2006)

(U Mons).

Modern gravimetric sorption devices mostly employ magnetic suspension balances that

allow contactless weighing of samples across the walls of closed high-pressure systems.

The principles of the different techniques are discussed by (Gensterblum et al., 2009b) and

(Goodman et al., 2007a; Goodman et al., 2004). Although these three methods make use of

different physical parameters, they usually provide accurate and comparable sorption

isotherms for simple and well-characterized sorption systems (e.g. methane on activated

carbon or natural coals) (Gensterblum et al., 2009b; Goodman et al., 2004). The primary

experimental parameter obtained from all these procedures is the excess sorption (Gibbs

sorption, Gibbs excess) (Sircar, 1999). A summary of the specifications of the various

setups used in this study is provided in Table 12.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

129 | P a g e

Table 12: Specifications of experimental devices used by different laboratories

RWTH Aachen TU Delft U Mons

Method manometric manometric gravimetric

PCO2_max (MPa) 25 ± 0.0125 20.7±0.001 16±0.016

T (K) 318.6±0.1

318.8±0.1

318.1±0.01

318.1±0.01

318.5±0.1

318.6±0.1

mcoal (g) 4-6±0.0001 23-48 ±0.03 1.2-1.9±0.00002

VSample cell (cm³) 13.085±0.001 78.33±0.06

76.9 ±1 -

VReference cell (cm³) 1.837±0.004

1.269±0.003

12.152±0.009

3.524±0.004 -

Vvoid (cm³)

10. 532±0.008

4.477±0.011

4.467±0.005

61.1±0.1

59.2±0.1 -

Vsystem (crucible +

sample) (cm³) - -

1.575 ±0.002

1.684 ±0.002

Tequilibration (h) 2-4 30

3 1-3

CO2 purity 99.995% 99.990% 99.996%

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

130 | P a g e

Parameterization of the experimental results

Excess sorption, the measurable quantity in volumetric, gravimetric and manometric

procedures, is a net effect arising from (i) transfer of molecules from the free (gaseous or

gas-like supercritical) phase to a condensed (sorbed) phase and (ii) the concomitant

volume increase of the condensed phases (solid + sorbed phase). At low pressures, density

differences between the free gas phase and the adsorbed phase are large and the volume

increase of the condensed phase is negligible. This is the typical situation in low-pressure

sorption experiments. At higher gas pressures (density) this volume effect becomes

significant in the experimental measurements via an increase in condensed phase volume

resulting in a decrease in void volume during the volumetric/manometric experiments and

in an increase in buoyancy during the gravimetric experiments.

The experimental high-pressure CO2 excess sorption isotherms obtained in this study were

approximated by the following excess sorption function:

s or be d

f r e e

abs ol ut ee x c e s s

pTpTspTs

),(1),(),(

[mmol/g] (1a)

The absolute amount of adsorbed substance, ),( pTsabsolut e , must be expressed by an

appropriate sorption function. In the present work the Langmuir function was used:

pK

pspTs

VL

abs ol ut eabs ol ut e

,

),(

(1b)

This is a steady function increasing monotonously with pressure, which is derived from the

concept of a dynamic equilibrium between free and adsorbed molecules. The Langmuir

parameters KL,V and

absolutes and the density of the adsorbed phase

( sorbed) in equation (1a) were fitted simultaneously by a least-square regression. The

Langmuir coefficient KL,V controls the excess sorption function (1b) in the low-pressure

region where the volume of the adsorbed phase is negligible. The density ratio

sorbed

free pT

),( of the free (“gas”) vs. the sorbed phase controls the shape of the isotherm

above the critical pressure when the density of the supercritical CO2 phase increases

rapidly.

For practical purposes, the Langmuir parameters KL,V and

absolutes were fitted first to the

excess sorption values in the pressure range up to 6 MPa, and the starting value for the

density of the adsorbed phase was taken from an extrapolation in the gas density vs. excess

sorption plot. In the second step of the fitting procedure the density of the adsorbed phase

was adjusted to obtain an optimal fit for the high-pressure values. These fitted values for

the Langmuir parameters KL,V and

absolutes and the density of the adsorbed phase were used

as starting values for the final fitting over the whole pressure range.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

131 | P a g e

Experimental Results

The pore space characterisation by low-pressure nitrogen adsorption at 77K reveals a

decrease in the specific surface area and the average pore size (using the Dubinin-Astakov

approach) with increasing rank. These results are in line with those of (Prinz and Littke,

2005b; Prinz et al., 2004).

Velenje lignite

The CO2 excess sorption isotherms measured by the different laboratories on the Velenje

lignite at 318 K are plotted on a linear and a logarithmic pressure axis in figure 35 a and b,

respectively. The isotherms increase with pressure up to a maximum value of

1.77 ± 0.07 mmol g-1 at 6.92 ± 0.6 MPa. Subsequently, the excess sorption decreases with an

inflection point around 10 MPa. This is due to the increasing gas density and therefore, the

increasing influence of the ratio between gas and sorbed phase density.

Table 13 lists the maximum excess sorption values of the individual experiments and the

respective measuring temperature (according to equation 1a and 1b). The deviation

between the isotherms from different laboratories is 1.5% at 3 MPa and 23% at 13 MPa

(Figure 43). The reproducibility of the isotherms is 2.7% and 1.4% at 5 MPa for Aachen and

Mons respectively (Table 8). The helium skeletal density values (ρsample) of

1578 ± 117 kg m-3 for the Velenje sample (Table 13) measured at RWTH Aachen, U Mons

and TU Delft differ by 7.4 %. This could be due to heterogeneities in the sample

composition and residual moisture. Systematic differences in the experimental methods

are unlikely to be the cause for these differences considering the excellent agreement for

activated carbons in an earlier study by the same laboratories (Gensterblum et al., 2009b).

Table 13: Comparison of experimental results for CO2 excess sorption isotherms on Velenje

coal (according to equ. 1a and 1b)

kL,V

(MPa)

S∞,dry

(mmol/g

)

ads,Fit

(kg/m³) s

ads,Plot

(kg/m³)

T

(K)

sample,dry

(kg/m³)

RWTH Aachen 1 2.01 2.77 941 0.07 1199±11 318.5 1763

RWTH Aachen 2 2.07 2.80 (1100*) 0.07 n.c. 318.2 1522

TU Delft 2.56 3.02 1172 0.29 3590±118 318 1581

U Mons 1 2.00 2.68 909 0.07 800±6 318.8 -/-

U Mons 2 1.98 2.79 (1344*) 0.07 n.c. 319.6 1447

average

std. dev.

2.13 2.81 1174 1000 1578

0.22 0.11 163 199 117

(n.c. not calculated, no high-pressure data available; *final pressure not high enough for

good reliability)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

132 | P a g e

Figure 35: CO2 excess sorption isotherms on dry Velenje coal plotted on a linear (a) and a

logarithmic (b) pressure scale.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 5 10 15

Pressure (MPa)

Exce

ss s

orp

tio

n (

mm

ol/g

)UMons 1

UMons 2

TU Delft

RWTH 1

RWTH 2

a)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0.01 0.1 1 10 100

Pressure (MPa)

Exce

ss s

orp

tio

n (

mm

ol/g

)

UMons 1

UMons 2

TU Delft

RWTH 1

RWTH 2

b)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

133 | P a g e

Also included in table 13 are the density values for the sorbed phase derived from

graphical extrapolation of the Gibbs adsorption isotherm in figure 36 (ads,Plot).

Gensterblum et al. (2009) have reported details on the procedure.

The discrepancy for the first two pressure points in the TU Delft isotherm is probably

caused by insufficient equilibration times. The standard drying procedure was applied, but

TU Delft used the largest (~23.65 g) amount of sample. Therefore, the long equilibration

time (>100h) might be due to insufficient moisture removal. These issues are further

discussed below.

Figure 36: CO2 excess sorption (dry basis) versus density of free gas phase for lignite

sample from the Velenje mine.

As shown in figure 36, after passing through a maximum, the Gibbs excess sorption shows a

linear decrease with an increase in gas density gas. The extrapolation of this linear relation

yields an x-axis intercept, where gas becomes equal to ads. The use of this technique

requires sufficient data in the linear (high-pressure) range beyond the maximum value of

the Gibbs adsorption. Estimates for the sorbed phase density obtained through the fitting

procedure (1174 ± 163 kg m-3) and the graphical method (1000 ± 199 kg m-3) are in good

agreement (table 13 through table 14) regarding the error.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 100 200 300 400 500 600 700 800

Gas density (kg m-3

)

Exce

ss s

orp

tio

n (

mm

ol/g

)

UMons 1

UMons 2

TU Delft

RWTH 1

RWTH 2

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

134 | P a g e

Brzeszcze 364 high-volatile bituminous coal

The sorption isotherms for the Brzeszcze 364 sample are plotted in figure 37 and figure 38.

Maximum excess sorption values of 1.37 ± 0.01 mmol g-1 are reached at a pressure of

6.68 ± 0.4 MPa. Subsequently, the excess sorption decreases with an inflection point

around 10 MPa.

The deviation of the isotherms from the different laboratories is 1.1% at 3 MPa and 6.0% at

13 MPa (Figure 43). The intra-laboratory reproducibility of the isotherms is 0.4% at 5 MPa

for both RWTH Aachen and U Mons and 2.6% at 11 MPa for RWTH Aachen (Table 18).

Table 14: Comparison of experimental results for CO2 excess sorption isotherms on Brzeszcze

364 coal sample

kL,V

(MPa)

S∞,dry

(mmol/g)

ads,Fit

(kg/m³)

s ads,Plot

(kg/m³)

T

(K)

sample,dry

(kg/m³)

RWTH Aachen 1 1.49 1.91 1414 0.04 1751 ± 18 318 1390

RWTH Aachen 2 1.49 1.88 1430 0.04 1920 ± 47 317.9 1460

TU Delft 1.47 1.84 1325 0.02 1874 ± 42 318 1476

U Mons 1 1.73 1.87 2100* 0.05 n.c.* 320 1402

U Mons 2 1.46 1.87 1169 0.04 1630

±176

319 1431

average

std. dev.

1.53 1.87 1353 1808 1432

0.1 0.02 107 127 33

(n.c. not calculated; *no high pressure data available)

The helium skeletal density values (ρsample) of the Brzeszcze 364 sample (Table 14) has an

average value of 1432 ± 33 kg m-3. The values measured at RWTH Aachen, U Mons and TU

Delft differ by 2.3 %. Estimates for the sorbed phase density obtained through the fitting

procedure (1301 ± 120 kg m-3) and the graphical method (1783 ± 118 kg m-3) show only

minor differences.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

135 | P a g e

Figure 37: CO2 excess sorption isotherms (dry basis) on Brzeszcze 364 LW plotted on a linear

(a) and a logarithmic (b) pressure scale.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 5 10 15

Pressure (MPa)

Ex

ce

ss

so

rpti

on

(m

mo

l/g

)

RWTH 1

RWTH 2

UMons 1

UMons 2

TU Delft

a)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.0E-02 1.0E-01 1.0E+00 1.0E+01 1.0E+02

Pressure (MPa)

Ex

ce

ss

so

rpti

on

(m

mo

l/g

)

RWTH 1

RWTH 2

TU Delft

UMons 1

UMons 2

b)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

136 | P a g e

Figure 38: CO2 excess sorption isotherms (dry basis) versus density of free gas phase for

bituminous coal sample Brzeszcze 364.

Selar Cornish semi-anthracite

The excess sorption isotherms measured for the Selar Cornish sample are plotted in figure

39 and figure 40. The isotherms of the participating laboratories show an excellent

agreement up to a pressure of 9 MPa. Maximum excess sorption values of

1.37 ± 0.05 mmol g-1 are reached at a pressure of 5.89 ± 0.6 MPa. At pressures above 10

MPa the excess sorption isotherms diverge substantially and final values at 14-15 MPa

range from ~0.6 to ~0.8 mmol/g.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 100 200 300 400 500 600 700 800 900

CO2 gas density (kg m-3

)

Ex

ce

ss

so

rpti

on

(m

mo

l/g

)

RWTH 1

RWTH 2

UMons 1

UMons 2

TU Delft

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

137 | P a g e

Table 15: Comparison of experimental results for CO2 excess sorption isotherms on Selar

Cornish anthracite

kL,V

(MPa)

S∞,dry

(mmol/g)

ads,Fit

(kg/m³)

s ads,Plot

(kg/m³)

T

(K)

sample,dry

(kg/m³)

RWTH Aachen 1 1.29 1.89 1066 0.03 1359 ± 11 317.8 1390

RWTH Aachen 2 1.26 1.95 1058 0.02 1139 ± 29 317.9 1451

TU Delft 1.43 1.91 1363 0.15 1915 ± 27 318.2 1370

U Mons 1 1.22 1.84 1200 0.02 2236 ± 93 319.2 1433

U Mons 2 1.41 1.94 1100 0.01 n.c.* 319.8 1433

average

std. dev.

1.35 1.92 1174 1527 1415

0.07 0.02 135 388 31

(n.c.* not calculated; no high pressure data available)

Table 15 lists the maximum excess sorption values of the individual experiments and the

respective measuring temperature. The deviation between the isotherms from different

laboratories is approximately 1% at 3 MPa and 26% at 12 MPa (figure 43).

The intra-laboratory reproducibility of the isotherms is 1.1% and 3.6% at 3 MPa for U

Mons and RWTH Aachen respectively and 1.5% at 11 MPa for RWTH Aachen (Table 18).

The helium skeletal density values (ρsample) 1415 ± 31 kg m-3 of the Selar Cornish sample

(Table 15) measured at RWTH Aachen, U Mons and TU Delft differ by 2.5 %. Estimates for

the sorbed phase density obtained through the fitting procedure (1174 ± 135 kg m-3) and

the graphical method (1527 ± 338 kg m-3) are in good agreement regarding the error.

The helium skeletal density values of the Selar Cornish anthracite determined in these tests

(Table 15) show slight differences (2.2%). This could be due to variations in the starting

conditions.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

138 | P a g e

Figure 39: CO2 excess sorption isotherms (dry basis) for the Selar Cornish semi-anthracite

sample plotted on a linear (a) and a logarithmic (b) pressure scale.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 5 10 15

Pressure (MPa)

Ex

ce

ss

so

rpti

on

(m

mo

l/g

)

UMons 1

UMons 2

TU Delft

RWTH 1

RWTH 2

a)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.E-01 1.E+00 1.E+01 1.E+02

Pressure (MPa)

Ex

ce

ss

so

rpti

on

(m

mo

l/g

)

UMons 1

UMons 2

TU Delft

RWTH 1

RWTH 2

b)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

139 | P a g e

Figure 40: Excess sorption versus density of free gas phase for Selar Cornish semi-anthracite

sample.

The results of the regressions for the individual experimental excess sorption isotherms

are listed in Table 13 through 15. The quality of the fit is expressed by the parameter s

according to equation (2) (Table 13,14,15). Here N is the number of data points of the

isotherm, and s and sfit are the measured and the fitted excess sorption values of data point

n.

2

1

1 N

f i tssN

s

(2)

Listed in table 16 are the parameter sets that provided the best fits (“Master fits”) of

equation (1) to the excess sorption isotherms measured by the three participating

laboratories (TU Delft, U Mons and RWTH Aachen). Also given in this Table are the average

values of the individual fits.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 100 200 300 400 500 600 700 800 900

CO2 gas density (kg m-3

)

Ex

ce

ss

so

rpti

on

(m

mo

l/g

)

UMons 1

UMons 2

TU Delft

RWTH 1

RWTH 2

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

140 | P a g e

Figure 41: Master isotherms of the a) Velenje b) Brzeszcze and c) Selar Cornish coals

Velenje

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0 2 4 6 8 10 12 14 16 18 20

Pressure (MPa)

Exce

ss s

orp

tio

n (

mm

ol/g

)

a)

Brzeszcze 365

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 2 4 6 8 10 12 14 16 18 20

Pressure (MPa)

Exce

ss s

orp

tio

n (

mm

ol/g

)

b)

Selar Cornish

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 2 4 6 8 10 12 14 16 18 20

Pressure (MPa)

Exce

ss s

orp

tio

n (

mm

ol/g

)

c)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

141 | P a g e

Table 16: Fitting parameters of the master isotherms in comparison with the average

values of the isotherms from table 13,14,15

kL,V

(MPa)

S∞

(mmol/g)

ads

(kg/m³)

Master

fit average

Master

fit average

Master

fit average n

Velenje 2.8 2.81 2.0 2.13 964 1174 0.13

Brzeszcze 364 1.39 1.53 1.82 1.87 1513 1353 0.06

Selar Cornish 1.31 1.35 1.97 1.92 1077 1174 0.27

Prior to this regression calculation the data densities of the experimental curves were

partly homogenized in order to avoid a bias that might result from different data densities

of individual isotherms in certain pressure intervals. As expected, the Langmuir sorption

parameters control the curve fit in the low pressure region, while the sorbate phase density

is important only in the high pressure region.

Figure 42 shows the deviations between measured data and the best fit regression

function.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

142 | P a g e

Figure 42: Normalized deviations of experimental CO2 excess sorption isotherms (318 K) for

a) Velenje b) Breszcze 364 c) Selar Cornish from the best fit vs. pressure

Velenje

-40%

-20%

0%

20%

40%

60%

80%

100%

120%

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0

Pressure (MPa)

No

rma

lis

ed

dev

iati

on

a)

Breszcze 364

-15%

-10%

-5%

0%

5%

10%

15%

0 2 4 6 8 10 12

Pressure (MPa)

No

rma

lis

ed

dev

iati

on

b)

Selar Cornish

-20%

-10%

0%

10%

20%

30%

40%

50%

0 2 4 6 8 10 12 14 16 18

Pressure (MPa)

No

rma

lis

ed

dev

iati

on

c)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

143 | P a g e

Discussion

In the present study, a good agreement of the CO2 excess sorption isotherms was observed

for the low-pressure range (0.1-3 MPa) with deviations of <2%. In the high-pressure range

(3-6 MPa), the isotherms measured in the three laboratories differed by 5%, which is

considered a satisfactory agreement. At even higher gas pressures, e.g. 10 MPa, deviations

of up to 25% were observed. For the three laboratories the deviations between the CO2

excess sorption values (from low to high maturity) were 1.2%, 1.4% and 1.5% at 3 MPa and

26%, 7% and 22% at 13 MPa, respectively. The intra-laboratory reproducibility of CO2

isotherms at 5 MPa was 2.7% and 3.6% at the University of Mons (U Mons) and RWTH-

Aachen University, respectively (table 13,14,15).

Table 17: Maximum CO2 excess sorption values at 318 K and corresponding pressures

Maximum excess sorption

capacity

(mol kg-1)

Corresponding

pressure

(MPa)

Velenje 1.77±0.07 6.89±0.5

Brzeszcze 364 1.37±0.01 6.68±0.3

Selar Cornish 1.37±0.05 5.89±0.6

The maxima of the excess sorption isotherms at 318 K were 1.77 ± 0.07, 1.37 ± 0.01 and

1.39 ± 0.05 mol kg-1 for the Velenje (lignite), Brzeszcze (high-volatile bituminous) and Selar

Cornish (semi-anthracite) coals, respectively (Table 17). Correspondingly the Langmuir

volumes have a minimum at the high-volatile bituminous coal rank level around 1% VRr

(see figure 9b), which is in agreement with observations reported by (Day et al., 2005).

Subsequently, the excess sorption decreases with an inflection point around 10 MPa. This is

due to the increasing gas density and therefore the increasing influence of the ratio

between gas and sorbed phase density. A small shift in the pressure of the maximum excess

sorption is also observed from approximately 6.89±0.5 MPa for the lignite to 6.68±0.4 MPa

for the bituminous coal and to 5.89±0.6 MPa for the semi-anthracite coal. This shift could

be related to the decrease of the Dubinin-Astakhov (DA) pore size with increasing rank.

Adsorption in micropores is energetically much more favourable than on the relatively flat

surfaces of meso- and macropores due to more intense interactions of the molecule with

the surrounding pore walls. Due to force - distance characteristics of adsorption forces in

small pores, the adsorbate interacts with a larger number of solid carbon atoms, as

compared to larger pores. Narrow pores will be filled at lower pressures and therefore a

comparable degree of saturation will be reached at lower pressures.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

144 | P a g e

Figure 43: a) Deviation between the CO2 isotherms at 3 and 13 MPa of the different

laboratories as a function of the volatile matter of the samples b) Langmuir pressure K,L

and maximum excess sorption capacity Smax as a function of volatile matter; error bars

reflect the standard deviations between the fitting parameters

0%

5%

10%

15%

20%

25%

30%

0 10 20 30 40 50 60 70 80

Volatile matter (%)

Devia

tio

n (

%)

low pressure range (3MPa)

high pressure range (13MPa)

a)

0

0.5

1

1.5

2

2.5

3

3.5

0 10 20 30 40 50 60 70 80

Volatile matter (%)

Lan

gm

uir

pre

ssu

re (

MP

a);

Sm

ax (

mm

ol/g

)

K,L

Smax

b)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

145 | P a g e

Comparison to previous inter-laboratory comparisons The inter-laboratory studies on CO2 sorption on natural (Argonne Premium) coals by

(Goodman et al., 2007a; Goodman et al., 2004) have revealed substantial discrepancies

among the results of the participating laboratories. The European inter-laboratory study on

high-pressure CO2 sorption, initiated as a joint attempt to overcome these experimental

problems, has yielded very promising results. The isotherms determined by the

participating laboratories on a FiltraSorb400 activated carbon sample showed an excellent

agreement with inter-laboratory deviations <5% and intra-laboratory reproducibility

variations of <1% (Gensterblum et al., 2009b). The comparison of gravimetric and

manometric procedures shows that both methods provide comparable results.

Table 18: Intra laboratory reproducibility

Corresponding

pressure (MPa) U Mons RWTH Aachen

Velenje 5 2.7% 1.4%

Brzeszcze 364 5

11

0.4%

-/-

0.4%

2.6%

Selar Cornish 3

12

1.1 %

-/-

3.6%

1.5%

For natural coals up to a pressure of 5 MPa, data reproducibility is better than 3% and in

the high-pressure region, better than 4%. This variance is attributed to the heterogeneity of

samples (Table 18). The deviation between the CO2 isotherms of different laboratories at 3

MPa are 1.2%, 1.4% and 1.5% and at 13 MPa 2.6%, 7% and 22%. The deviation between

the CO2 isotherms of different laboratories as a function of volatile matter content passes

through a minimum (Figure 43) and is much lower compared to the deviation of the

participating laboratories in the round robin of (Goodman et al., 2007a; Goodman et al.,

2004). The results of the first (Gensterblum et al., 2009b) and second round robins were

more accurate and reliable than in the calculation of Mohammed and co-workers

(Mohammed S, 2009). This is due the fact that the published used larger uncertainties for

error calculations, then the uncertainties of the measuring devices from the participating

labs of this round robin. In general, the results from (Gensterblum et al., 2009b) and

(Mohammed S, 2009) confirm our statement that CO2 sorption isotherms measured with

manometrical devices need a very accurate method of recording pressure and temperature.

Coal swelling

The excess sorption values in the high-pressure range are dominated by the ratio of the gas

and the sorbed phase density and volumetric effects of the sample. Coal is a very

heterogeneous material because it is composed of various macerals and mineral matter

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

146 | P a g e

(“ash”). Different macerals and ash components show different swelling behaviour (Day et

al., 2008b; Karacan, 2003; Karacan, 2007; Kelemen, 2006; Mazumder, 2005; Milligan,

1997). There is a relationship between coal swelling and the amount of CO2 adsorbed by

the coal. At pressures below a few atmospheres, swelling is low and generally unaffected by

the amount of gas adsorbed. But at elevated pressures, swelling increases nearly linearly

with the amount of CO2 adsorbed (Kelemen, 2006; van Bergen et al., 2009). Above about 8

MPa, this relationship is no longer linear: adsorption continues to increase but the coal

matrix volume remains constant (no swelling) (Kelemen, 2006). Additionally, the intensity

of swelling changes with coal rank (Siemons and Busch, 2007), which may affect the

absolute error of the isotherm in the terms of absolute sorption capacity, but not the

comparability between different laboratories.

Residual moisture

Although other processes may also be of some relevance, residual moisture appears to play

a dominant role in affecting the measured adsorption isotherms of CO2 on dried coals. This

could also explain the rank-dependence of the agreement of inter-laboratory results. In

future studies, a strict procedure for controlling coal moisture content will be required.

Throughout the long history of specific surface area measurements by low-pressure gas

sorption on coal (nitrogen BET, CO2, and others) (Mahajan, 1991; Prinz and Littke, 2005b;

Prinz et al., 2004), it was thought that drying and degassing the coals in the measuring cell

provides the best method for obtaining reproducible isotherms. Drying in the sample cell

prevents remoisturisation, but it requires longer drying times due to slow removal of

moisture from the very thick powder bed. If adsorption isotherms are to be obtained for

dried coals, stricter control on the drying and/or handling conditions will be required for

good inter-laboratory comparisons.

Particle size

It is well-known that smaller particle-size fractions of powdered coals tend to have higher

ash contents than larger particle-size fractions (Busch et al., 2004). Therefore, the optimum

particle size for representative isotherms is generally >100µm (Spears AD, 2002). Particle

size fractionation of coal powders may also occur due to vibrations of the sample container

(e.g. transportation, carrying). This has to be taken into account when filling the sample

cell. This effect is considered of almost no relevance, but should be taken into account when

handling very heterogeneous samples.

Supercritical CO2 as solvent

Supercritical CO2 is an excellent solvent for hydrocarbons (Larsen, 2004). The pore

structure of coal is not a constant state but changes during the coalification process.

Through coalification pore size redistribution occurs and free hydrocarbon (bitumen)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

147 | P a g e

generated as a product of coalification fills part of the available pores and could be

dissolved by supercritical CO2 during isotherm measurement. This effect of dissolving

hydrocarbons is coal rank-dependent (Larsen, 2004) and would also depend on the

moisture content and the primary composition of the coal. For the CO2 sorption isotherms

in this paper extraction it is a minor source of error because no repeat measurements on

the same samples have been performed.

Gas impurities and uncertainties in temperature measurements

Impurities of the measuring gas might be another reason for differences between the

isotherms from the different laboratories. The purity of the CO2 used in the different

laboratories varied between 0.9999 (4.0) and 0.999999 (6.0). Assuming helium as the

impurity and using the equation of state for gas mixtures by (Kunz et al., 2007), at 10 MPa

and 318K the density difference between pure CO2 (502.7 kg m-3) and the mixture amounts

to 0.5 kg m-3 corresponding to a difference of 0.1%.

An error of one degree in the temperature measurement (318K instead of 319K) will result

in a maximum error in calculated CO2 density at 10 MPa. This corresponds to a density

difference of 28.4 kg·m-3 (5.6%) at an absolute gas density of pure CO2 = 502.7 kg·m-3

(Figure 45).

Another possible source of error might be the contamination of CO2 by helium during the

manometric measuring procedure. Thus, prior to the CO2 sorption test the void volume of

the measuring cell is determined by helium in the manometric method. If this helium is not

completely removed from the system by evacuation and flushing with CO2, this may result

in erroneous values for the gas density. Assuming 0.1 MPa of residual He in the sample cell

at the start of the test, a minimum error would occur at 7.1 MPa and a maximum error at

10.2 MPa (Figure 44). This maximum error corresponds to an over-estimation of CO2

density in the cell by 31.9 kg·m-3 (6.1%) at density of 502.7 kg·m-3 for pure CO2.

Figure 45 documents the magnitude of potential effects of experimental errors on CO2

excess sorption isotherms and their influence on the shape of CO2 isotherms. The

uncertainties assumed in the calculations are higher than the maximum uncertainties in

our devices (Gensterblum et al., 2009b). Only one input parameter is varied per calculation.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

148 | P a g e

a)

b)

Figure 44: a) Pure CO2 density for 318K and 319K. b) Density of pure CO2 and a mixture of

CO2 with a residual gas content of 1 bar He at for 318 K. Equations of State were calculated

using the “Software for the Reference Equation of State for Natural Gases (GERG-2004)” of

Ruhr Universität Bochum1 (Kunz et al., 2007)

1 http://www.ruhr-uni-bochum.de/thermo/Software/Seiten/GERG_2004-eng.htm

0

100

200

300

400

500

600

700

800

900

1000

0 5 10 15 20

Pressure (MPa)

Den

sit

y (

kg

/m³)

0

5

10

15

20

25

30

35

Den

sit

y d

iffe

ren

ce (

kg

/m³)

EOS CO2 318K

EOS CO2 319K

dT =1

0

100

200

300

400

500

600

700

800

900

1000

0 5 10 15 20

Pressure (MPa)

Den

sit

y (

kg

/m³)

0

5

10

15

20

25

30

35

Den

sit

y d

iffe

ren

ce (

kg

/m³)

pure CO2

CO2 with residual He

density differences

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

149 | P a g e

Figure 45: Potential effects of experimental errors on CO2 excess sorption isotherms: a) gas

impurities (Kunz et al., 2007) and uncertainties in temperature measurements; b) errors

in reference and void volume determination

0 2 4 6 8 10 12 14 16

Pressure (MPa)

Temperature +1K

residual N 1bar

residual He 1bar

pure CO2

a)

0 2 4 6 8 10 12 14 16

Pressure (bar)

Vref*1.05

Vref*0.95

pure CO2

Vvoid*0.95

Vvoid*1.05

b)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

150 | P a g e

Minimizing experimental errors for manometric methods Based on the results of this study, the following factors are considered critical in

minimizing the uncertainties in the measurement of CO2 excess sorption isotherms on

natural coals:

(1) In the manometric method the ratio of the measuring cell void volume (Vm) to the

reference cell volume (Vr) must be optimized: A larger Vm/Vr ratio results in lower errors

but also in smaller pressure steps and, in consequence, more pressure steps are required to

reach the final pressure. Since errors for each pressure step add up, the total number of

pressure steps should not exceed a value of 20. Within this study, errors for the

manometric setups were estimated as 0.1% to 0.5% for each pressure step, which results

in a total experimental error of 2-10%.

Calculations show that for CO2 isotherms, a Vm/Vr ratio between 1 and 10 is a good

compromise in terms of pressure-step size and minimizing error.

(2) The void volume in the sample cell should be minimized by using a large amount of

sample. The ratio between the initial pressure and the pressure drop due to the sorption

should be as large as possible.

(3) For manometric setups, the pressure range from 7-13 MPa (Figure 44) should be

avoided while pressurising the reference cell. The potential error in this pressure range is

high due to the slope of the equation of state, and it will affect mass balance calculations

substantially. Furthermore, small deviations in temperature will result in high deviations in

density.

(4) Much care should be taken when determining the void volume of the sample cell, as

well as the volume of the reverence cell, because small errors have a large impact on the

mass balance calculation. It is suggested to perform multiple series of helium pycnometry

runs to minimize the error by averaging.

(5) The temperature should be measured by high-precision resistance temperature

detectors (RTD), to avoid absolute errors of 1K associated with the cold junction

compensation of thermocouples. Temperatures should be measured directly within or

close to the sample and reference cells.

Minimizing experimental errors for gravimetric methods Much care should be taken when determining the volume of the sample and gas density

close the critical point, because small errors have an impact on the excess mass calculation

(calculation of the buoyancy effect). It is suggested to perform multiple series of helium

pycnometry runs to minimize the error in sample volume by averaging.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

151 | P a g e

Conclusion

In the present inter-laboratory study, the overall agreement of CO2 excess sorption

isotherms among the different laboratories was good for the Brzeszcze 364 (high-volatile

bituminous) and Selar Cornish (anthracite) coals, especially at low pressures (<6-8 MPa).

The isotherms for the Velenje lignite showed large discrepancies, especially at high

pressures. This rank-dependence of the inter-laboratory agreement is likely to be due to

different residual moisture contents.

The observed excess sorption maxima for the Velenje, Brzeszcze and Selar Cornish coals

are 1.77±0.07, 1.37±0.02 and 1.41±0.02 mol kg-1 in the order of increasing vitrinite

reflectance. A decrease of the pressure at maximum excess sorption has been observed

from lignite to bituminous coal to semi-anthracite. This shift varies from roughly

6.89±0.5 MPa for the lignite, 6.68±0.4 MPa for the bituminous coal, to 5.89±0.6 MPa for the

semi-anthracite. These could be related to the increasing micropore volume.

The deviation between the CO2 isotherms of different laboratories shows a minimum when

related to the volatile matter content (Figure 43).

The deviations between the CO2 isotherms of the different laboratories are higher than the

intra-laboratory deviations. This is a common feature reflecting slight differences in

procedures, in particular drying and sample preparation, but could also be related to

sample heterogeneity. Intra-laboratory comparison is an indication for statistical or

random deviation but may also be affected by sample heterogeneity and their influence on

the isotherms, because no repeat measurements on the same samples have been

performed. The results show that CO2 sorption on natural coals in the supercritical range

can be determined accurately with both gravimetric and manometric equipment but

requires thorough optimisation of instrumentation and measuring procedures as well as

well-defined sample preparation procedures. For moderate-pressure CO2 isotherms (up to

5 MPa), the intra-laboratory reproducibility is nearly equal to the difference between the

different laboratories. For the high-pressure region, some improvements in the

experimental procedure are still needed, especially regarding starting conditions. Small

differences in measuring procedures resulted in deviations of 26%, 7% and 22% (in order

of vitrinite reflectance). Methodological differences among the participating laboratories

are negligible since the intra-laboratory comparison and the inter-laboratory comparison

of previous round robins with activated carbons are very good.

Acknowledgements

This research was conducted in the context of the MOVECBM (Monitoring and Verification

of Enhanced Coalbed Methane) Project supported by the European Commission under the

6th Framework Programme. A major part of the results were obtained within the RECOPOL

and CATO-1 programme. The fruitful discussion and remarks of Prof. Dr. J. Bruining and Dr.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

152 | P a g e

Dirk Prinz are greatly appreciated. The paper benefited from helpful comments and

suggestions by the two reviewers.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

153 | P a g e

Notation

Symbols Units Physical

KL,V [MPa] Langmuir pressure

mtransferred [g] Mass transferred from reverence to sample cell

mcoal,msample [g] mass of coal

mexcess [g] Excess sorption mass

m [g] Incremental apparent sample weight in gravimetric

experiment

n [mmol] amount of substance of gas

p [MPa] Gas pressure

T [K] Temperature

tequil [h] Time taken for equilibration

sexcess [mmol g-1] Excess sorption

sabsolute [mmol g-1] Absolute sorption

absolutes

[mmol g-1] Langmuir sorption capacity (“Langmuir volume”)

VR , Vref [cm³] reference cell volume

Vm, Vvoid [cm³] void volume of sample cell

Vsample [cm³] Skeletal volume

i,CO2: [g cm-3] Free gas density of CO2 to the time i , e=equilibrium

sample: [g cm-3] Skeletal density of the sample

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

154 | P a g e

References

Bae, J.S. and Bhatia, S.K., 2006. High-Pressure Adsorption of Methane and Carbon Dioxide

on Coal. Energy and Fuels, 20: 2599-2607.

Busch, A., Gensterblum, Y. and B.M., K., 2003a. Methane and CO2 sorption and desorption

measurements on dry Argonne Premium Coals: Pure components and mixtures.

International Journal of Coal Geology, 55: 205-224.

Busch, A., Gensterblum, Y., Krooss, B.M. and Littke, R., 2004. Methane and carbon dioxide

adsorption-diffusion experiments on coal: upscaling and modeling. International

Journal of Coal Geology, 60: 151-168.

Busch, A., Gensterblum, Y., Krooss, B.M. and Siemons, N., 2006. Investigation of high-

pressure selective adsorption/desorption behaviour of CO2 and CH4 on coals: An

experimental study. International Journal of Coal Geology, 66: 53-68.

Busch, A., Krooss, B.M., Gensterblum, Y., van Bergen, F. and Pagnier, H.J.M., 2003b. High-

pressure adsorption of methane, carbon dioxide and their mixtures on coals with a

special focus on the preferential sorption behaviour. Presented at Geofluids IV, May

12-16, Utrecht., J. Geochem. Expl., 78-79: 671-674.

Day, S., Duffy, G., Sakurovs, R. and Weir, S., 2008a. Effect of coal properties on CO2 sorption

capacity under supercritical conditions. International Journal of Greenhouse Gas

Control, 2: 342-352.

Day, S., Fry, R. and Sakurovs, R., 2008b. Swelling of Australian coals in supercritical CO2.

International Journal of Coal Geology, 74: 41-52.

Day, S.J. et al., 2005. Factors controlling CO2 absorption in Australian coals, Greenhouse Gas

Control Technologies 7. Elsevier Science Ltd, Oxford, pp. 2221-2224.

De Weireld, G., Frere, M. and Jadot, R., 1999. Automated determination of high-temperature

and high-pressure gas adsorption isotherms using a magnetic suspension balance.

Meas. Sci. Technol., 10: 117–126.

Fitzgerald, J.E. et al., 2005. Adsorption of methane, nitrogen, carbon dioxide and their

mixtures on wet Tiffany coal. Fuel, 84: 2351-2363.

Gensterblum, Y. et al., 2009. European inter-laboratory comparison of high pressure CO2

sorption isotherms. I: Activated carbon. Carbon, 47: 2958-2969.

Goodman, A.L. et al., 2007. Inter-laboratory comparison II: CO2 isotherms measured on

moisture-equilibrated Argonne premium coals at 55°C and up to 15 MPa.

International Journal of Coal Geology, 72: 153-164.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

155 | P a g e

Goodman, A.L. et al., 2004. An Inter-laboratory Comparison of CO2 Isotherms Measured on

Argonne Premium Coal Samples. Energy and Fuels, 18: 1175-1182.

Karacan, C.Ö., 2003. Heterogeneous Sorption and Swelling in a Confined and Stressed Coal

during CO2 Injection. Energy and Fuels, 17: 1595-1608.

Karacan, C.Ö., 2007. Swelling-induced volumetric strains internal to a stressed coal

associated with CO2 sorption. International Journal of Coal Geology, 72: 209-220.

Kelemen, S.R. and Kwiatek, L.M., 2009. Physical properties of selected block Argonne

Premium bituminous coal related to CO2, CH4, and N2 adsorption. International

Journal of Coal Geology, 77: 2-9.

Kelemen, S.R., Kwiatek, L.M., Lee, A.G.K., 2006. Swelling and Sorption Response of Selected

Argonne Premium Bituminous Coals to CO2, CH4, and N2, International CBM

Symposium, Tuscaloosa, Alabama.

Kunz, O.K., R., Wagner, W. and Jaeschke, M., 2007. The GERG-2004 wide-range reference

equation of state for natural gases. GERG Technical Monograph 15 Fortschritts-

Berichte Verein Deutscher Ingenieue (VDI), 6. VDI-Verlag, Düsseldorf.

Larsen, J.W., 2004. The effects of dissolved CO2 on coal structure and properties. Int J. Coal

Geology, 57: 63– 70.

Li, D. et al., 2010. High-pressure sorption isotherms and sorption kinetics of CH4 and CO2

on coals. Fuel, 89: 569-580.

Mahajan, O.P., 1991. CO2 surface area of coals: A 25 year paradox. Carbon, 29: 735-742.

Mazumder, S., Siemons, N., Wolf, K.-H., 2005. Differential swelling and permeability change

of coal in response to CO2 injection for ECBM. SPE, 98475.

Mazumder, S., van Hemert, P., Busch, A., Wolf, K.H.A.A. and Tejera-Cuesta, P., 2006. Flue gas

and pure CO2 sorption properties of coal: A comparative study. International Journal

of Coal Geology, 67: 267-279.

Milligan, J.B., Thomas, K. M., Crelling, J. C., 1997. Solvent Swelling of Maceral Concentrates.

Energy and Fuels, 11: 364-371.

Mohammed S, J.F., RL Robinson Jr, KAM Gasem, 2009. Experimental uncertainties in

volumetric methods for measuring equilibrium adsorption. Energy and Fuels, 23:

2810-2820.

Ottiger, S., Pini, R., Storti, G. and Mazzotti, M., 2008. Competitive adsorption equilibria of

CO2 and CH4 on a dry coal. Adsorption, 14: 539-556.

Pini, R., Ottiger, S., Rajendran, A., Storti, G. and Mazzotti, M., 2006. Reliable measurement of

near-critical adsorption by gravimetric method. Adsorption, 12: 393-403.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

156 | P a g e

Prinz, D. and Littke, R., 2005. Development of the micro- and ultramicroporous structure of

coals with rank as deduced from the accessibility to water. Fuel, 84: 1655-1662.

Prinz, D., Pyckhout-Hintzen, W. and Littke, R., 2004. Development of the meso- and

macroporous structure of coals with rank as analyzed with small angle neutron

scattering and adsorption experiments. Fuel, 83: 547-556.

Siemons, N., Bruining, H., Castelijns, H. and Wolf, K.-H., 2006. Pressure dependence of the

contact angle in a CO2-H2O-coal system. Journal of Colloid and Interface Science, 297:

755-761.

Siemons, N. and Busch, A., 2007. Measurement and interpretation of supercritical CO2

sorption on various coals. International Journal of Coal Geology, 69(4): 229-242.

Sircar, S., 1999. Gibbsian Surface Excess for Gas Adsorptions Revisited. Ind. Eng. Chem. Res.,

38: 3670-3682.

Spears AD, C.B., 2002. The composition of size-fractionated pulverised coal and the trace

element associations. Fuel, 81: 683-690.

van Bergen, F., Spiers, C., Floor, G. and Bots, P., 2009. Strain development in unconfined

coals exposed to CO2, CH4 and Ar: Effect of moisture. International Journal of Coal

Geology, 77: 43-53.

van Hemert, P., Wolf, K.-H.A.A. and Bruining, J., 2007. The Intrinsic Reliability of

Manometric Sorption Apparatus Using Supercritical Carbon Dioxide, SPE Annual

Technical Conference and Exhibition, Anaheim, CA, pp. 11.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

157 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

158 | P a g e

"Any intelligent fool can make things bigger, more complex, and more violent.

It takes a touch of genius -- and a lot of courage -- to move in the opposite direction."

(Albert Einstein)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

159 | P a g e

4. Molecular concept of H2O, CH4 & CO2

adsorption on organic material

Fuel (2014), Vol 115 (2), page 581-588

Yves Gensterblum, and Bernhard M. Krooss RWTH-Aachen University, Institute of Geology and Geochemistry of Petroleum and

Coal, Lochnerstr. 4-20, D-52056 Aachen, Germany

Andreas Busch

Shell Global Solutions International B.V., Kessler Park 1, 2288GS Rijswijk, The

Netherlands

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

160 | P a g e

Abstract

Unconventional gas, such as shale gas or coalbed methane offers an attractive low-

carbon solution and furthermore provides possibilities for CO2-storage and coevally for

enhanced gas recovery. In order to better understand gas and water interaction with

organic matter (natural coal) of different maturity we developed a molecular concept

with experimental and literature support for sorption of CH4, CO2 and H2O on organic

material over a broad range of thermal maturity (0.5~3.3% vitrinite reflectance).

We present here a conceptual model to explain CO2 and CH4 sorption in the presence of

water on coal with varying coal maturity (from sub-bituminous to anthracite).

Adsorption experiments have been performed on different maturity coals at

temperatures between 303 and 350 K, pressures up to 20 MPa and under dry and

moisture-equilibrated conditions. With increasing coal maturity we find for both gases a

linear sorption capacity trend for the moist and a more parabolic trend for the dry coal

samples. When investigating the difference in CH4 and CO2 sorption capacity on coal of

different maturity as a function of moisture content we infer that oxygen containing

functional groups account for the primary sorption sites where the competition between

H2O and CO2 or CH4 takes place. The competitive interaction turns out to be a volumetric

displacement from the gas type independent. A pore blocking mechanism could not be

affirmed. Adsorbed molecules on anthracite are mobile within the adsorbed phase at

low surface coverage. Additionally restrictions in translational and vibrational

movements of the sorbed gas molecules induced by adsorbed water molecules are

observed. Therefore we conclude that sorbed molecules are more localised when water

is present in the adsorbed phase. Whereas at high surface coverage, the thermodynamic

properties of adsorbed molecules are dominated by adsorbate-adsorbate interactions.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

161 | P a g e

Article

Adsorption of supercritical gases such as methane and carbon dioxide on coals and

shales under geological reservoir conditions occurs in the presence of water. The

amount of experimental data on gas sorption under these conditions is still very limited

and the fundamental processes are only partly explored. This paper aims at shedding

light on this issue, based on new data sets and data presented previously.

The interaction of water with natural coal is more complex than the interaction of non-

polar gases like helium, methane or carbon dioxide. This complexity is due to the weak

dispersion interaction of water with coal, the tendency of water to form hydrogen bonds

with other sorbed water molecules and surface-chemical species, and the physisorptive

interaction with the coal mineral matter (s. ref. (Busch and Gensterblum, 2011) for a

more comprehensive discussion).

The sorption of water vapour and methane on coal surfaces is controlled by polar (e.g.

carboxylic and hydroxylic) functional groups (Given, 1986; Gutierrez-Rodriguez, 1984;

Mu and Malhotra, 1991; Nishino, 2001). Already in 1936, Coppens reported a decrease

in methane sorption capacity due to the presence of moisture in several Belgian

Carboniferous coals (Coppens, 1936). Polar sites on the coal surface are preferentially

occupied by water and, hence reduce the capacity for CO2 and CH4 (and other gases).

Therefore moist coals have a significantly lower maximum sorption capacity for either

gas than dry coal. However, the extent to which the capacity is reduced depends on coal

rank. Coals of higher rank are less affected by the presence of moisture than low rank

coals (Day et al., 2008c). Gas sorption capacity decreases with increasing moisture

content up a certain level of water saturation. Above this saturation level the gas

sorption capacity remains constant (Day et al., 2008c; Joubert et al., 1973b; Joubert et al.,

1974). On the molecular level the coal surface can be envisaged to consist of sets of

rather water-prone (hydrophilic) and rather gas-prone (hydrophobic) sorption sites and

there is a partial overlap of these sets of sorption sites.

The gas sorption capacity of dry coals decreases with increasing rank up to a vitrinite

reflectance of 1.1-1.3% (medium volatile bituminous coal) and increases subsequently.

In contrast to this parabolic (“u-shaped”) trend, the gas sorption capacity of moist coals

increases linearly with rank (Busch and Gensterblum, 2011; Mosher et al., 2013;

Ozdemir and Schroeder, 2009).

This contribution combines single published observations with the comprehensive set of

isotherms measured by the authors (Gensterblum et al., 2013a) to demonstrate how the

competition between preadsorbed water and CO2 or CH4 can be envisaged on a

molecular level.

Thermodynamics

The van 't Hoff equation relates the equilibrium constant K of a physico-chemical

(reaction) equilibrium to a change in temperature T through the standard enthalpy

change ΔH for the process:

or

(28)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

162 | P a g e

Here R is the universal gas constant

The equation is approximate in that both enthalpy and entropy changes of a process

(e.g., sorption or diffusion) are assumed to be constant with temperature. The equation

provides a better fundamental physical understanding of the bonding characteristics.

characteristics (Ruthven, 1984). From the definition of the Gibbs energy or free

enthalpy:

(29)

HereS is the standard entropy. The Gibbs free energy of a real gas is related to its

fugacity by:

(30)

The Henry coefficient is defined as the slope at zero pressure. Further we use as

reference the fugacity of f0=1MPa.

(31)

We use the slope from the origin of the isotherm to the first pressure point as the best

experimental approximation.

Combining the Gibbs energy and the definition of the Henry coefficient KH and the

Langmuir pressure pL results in equation 4 and 5:

(32)

(33)

Where the Henry constant represents the CH4 and CO2 sorption behaviour at low surface

coverage. The Langmuir pressure pL is a function of thermodynamic adsorption

parameters and the enthalpy H is equal to the heat of adsorption qst when the

experimental system is isothermal (Xia et al., 2006). The standard entropy is segmented

in the entropy of the gas Sgas at standard pressure (0.10132 MPa) and entropy of the

adsorbed phase Sads: (34)

Entropy Entropy values derived from experimental data are a combination of the translation

, vibration

and rotation entropies following (Ruthven, 1984):

(35)

It is common practice to treat adsorbed molecules thermodynamically as a 2-

dimensional (2D) gas, with molecules moving along the 2D surface in all directions. This

2D gas is thermodynamically parameterised through e.g. the spreading pressure and

surface area, etc. The mean free path in this 2D movement is far smaller and the density

of the adsorbed phase higher than in a 3D gas:

(36)

where S3Dtrans is the translation entropy in 3D, M the molar mass, and “v” the volume

occupied by each molecule (Xia et al., 2006).

When the Lennard-Jones potential energy trough is low and a vibration of low frequency

replaces perpendicular translational motion, then the entropy change upon adsorption

is less than what would be expected in case the movement is restricted to the 2D surface.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

163 | P a g e

This type of adsorption was denoted as “supermobile”(Kemball et al., 1950). There are

little considerable restrictions on the freedom of the vibration:

(37)

where is the 2D translation entropy :

(38)

Here is the translation entropy in 3D and T is temperature(Gregg, 1967). When

the entropy of adsorption is roughly equal to the entropy change on losing the degree of

translational freedom (S3D - S2D) is termed “mobile” adsorption (Kemball et al., 1950). It

does not imply that no vibration of the molecules occurs with respect to the surface.

When the strength of adsorption is higher, the molecules are more restricted. Localized

physical adsorption occurs when the interaction between the adsorbed molecules and

the adsorbent surface is very strong and the kinetic energy of the adsorbed molecules is

low (Kemball et al., 1950). The molecules are virtually kept at fixed positions on the

surface apart from weak vibrations. An adsorbate molecule is considered to be localized

when it is at the bottom of a potential well whose depth is far greater than the thermal

energy of the molecule (Kemball et al., 1950).

Results and Discussion

In this study we investigate the effect of coal rank on the sorption enthalpy and entropy.

Therefore we use a differential approach by comparing CH4 and CO2 sorption isotherms

on dry and moisturised coal samples. For this effort we selected on three coal samples of

different coal rank: one high-volatile bituminous (hvb) and one anthracite coal sample

from the German Ruhr Basin, and one low rank (sub-bituminous) coal from the Surat

Basin, Australia. Details on the geological setting for all three coal samples have been

described earlier (Gensterblum et al., 2013a). Coal properties and adsorption data are

provided in the supplementary information.

Sorption isotherms for CO2 and CH4 have been measured on dry and moisture-

equilibrated coal samples at temperatures between 303 and 350K. The 36 sorption

isotherms and technical and methodical details are reported elsewhere (Gensterblum et

al., 2013a).

Figure 46 infers that coal samples used in this study are comparable to other datasets in

their adsorption properties. A comparison on the basis of surface chemistry properties

would be more preferable; however this type of characterisation is very rare.

All isotherms show a decrease in excess sorption capacity following the specific isosters

with increasing temperature due to the exothermic nature of the adsorption process

(Figure 46). It is not surprising that almost all isotherms for dry coals have significantly

higher excess sorption capacities than for moist coals. Furthermore, all CO2 isotherms

show higher excess sorption capacities compared to their corresponding CH4 isotherms.

Experimental adsorption data have been fitted to a temperature-dependent Langmuir

model (Equation 36) which interprets the adsorbed phase as an adsorbed layer which is

not able to provide additional sorption sites (Langmuir, 1932):

(39)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

164 | P a g e

This is commonly expressed in the way that sorption is restricted to a monolayer. The

surface is assumed to be energetically (chemical composition) and structurally (pore

size distribution) homogeneous. Therefore, the main assumption is that no energy

distribution of the sorption sites exists. All sorption sites are equal in both, density and

sorption enthalpy. Further, it is assumed that there is no interaction between the sorbed

molecules in the sorbed phase.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Me

tha

ne

so

rpti

on

cap

aci

ty (

mm

ol/

g)

Vitrinite reflectance VRr (%)

dry moist

moist - this study dry - this study

Methane on moist coalsnmoist = 0.303VRr + 0.004

R² = 0.689

h.v.b.lignite/subbit.

m.v.b. l.v.b. semi-anthracite / anthracite

Methane on dry coalsndry = 0.098VRr

2 - 0.210VRr + 0.753R² = 0.58

A)

nMOIST = 0.1985VRr + 0.6282R² = 0.37

nDRY = 0.1388VRr2 - 0.3997VRr + 1.5434

R² = 0.23

0

0.5

1

1.5

2

2.5

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

CO

2so

rpti

on

cap

acit

y (m

mo

l/g)

Vitrinite reflectance VRr (%)

moist dry

moist - this study dry - this study

h.v.b.lignite/subbit.

m.v.b. l.v.b. semi-anthracite / anthracite

B)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

165 | P a g e

Figure 46: Methane (a) and carbon dioxide (b) sorption capacity at reference pressure of 5

MPa as a function of coal rank for moist and dry coals. Additional isotherm data are

used from previous studies(Busch and Gensterblum, 2011) 1. (c) CO2/CH4 sorption ratio

for different moist and dry coal samples at various temperatures as a function of coal

rank. Values were selected between 1 and 5 MPa. Isotherm data are taken from Ref.

(Busch and Gensterblum, 2011)1. Ratios from this study are taken at 318K on dry and

moist coal samples at 1 MPa (small symbols) and 5 MPa (large symbols).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

166 | P a g e

The excess sorption isotherms are approximated by the following equation:

(40)

Here n represents the Langmuir volume, p is the gas pressure, PL(T) and are the

temperature-dependent Langmuir pressure and adsorbed phase density. Usually this

expression with three adjustable parameters provides an accurate fit to the

experimental data. The following discussion will show that from a numerical

approximation point of view the choice is appropriate (very low error sum), but several

conclusions from this comprehensive set of isotherms are not appropriate with the

fundamental assumptions (e.g. energetically (chemical composition) and structurally

(pore size distribution) homogeneous) of the Langmuir model.

It is observed that the Langmuir pressure is generally higher for moist than for dry coals,

and it generally decreases with coal rank. As expected, CH4 and CO2 Langmuir pressures

on dry coals decrease with increasing maturity. This means with increasing coal

maturity lower pressures are needed to achieve equal surface coverage. This trend is

also valid for sorption of CO2 and CH4 on moist coals.

The Langmuir sorption amount n is always higher for CO2 than for CH4. In addition, the

difference between the different Langmuir sorption amounts n, as observed for moist

and dry low rank coals, becomes smaller with increasing with increasing rank (Figure

46a,b) (Busch and Gensterblum, 2011).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

167 | P a g e

Figure 47: The difference of CH4 and CO2 Langmuir volume n between dry and moist coals

as a function of oxygen content. (a) shows the absolute and (b) the relative reduction

in sorption capacity due to the presence of water. This indicates that the amount of

oxygen is proportional to the reduction in sorption capacity due to sorbed water. In

addition, the reduction is, in relation to the total amount of sorption sites, the some

and independent of the type of gas. This indicates that the reduction in sorption

capacity is only caused by the water-occupied oxygen-containing functional groups.

nCH4 = 0.0108cOxygen - 0.1179R² = 0.9737

nCO2 = 0.0242 cOxygen - 0.0827R² = 0.9894

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 5 10 15 20 25 30 35 40

CH

4, C

O2

Lan

gmu

ir s

orp

tio

n c

apac

ity

(nd

ry-n

mo

ist)

(m

mo

l g-1

)

Oxygen content (%)

CH4

CO2

A)

dnCH4 = 0.010coxygen - 0.103R² = 0.97

dnCO2 = 0.010coxygen - 0.015R² = 1.0

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 5 10 15 20 25 30 35 40no

rmal

ised

CH

4, C

O2

Lan

gmu

ir s

orp

tio

n c

apac

ity

(nd

ry-n

mo

ist)

/nd

ry

Oxygen content (%)

CH4

CO2

B)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

168 | P a g e

In most cases the dry/moist sorption capacity ratio for CH4 and CO2 increases with

increasing pressure which is equivalent to an increasing surface coverage and its

difference increases with increasing pressure (Figure 46) (Gensterblum et al., 2013a).

Therefore, the influence of water on sorption capacity is higher at low compared to high

surface coverage. The first sorption sites occupied are the highest energetic ones and

preferably occupied by water, because of its dipole moment favouring a relatively strong

interaction between the water molecule and this part of the coal surface. These high-

energy sorption sites are commonly attributed to oxygen-containing functional groups

(Figure 47) (Svábová et al., ; Svábová et al., 2012). Fourier Transformation Infrared

Spectroscopy (FTIR) studies confirm high amounts of oxygen-containing functional

groups (e.g., hydroxylic and carboxylic groups) for the sub-bituminous coal, lower values

for the hvb coal and no such groups for the anthracite (Appendix figure A1). This trend

is consistent with literature (van Krevelen, 1993). As mentioned, from molecular

dynamic simulation it is known that carboxylic- (-COOH) followed by hydroxylic groups

(-OH) are most likely occupied by CO2 (Liu and Wilcox, 2012a; Tenney and Lastoskie,

2006).

These functional groups are also preferential adsorption sites of water molecules.

Furthermore oxygen-containing functional groups increase the CO2 density in graphitic

slit pores and induce an efficient packing pattern for more efficient usage of the pore

space by CO2 when carboxylic group (-COOH) followed by hydroxylic groups (-OH) are

present (Liu and Wilcox, 2012a). It is therefore evident that the main competition

between gas and water molecules will be at low surface coverage (i.e. low pressures)

(Gensterblum et al., 2013a). For both gases, CH4 and CO2 we observe a clear decreasing

influence of water with increasing coal rank. Figure 47b clearly illustrates the linear

volumetric displacement of CH4 and CO2 by water as a linear function of oxygen content,

while oxygen content is inversely proportional to coal rank and decreases from 31% to

17% and 7% for the three coal sample studied here. This linear correlation between

sorption capacity reduction by the pre-adsorbed water and the oxygen content in Figure

47 is in disagreement with the published hypothesis of sorbed water on oxygen-

containing functional groups blocking nanopores, hence leading to a reduced gas

sorption capacity (e.g. (Narkiewicz and Mathews, 2009; Ozdemir and Schroeder, 2009)).

Because, otherwise, the linear dependence would depend on the enclosed pore volume

which would be specific for each sample and therefore would not lead to a linear

dependency as in Figure 47 (Joubert et al., 1974).

When pre-adsorbed water is present then the adsorbed molecules are localised

independently of coal rank or gas type. Figure 47b illustrates that the mechanism is a

gas type independent volumetric displacement of CH4 or CO2 by water (Day et al.,

2008c). Similar pore filling approach was suggested by Sakurovs et al. 2008. One CH4

molecule or 2.2 CO2 molecules are displaced by water on the sorption sites attributed to

functional groups. In maximum (sub-bituminous coal), 30% of CO2 and 25% of the CH4

total sorption sites are affected by preadsorbed water (Figure 47b). However, 6% of the

CO2 total sorption sites are affected by preadsorbed water, but very likely not attributed

to the oxygen containing functional groups (hydroxylic and carboxylic) (Gensterblum et

al., 2013a).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

169 | P a g e

It is observed that the CO2/CH4 sorption ratios at low surface coverage are generally

higher for moist than for dry coals. These differences between dry and moist samples at

low surface coverage however are not observed for the anthracite. In general the

amount of functional groups decreases and the carbon content increases with increasing

coal rank (van Krevelen, 1993). Therefore the functional groups act as primary (high

energetic) sorption sites for water or, if no water is present, for CO2 and CH4. However,

these oxygen-containing functional groups roughly take up twice as many CO2 as CH4

molecules. This becomes clear when dividing the different slopes of the trend lines in

Figure 47a for the CO2 and CH4 sorption capacity reduction in the presence of water

(Figure 47). This 2:1 ratio is also shown in the CO2/CH4 sorption capacity ratio for dry

coals in Figure 46c. Therefore, the pore radius contribution is of lesser importance for

low rank coals due to the presence of functional groups. Increasing the surface density of

oxygen-containing functional groups increases the CO2 adsorption capacity and lowers

the pore-filling pressure (Tenney and Lastoskie, 2006). This results in a high impact of

the pore size distribution for anthracites because most of the functional groups are

decomposed during coalification.

In Figure 46c we observe an exponential decay of the CO2/CH4 sorption capacity ratio

with coal rank when water is pre-adsorbed. At first sight this is not in agreement with

the expectation that the sorption capacity for CO2 is more affected by pre-adsorbed

water than the CH4 sorption capacity. Because we are only able to compare the

maximum excess sorption capacity at 5MPa, affinity changes related to different surface

coverage have to be considered as well. The CH4 Langmuir pressure is much more

affected by pre-adsorbed water than the one of CO2. Therefore we can assume that the

affinity shift to higher pressure to achieve equal surface coverage (comparing dry and

moist coals) causes the exponential decay of the CO2/CH4 sorption capacity ratio in

Figure 46c.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

170 | P a g e

Figure 48: Langmuir sorption capacity and Langmuir pressure as a function of maturity

(vitrinite reflectance). The Langmuir sorption capacity is calculated according to

equation 40.

0

0.5

1

1.5

2

2.5

0 0.5 1 1.5 2 2.5 3 3.5

Lan

gmu

ir s

orp

tio

n c

apa

city

(m

mo

l/g D

AF)

Vitrinite reflectance (%)

CH4 - dry

CH4 - moist

CO2 - dry

CO2 - moist

0

2

4

6

8

10

12

0 0.5 1 1.5 2 2.5 3 3.5

CH

4, C

O2

Lan

gmu

ir p

ress

ure

(MP

a)

Vitrinite reflectance (%)

CH4 - dry

CH4 - moist

CO2 - dry

CO2 - moist

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

171 | P a g e

Low-rank, sub-bituminous coals have high CO2 and CH4 sorption capacities, which is

attributed to the presence of oxygen-containing functional groups achieving higher

packing densities for CO2 and CH4 by space-efficient adjustments on the pore surfaces

(Liu and Wilcox, 2012a). These oxygen-containing functional groups are preferentially

occupied by water in moist coal due to its polarity. For CO2 and CH4 which have no

permanent dipole moment, adsorbed phase densities are considered to be lower(Liu

and Wilcox, 2012a) Therefore we observe experimentally lower adsorbed phase

densities and lower sorption capacities for gases on low rank coals in the presence of

water (Gensterblum et al., 2013a). The higher CO2/CH4 ratios observed at low surface

coverage can be explained by a stronger ordering of CO2 as compared to CH4 (Heuchel,

1999). Due to the high enthalpy of sorption between the sorptives CO2 and CH4 and the

functional groups, and additionally the enhanced packing of CO2 molecules caused by

oxygen-containing functional groups, lower partial pressures are required to reach a

certain surface coverage. Therefore, when water occupies these particular sorption sites,

higher Langmuir pressures pL are observed for CO2 and CH4 (Figure 48). The

predominant influence of functional groups on water and gas sorption on low rank coals

is considered to be caused by the higher average pore diameters for low rank coals (Liu

and Wilcox, ; Prinz and Littke, 2005a).

During coalification the surface density of oxygen-containing functional groups

decreases due to decarboxylation and dehydration reactions (van Krevelen, 1993). This

explains the minimum of the CH4 and CO2 sorption capacity vs. rank trend at

intermediate coal rank. Further, it explains the change to a linear trend as well as the

small changes at high rank coals when water is already sorbed on the surface of coal

micropores. The progressive coalification process generates additional micro and ultra-

micro porosity (Prinz and Littke, 2005b; Sakurovs et al., 2012) by orientation of the

aromatic rings. The increasing sorption capacity for high rank coals can be attributed to

the overlapping Lennard-Jones potentials in the ultra-micropores (strong pore wall –

wall interactions) and the polarizability per carbon atom becomes larger as the aromatic

rings become larger (Behar and Vandenbroucke, 1987). This is due to the delocalization

of the electrons (Schuyer et al., 1953); an increase in adsorbed phase density and

sorption capacity of the gases with rank can be observed irrespective of the presence of

water (Liu and Wilcox). Consequently, the differences between dry and moist samples at

low surface coverage diminish and are generally much smaller in anthracites.

Furthermore, due to overlapping Lennard-Jones potentials in the micropores, lower

partial pressures are necessary to completely fill these pores, resulting in the observed

low Langmuir pressures for CH4 and CO2 for high rank coals such as anthracites. Figure

48b shows the decreasing differences in Langmuir pressure between dry and

moisturized coals with increasing maturity. This can be explained by the absence of

induced efficient packing pattern by the oxygen containing functional groups (Liu and

Wilcox, 2012a).Whereas for the high rank coal (such as anthracite) the sorption

properties are predominated by micro pore formed of oriented large aromatic rings.

Based on the observation that at high pressures (equivalent to high surface coverage)

the difference between the samples of different maturity reduces, it is concluded that the

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

172 | P a g e

carbon-CO2 or carbon-CH4 interaction dominates the growth of sorption layers. This is in

line with molecular simulations. Thus, it is concluded that the influence of water on the

individual sorption capacities decreases with increasing coal rank, while the difference

in the CO2/CH4 sorption capacity ratio between the dry and moist samples decreases

with increasing pressure.

The effect of moisture on the adsorption capacity for other species such as activated

carbons; however, there is limited information available for the effect of packing of

water molecules in confined pores (Billemont et al., 2011; Billemont et al., 2013). Such

enhanced entropic configurations will have a notable effect on water adsorption and

subsequent access of the pore space. The CH4 or CO2 isotherms did not show CH4 or CO2

hydrates (no sorption isotherm of IUPAC type VI, the isotherms used in this study are of

type 1) or ice like structures. Therefore we can conclude these effects are negligible or at

least of minor importance. However, water clustering or ice like structures of water

would indeed influence the thermodynamic parameters and the uptake kinetics, but the

published data basis is limited and therefore at this point a valuable entropic discussion

on the influence of preadsorbed water configurations is premature.

In order to improve the fundamental understanding of the CH4 and CO2 sorption

processes in the presence of moisture, we will move the focus from more

phenomenological observations to thermodynamics. Adsorbed molecules are

thermodynamically treated as 2D gases, with molecules moving along the sorbent

surface (2D) in all directions (Do, 1998). The mean free path length for these molecules

is far smaller and the density of the adsorbed phase is significantly higher than for a 3D

gas. When the entropy change upon adsorption is less than what would be expected if

the movement is restricted to the 2D surface, the Lennard-Jones potential is low and

perpendicular translational motions are replaced by low-frequency vibrations. This type

of adsorption is generally defined as “mobile”(Kemball et al., 1950). Localized physical

adsorption occurs when the interaction between the adsorbed molecules and the

adsorbent surface is strong and the kinetic energy of the adsorbed molecules is low

(Kemball et al., 1950). The molecules are virtually restricted to defined positions on the

surface apart from weak vibrations.

It is obvious from Figure 49 that the adsorption entropies and enthalpies of CH4 or CO2

on moist coals show only small differences and further no trend with rank. This

indicates that the selective adsorption properties are blocked by water and is

particularly the case for CH4 where the pore size distribution is energetically of lesser

importance.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

173 | P a g e

-35

-30

-25

-20

-15

-10

-5

0

0.000 0.020 0.040 0.060 0.080 0.100 0.120

Iso

ste

ric

he

at o

f CH

4&

CO

2ad

sorp

tio

n (k

J m

ol-1

)

Entropy (kJ mol-1 K-1)

CH4 -dry- VRr=0.5% CH4 -dry- VRr=0.9% CH4 -dry- VRr=3.3%

CH4 -moist- VRr=0.5% CH4 -moist- VRr=0.9% CH4 -moist- VRr=3.3%

CO2 -dry- VRr=0.5% CO2 -dry- VRr=0.9% CO2 -dry- VRr=3.3%

CO2 -moist- VRr=0.5% CO2 -moist- VRr=0.9% CO2 -moist- VRr=3.3%

Kerogen types (Zhang et al. 2012)

localisedmobile

A) q<0.3

D

trans

D

transtrans SSS 23

CH4 CO2

I

III

II

CH4

CO2

subbituminous high-volatile bituminous anthracitedrymoistdrymoist

-35

-30

-25

-20

-15

-10

-5

0

0 0.02 0.04 0.06 0.08 0.1 0.12

Iso

ste

ric

he

at o

f CH

4&

CO

2ad

sorp

tio

n (k

J m

ol-1

)

Entropy (kJ mol-1 K-1)

CH4 -dry- VRr=0.5% CH4 -dry- VRr=0.9% CH4 -dry- VRr=3.3%

CH4 -moist- VRr=0.5% CH4 -moist- VRr=0.9% CH4 -moist- VRr=3.3%

CO2 -dry- VRr=0.5% CO2 -dry- VRr=0.9% CO2 -dry- VRr=3.3%

CO2 -moist- VRr=0.5% CO2 -moist- VRr=0.9% CO2 -moist- VRr=3.3%

Kerogen type (Zhang et al. (2012))

localisedmobile

D

trans

D

transtrans SSS 23

CH4 CO2

0<q<0.7B)

CH4

CO2

I

II

III

CH4

CO2

subbituminous high-volatile bituminous anthracitedrymoistdrymoist

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

174 | P a g e

Figure 49: Isosteric heats of adsorption for CO2 and CH4 at low (part A) and high surface

coverage (part B) on dry and moist coals and different Kerogen types as a function of

entropy. Data from this study and Ref. (Li et al., 2010; Zhang et al.). Figure part A)

Surface coverage up to 30% Figure part B) and C) Surface coverage up to 70%. The

dashed line indicates the translation entropy change from the free gas 3D on the 2D

surface (45 and 49 J mol-1K-1 for CH4 and CO2 respectively). In part B) to illustrate the

thermodynamic parameter of the different coal rank we use increasing grayscale for

increasing rank.

Figure 49 shows the thermodynamic parameters at low surface coverage (here roughly

30%) calculated using Henry’s coefficients. This Figure provides a better differentiation

on how functional groups affect the sorption process due to the limited amount of

sorption sites provided by the functional groups. Figure 49 also illustrates that the

adsorbed CH4 and CO2 molecules on a dry high rank coal are very mobile in the adsorbed

phase. Contrary, Figure 49a and c show that the highest energetically localised CH4

molecules are adsorbed on a dry low rank coal. This indicates that the oxygen containing

functional groups dominate CH4 adsorption properties, whereas this contribution is less

significant for CO2 adsorption.

At least the first 30% of the adsorbed CO2 and CH4 molecules are mobile on high rank

coals (anthracite) where the surface chemistry is carbon dominated. However with

increasing surface coverage this rank effect is superimposed by translation restrictions

in the adsorbed phase due to a higher amount of adsorbed molecules (for example VRr

=3.3% in Figure 49b) in comparison to Figure 49a (low surface coverage)).

When pre-adsorbed water is present the adsorbed molecules are localised

independently of coal rank or gas type.

-35

-30

-25

-20

-15

-10

-5

0 0.02 0.04 0.06 0.08 0.1

Iso

ste

ric

he

at o

f ad

sorp

tio

n (

kJ m

ol-1

)

Entropy (kJ mol-1 K-1)

CO2 -dry- Li et al. (2010) CH4 -dry- Li et al. (2010)CO2 -moist- this study CO2 -dry- this studyCH4 -dry- this study CH4 -moist- this studyCH4 - Kerogen dry - Zhang et al. (2012) CH4 - Gasparik et al. (2012)

I

III

II

VRr =0.6%

S3d-2d (CH4) S3d-2d (CO2)

VRr =3.8%

VRr =3.3%

C) 0<θ<0.7

mobile

localised

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

175 | P a g e

Figure 49 illustrates that high rank coals (VRr =3.8%) are characterised by a high

sorption capacity and a comparable low change in entropy. This means that CH4

molecules are mobile in the adsorbed phase, indicating that the adsorbed phase has a

homogeneous energy distribution. Otherwise, energy barriers would reduce the degree

of freedom of the CH4 molecules.

At high surface coverage, the thermodynamic parameters for CH4 and CO2 only show

minor differences between the different coal ranks and pre-adsorbed water at high

surface coverage. This indicates that the interactions between the adsorbed molecules

(adsorbate-adsorbate) superimpose the thermodynamic properties at high surface

coverage.

The observation in Figure 49 that type I kerogen (low O/C atomic ratio) is characterised

by low adsorption energy and entropy values is in agreement with the generally low

amount of oxygen-containing functional groups. Contrary, CH4 sorption on type III

kerogen (high O/C atomic ratio) is controlled by localised molecules on high-energy

sorption sites (Zhang et al., 2012). It can therefore be concluded that the oxygen-

containing functional groups are the primary adsorption sites in the different kerogen

types. However observations from coals indicate that the CH4 adsorption capacity of

type III kerogen will be significantly affected by the presence of water, whereas it is

likely that this is marginal for type I kerogen. These properties of the adsorbed

molecules are more pronounced at low surface coverage. At high surface coverage

interactions between the adsorbate molecules and further, between carbon atoms of the

adsorbent and the adsorbate molecules are predominating the thermodynamic

properties.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

176 | P a g e

Conclusion

• The preferential sorption of CO2 over CH4 is highest for low rank coals and at low

pressures. This is caused by the availability of functional groups. The reduction in

sorption capacity by the presence water is also highest for low rank coals. The

reduction in sorption capacity due to pre-adsorbed water is very low for the

anthracite. Whereas, we observe that the reduction in sorption capacity by pre-

adsorbed water is very low on the anthracite.

The surface coverage dependence could be explained due to (1) high molecular

specific affinities of the high energetic sorption sites (low surface coverage) and

(2) related to low molecular specific affinities due to the less favourable ‘van der

Waals’-interactions between H2O, CH4 or CO2 and the carbon atoms of the coal

surface at high surface coverage. This decrease in the amount of functional

groups with increasing rank is caused by the decarboxylation and dehydration

during the coalification (van Krevelen, 1993).

Gas sorption capacity decreases with increasing moisture content up a certain

level of water saturation. Above this saturation level the gas sorption capacity

remains constant (Day et al., 2008c; Joubert et al., 1973b; Joubert et al., 1974). This

saturation level is a function of rank.

Furthermore, this reduction in sorption capacity can be clearly correlated with the

oxygen content of the coals.

• The proposed molecular concept explains the higher Langmuir pressure, the

reduction in sorption capacity by the presence of water, the parabolic trend in

sorption capacity as function of coal rank for dry coals, the linear trend for

moisturized coals, the exponential decay of the CO2/CH4 sorption capacity ratio as

function of rank in the presence of moisture and the constant CO2/CH4 sorption

capacity ratio of dry coals.

CH4 and CO2 adsorbed molecules are mobile within the adsorbed phase of

anthracite.

Whereas when water is pre-adsorbed, the adsorbed molecules have more

restrictions and are therefore localized within the adsorbed phase.

At high surface coverage the adsorbent-adsorbent interactions dominate the

thermodynamic properties.

Surface chemistry (oxygen-containing functional groups) controls the competition

between water and CO2 or CH4 for sorption sites.

The mechanism of the competition is occupation of potential CH4 or CO2 sorption

sites by water molecules. On sorption sites attributed to functional groups one

water molecule on average occupies the space of one CH4 and 2.2 CO2 molecules.

At high surface coverage the thermodynamic properties of the adsorbed phase are

dominated by adsorbate-adsorbate interactions. Surface chemistry, pore size

distribution or preadsorbed water has only minor influence on the thermodynamic

properties.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

177 | P a g e

Acknowledgments

Authors would like to thank Shell Global Solutions International for financial support

and the opportunity to publish this article. We appreciate the fruitful comments and

discussions with Rick Wentinck, Christian Ostertag-Henning and Ralf Littke.

Appendix

Figure A1: FTIR- spectra of subbituminous coal, high volatile bituminous coal and

anthracite. Hydroxyl groups have a characteristic peak at a wavenumber of

3360 cm-1, carbonyl-groups at 1800–1600 cm-1, carboxyl groups and ketones at

1700 cm-1. It is obvious from the spectra that the subbituminous coal contains the

highest amount of functional groups, whereas the anthracite contains only small

amounts of functional groups.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

178 | P a g e

References

1. Karacan, C.O., et al., Coal mine methane: A review of capture and utilization

practices with benefits to mining safety and to greenhouse gas reduction.

International Journal of Coal Geology. 86(2-3): p. 121-156.

2. Moore, T.A., Coalbed methane: A review. International Journal of Coal Geology,

2012. 101: p. 36-81.

3. Busch, A. and Y. Gensterblum, CBM and CO2-ECBM related sorption processes in

coal: A review. International Journal of Coal Geology, 2011. 87: p. 49-71.

4. Gutierrez-Rodriguez, J.A., Purcell, R.J. Jr.,Aplan F.F., Estimating the Hydrophobicity

of coal. Colloids and Surfaces, 1984. 12: p. 1-25.

5. Given, P.H., Marzec, A., Barton, W.A., Lynch, L.J., Gerstein, B.C., The concept of a

mobile or molecular phase within the macromolecular network of coals: A debate.

Fuel, 1986. 65: p. 155-163.

6. Mu, R. and V.M. Malhotra, A new approach to elucidate coal-water interactions by

an in situ transmission FT-i.r. technique. Fuel, 1991. 70(10): p. 1233-1235.

7. Nishino, J., Adsorption of water vapor and carbon dioxide at carboxylic functional

groups on the surface of coal. Fuel, 2001. 80: p. 757-764.

8. Coppens, L., Ann. Mines Belg., 1936. 37: p. 173.

9. Day, S., R. Sakurovs, and S. Weir, Supercritical gas sorption on moist coals.

International Journal of Coal Geology, 2008. 74(3-4): p. 203-214.

10. Joubert, J.I., C.T. Grein, and D. Bienstock, Sorption of methane in moist coal. Fuel,

1973. 52(3): p. 181-185.

11. Joubert, J.I.G., T. & Bienstock, D., Effect of moisture on the methane capacity of

American coals. Fuel, 1974. 53: p. 186-191.

12. Ozdemir, E. and K. Schroeder, Effect of Moisture on Adsorption Isotherms and

Adsorption Capacities of CO2 on Coals. Energy & Fuels, 2009. 23(5): p. 2821–2831.

13. Mosher, K., et al., Molecular simulation of methane adsorption in micro- and

mesoporous carbons with applications to coal and gas shale systems. International

Journal of Coal Geology, 2013. 109-110: p. 36-44.

14. Gensterblum, Y., et al., High-pressure CH4 and CO2 sorption isotherms as a function

of coal maturity and the influence of moisture. International Journal of Coal

Geology, 2013. submitted.

15. Ruthven, D.M., Principles of adsorption and adsorption processes. 1984: Wiley.

16. Xia, X., S. Litvinov, and M. Muhler, Consistent Approach to Adsorption

Thermodynamics on Heterogeneous Surfaces Using Different Empirical Energy

Distribution Models. Langmuir, 2006. 22(19): p. 8063-8070.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

179 | P a g e

17. Kemball, C., V.I.K. W.G. Frankenburg, and E.K. Rideal, Entropy of Adsorption, in

Advances in Catalysis. 1950, Academic Press. p. 233-250.

18. Gregg, S.J.a.S., K.S.W., Adsorption, surface area and porosity. Academic Press. 1967,

London and New York. 195.

19. Langmuir, I., Surface chemistry, in Nobelprize Lecture. 1932, Swedish Academy of

Sciences: Stockholm.

20. Svábová, M., Z. Weishauptová, and O. Pribyl, Water vapour adsorption on coal.

Fuel. 90(5): p. 1892-1899.

21. Svábová, M., Z. Weishauptová, and O. Pribyl, The effect of moisture on the sorption

process of CO2 on coal. Fuel, 2012. 92(1): p. 187-196.

22. van Krevelen, D.W., Coal: Typology-Physics-Chemistry-Constitution. 1993: Elsevier.

1002.

23. Tenney, C.M. and C.M. Lastoskie, Molecular simulation of carbon dioxide

adsorption in chemically and structurally heterogeneous porous carbons.

Environmental Progress, 2006. 25(4): p. 343-354.

24. Liu, Y. and J. Wilcox, Effects of surface heterogeneity on the adsorption of CO 2 in

microporous carbons. Environmental Science and Technology, 2012. 46(3): p.

1940-1947.

25. Narkiewicz, M.R. and J.P. Mathews, Visual representation of carbon dioxide

adsorption in a low-volatile bituminous coal molecular model. Energy and Fuels,

2009. 23(10): p. 5236-5246.

26. Heuchel, M., Davies, G.M., Buss, E., Seaton, N.A., Adsorption of Carbon Dioxide and

Methane and Their Mixtures on an Activated Carbon: Simulation and Experiment.

Langmuir, 1999. 15: p. 8695-8705.

27. Liu, Y. and J. Wilcox, Molecular simulation of CO2 adsorption in micro- and

mesoporous carbons with surface heterogeneity. International Journal of Coal

Geology, 2012. 104(0): p. 83-95.

28. Prinz, D. and R. Littke, Development of the micro- and ultramicroporous structure

of coals with rank as deduced from the accessibility to water. Fuel, 2005. 84(12-

13): p. 1645-1652.

29. Prinz, D. and R. Littke, Development of the micro- and ultramicroporous structure

of coals with rank as deduced from the accessibility to water. Fuel, 2005. 84(12-

13): p. 1655-1662.

30. Sakurovs, R., et al., Pore size distribution and accessible pore size distribution in

bituminous coals. International Journal of Coal Geology, 2012. 100(0): p. 51-64.

31. Behar, F. and M. Vandenbroucke, Chemical modelling of kerogens. Organic

Geochemistry, 1987. 11(1): p. 15-24.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

180 | P a g e

32. Schuyer, J., L. Blom, and D.W. Vankrevelen, The molar refraction of condensed

aromatic compounds. Transactions of the Faraday Society, 1953. 49(12): p. 1391-

1401.

33. Billemont, P., B. Coasne, and G. De Weireld, An experimental and molecular

simulation study of the adsorption of carbon dioxide and methane in nanoporous

carbons in the presence of water. Langmuir, 2011. 27(3): p. 1015-1024.

34. Billemont, P., B. Coasne, and G. De Weireld, Adsorption of carbon dioxide, methane,

and their mixtures in porous carbons: Effect of surface chemistry, water content,

and pore disorder. Langmuir, 2013. 29(10): p. 3328-3338.

35. Do, D.D., Adsorption Analysis: Equilibria and Kinetics. 1998: Imperial College

Press.

36. Zhang, T., et al., Effect of organic-matter type and thermal maturity on methane

adsorption in shale-gas systems. Organic Geochemistry, 2012. 47(0): p. 120-131.

37. Li, D., et al., High-pressure sorption isotherms and sorption kinetics of CH4 and CO2

on coals. Fuel, 2010. 89(3): p. 569-580.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

181 | P a g e

"God does not care about our mathematical difficulties. He integrates empirically."

Albert Einstein

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

182 | P a g e

5. High-pressure CH4 and CO2 sorption

isotherms as a function of coal

maturity and the influence of

moisture

International Journal of Coal Geology,2013, 118, page 45-57

Yves Gensterblum,

Alexej Merkel and Bernhard M. Krooss

RWTH-Aachen University, Institute of Geology and Geochemistry of Petroleum and

Coal, Lochnerstr. 4-20, D-52056 Aachen, Germany

Andreas Busch Shell Global Solutions International B.V., Kessler Park 1, 2288GS Rijswijk, The

Netherlands

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

183 | P a g e

Abstract

Methane (CH4) and carbon dioxide (CO2) sorption isotherms have been measured on an

Australian subbituminous, a German high-volatile bituminous coal and a German

anthracite in the dry and moisture-equilibrated state. The purpose was to study the

variation of CH4 and CO2 sorption capacities of the dry coals as a function of rank and the

influence of water on the sorption properties. Methane sorption isotherms were

measured at 303, 308, 318 and 334 K (30, 35, 45 and 61 °C), and CO2 isotherms at 318,

334 and 349 K (45, 61 and 76 °C).

The excess sorption capacity of coals is always higher for CO2 than for CH4. The CO2 and

CH4 sorption capacity of dry coals as a function of rank follows a parabolic trend

reported in earlier studies, with a minimum at ~1% vitrinite reflectance. This trend is

more pronounced for CO2 than for CH4. For moisturized coals a linear increase in CO2

and CH4 sorption capacity with coal maturity was observed. Moisture reduces the gas

sorption capacity of coals significantly. Moisture content therefore is a first-order

control for low rank coals up to bituminous rank, with much higher impact than

temperature or rank. The moisture-induced reduction in CO2 and CH4 sorption capacity

decreases with increasing coal rank. It correlates linearly with the oxygen content,

which in turn correlates qualitatively with the amount of hydrophilic and carboxylic

functional groups as evidenced by FTIR analysis.

The influence of sorbed water on the sorption capacity is highest at low pressures (low

surface coverage θ<0.3). The dry/moist sorption capacity ratios converge towards 1

with increasing pressure (high surface coverage θ0.7).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

184 | P a g e

Introduction

The increasing interest in the recovery of gas from natural unconventional deposits such

as coal beds or shale formations has led to increased efforts in studying the gas sorption

behaviour of coal and organic-rich shales (Gasparik et al., 2012; Li et al., 2010). The

accurate prediction of equilibrium sorption capacities is important to estimate gas

saturation and recovery potential in basin modelling and/or production simulators. To

accurately represent reservoir conditions, laboratory sorption isotherm data are

generally determined at elevated temperatures, usually between 298 and 349 K, and

elevated pressures (up to 20 MPa).

Despite significant progress in the experimental measurement of high-pressure/high-

temperature sorption isotherms, data on the physical fundamentals of the adsorption of

supercritical gases on coals or other geological samples in the presence of water are still

scarce. In addition the issue of excess vs. absolute sorption isotherms and the related

problem of the density of the sorbed phase for the respective gas type and its changes

with pressure and temperature have resulted in some confusion and controversy. This

paper aims at shedding light on these two aspects based on the data sets determined

here.

Competitive water/gas sorption The interaction of carbonaceous materials like natural coal with water is more complex

than with non-polar gases like helium, argon, nitrogen, methane, or carbon dioxide. This

complexity is due to the weak dispersion interaction of water with coal, the tendency of

water to form hydrogen bonds with other sorbed water molecules and functional groups

on the surface, and the chemisorptive interaction (e.g. in combination with CO2) with the

coal mineral matter (see discussion in previous chapters 1 and 4).

The sorption of water vapour and methane on coal surfaces has been studied for a long

time, and the importance of carboxylic and hydroxylic functional groups for this process

is well established (Given, 1986; Gutierrez-Rodriguez, 1984; Joubert et al., 1973b;

Joubert et al., 1974; Lynch and Webster, 1982; Mu and Malhotra, 1991; Nishino, 2001;

Suárez et al., 1993). Joubert et al. (1974) determined that methane sorption capacities of

oxygen-rich coals undergo a much greater reduction when saturated with moisture than

do their low-oxygen counterparts.

As early as 1936, Coppens reported the moisture-related decrease in methane sorption

capacity for several Belgian coals (Coppens, 1936). Polar sites such as hydroxyl groups

on the coal surface are preferentially occupied by water, which results in a reduction of

the sorption capacity for CO2 and CH4. Therefore moist coals have a significantly lower

maximum sorption capacity for either gas than dry coals. The extent to which the

capacity is reduced depends on coal rank. High-rank coals are less affected by the

presence of moisture than low-rank coals (Day et al., 2008c). Gas sorption capacity

decreases with increasing moisture content until the “limiting moisture content” is

reached. Above this moisture content the gas sorption capacity remains constant (Day et

al., 2008c; Joubert et al., 1973b; Joubert et al., 1974). This limiting moisture content

depends on the rank of the coal and, for Australian low-rank coals, corresponds

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

185 | P a g e

approximately to the equilibrium moisture content attained when exposing the coal to

about 40–80% relative humidity.

Conceptually the coal surface can be envisaged to consist of sets of more water prone

(hydrophilic) and more gas-prone sorption sites and there is a partial overlap of these

sets of sorption sites. For dry coals, sorption capacity shows a parabolic dependence on

rank, whereas coals with an inherent moisture content (“as-received”) display a more

linear correlation of sorption capacity with fixed carbon content (Busch and

Gensterblum, 2011; Ozdemir and Schroeder, 2009).

The second key parameter to characterize and describe gas sorption on coals in

competition with water is the presence and size distribution of micro-pores. This

contribution seems to be of lesser importance for low rank coals due to the

predominating effect of functional groups (Liu and Wilcox, ; Prinz and Littke, 2005a).

However for highly mature coals like anthracites the micro-pore radius distribution is

dominant because most of the functional groups have been lost during the coalification

process (Heuchel, 1999).

Excess and absolute adsorption and adsorbed phase density Most of the fundamental gas adsorption studies have been performed at low pressures

(and temperatures). Under these conditions the density of the bulk gas phase in

equilibrium with the sorbed phase is very low as compared to the adsorbed phase

density, especially for CO2. It is therefore assumed that the volume occupied by the

sorbed phase is negligible and the difference between “absolute” and “excess”

adsorption is negligibly small. For high-pressure sorption isotherms and especially with

CO2 as test gas, this idealization is however no longer valid. Here the volume of the

sorbed phase has measureable effects (increasing buoyancy in gravimetric

measurements and reduction of the void volume in volumetric measurements).

Adsorption takes place on the boundary surface between the bulk solid phase and the

free gas phase. The absolute amount of sorbed molecules will always increase with gas

pressure and eventually reach a saturation level. In experimental measurements the

consumption of volume by the adsorbed phase and, in particular at high pressures, the

decreasing difference between the density of the free gas phase and the adsorbed phase,

result in an additional effect that prohibits the direct determination of “absolute

sorption”. All sorption isotherm measurements (volumetric, manometric, gravimetric)

thus yield the “excess sorption”, which is based on an operational definition. The excess

adsorption describes the amount of molecules that can be accommodated in a system in

excess of the amount of the same molecules under the condition of non-sorption under

the same pressure and temperature conditions (Busch et al., 2003a). The “non-sorption”

case is usually defined by means of a reference measurement with a (by definition) non-

sorbing gas, most commonly helium. Taking explicitly into account that the adsorbed

molecules do not “disappear” but form a phase with a finite density and occupy a small

but measurable volume, the following general equation can be derived relaying excess

sorption to absolute sorption according to (Busch et al., 2006):

Equation 41

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

186 | P a g e

Here is the free gas density and (T) the density of the adsorbed phase. It is

evident from this equation that the excess sorption will invariably decrease as the free

gas density increases and its value becomes significant relative to the sorbed phase

density.

Thermodynamic considerations usually rely on absolute sorption values, but, as outlined

above, cannot be determined directly from sorption isotherm experiments and

certain assumptions have to be made. Thus, it is commonly assumed that is

independent of pressure and temperature (constant density approach (Murata et al.,

2001)). It was found experimentally that for methane on silica gel is close to the

inverse of the van der Waals volume (bvdW = 0.04286 and 0.04301 L/mol (Weast, 1972))

of the adsorbate (Humayun and Tomasko, 2000). The interaction parameters between

CO2 and silica are slightly different from those for CO2 and carbon atoms. However,

many experimental studies use 1027 and 373 kg m−3 as sorbed-phase densities of CO2

and CH4 respectively, with reasonably good fits to the data (Day et al., 2008a;

Gensterblum et al., 2010b; Gensterblum et al., 2009a; Sakurovs et al., 2010; Sakurovs et

al., 2013). For the coal samples used in this study, however, the corresponding fits

resulted in higher adsorbed phase densities (see below).

The Langmuir equation (Langmuir, 1918) is widely used as a first approach to describe

gas sorption on solid surfaces as a function of pressure and at constant temperature

(equation 42). In this study we use the Langmuir equation (Langmuir, 1918) as a basis

for fitting experimental isotherms.

Equation 42

Here [mmol g-1] is the amount of substance adsorbed at pressure p [MPa], n

[mmol g-1] is the total sorption capacity, defined as the amount of substance adsorbed at

infinite pressure and PL(T) [MPa] is the Langmuir pressure which is the pressure where

half the sorption sites are occupied. Further the Langmuir pressure is an inverse

indicator for the affinity (intensity of physical interaction) between adsorbent and

adsorbate. The Langmuir equation was originally derived on the basis of simple

molecular dynamic considerations and verified at very low pressures where the volume

of the adsorbed phase was negligible.

For high-pressure sorption experiments, in particular with supercritical CO2, the density

of the adsorbed phase is no longer negligible and has to be taken explicitly into account.

The excess sorption function (equation 43) is explained in detail by Gensterblum et al.

(2009) and corresponds to equation 42 with the Langmuir formula used as the “absolute

sorption function”.

Equation 43

Equation 43 was used in this study to parameterize the experimental sorption isotherm

and to compare the different isotherms at equal surface coverage. Note that some of the

fundamental assumptions of the Langmuir (1918) model (monolayer sorption;

energetically and structurally homogeneous surface) are not met or verified for the

samples and experimental conditions considered here. But due to the fact that the

Langmuir equation represents the simplest model for an “absolute sorption isotherm”

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

187 | P a g e

and that equation 43 described our results within the range of experimental error, we

chose it as an “Ockham’s razor” approach (“entities must not be multiplied beyond

necessity”; (Glymour, 1980) for the parameterisation of the excess sorption isotherms

measured in this study.

Samples and sample preparation

To investigate the effect of rank on the sorption enthalpy and entropy, two coal samples

of intermediate and high maturity from the Ruhr Basin, Germany and one low rank coal

from the Surat Basin, Australia have been selected. Details on the geological setting for

all three coal samples have been described elsewhere (Gensterblum et al. 2013,

accepted). A summary, including available data for proximate and ultimate analysis, is

provided in Table 19.

For measurements on powdered coal material, crushed samples were divided and

aliquots were ground to pass a sieve size of 2 mm. For sorption measurements, the sieve

fraction of 0.5 to 1mm was used. The coal samples were moisture-equilibrated at 97%

relative humidity in a desiccator for 7 to 14 days following standard procedures. The

sample material was then transferred immediately into the adsorption cell; an aliquot

was used for the determination of the moisture content.

The equilibrium moisture contents for the high-volatile bituminous coals and the

anthracite were higher than those of other coal samples of similar rank in the literature

data compilation by Busch and Gensterblum (2011). However the values are averages of

9 to 11 measurements and the standard deviations indicate their accuracy and

reproducibility.

Carbon, hydrogen and nitrogen contents were determined using a Leco CHN-2000

analyser following the DIN 51732. Sulfur contents were determined on a Leco SC 144 DR

analyser following DIN 51724. Oxygen content was calculated by difference.

The results of the ultimate and proximate analysis of the coal samples are listed in Table

19. The oxygen contents decrease from 31% to 17% and down to 6.6% with increasing

maturity. FTIR spectroscopy confirmed the high amounts of oxygen-containing

functional groups such as hydroxyl and carboxyl groups present in the subbituminous

coal (Figure 50). For the high-volatile bituminous coal the amount of oxygen-containing

functional groups is lower and the IR spectrum of the anthracite does not show any

oxygen-containing functional groups.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

188 | P a g e

Table 19: Ultimate and proximate analysis of coal samples used in this study.

For the analyzing carbon, hydrogen and nitrogen content a Leco CHN-2000 was used

and we follow the measurement procedure DIN 51732. For the sulfur content we use a

Leco SC 144 DR and we follow the measurement procedure DIN 51724.Oxygen was

calculated by the difference approach (100%-C%-N%-S%-H%=O%). For the

determination of the vitrinite reflectance we use 100 points. For the maceral group

analysis we count 500 points.

Queensland,

AUS

Prosper Haniel

mine, GER

Ibbenbüren mine,

GER

Coal Basin Surat Ruhr Ruhr

Geologic age Middle

Jurassic Pennsylvanian Pennsylvanian

Rank Sub-

bituminous A

High-volatile

bituminous A anthracite

VRr (%) 0.5 0.93 3.3

Volatile mater, mf (%) 43.5±0.4 31.7±1.2 6.0

Fixed carbon, mf (%) 49.5±0.1 58.0±1.7 88.2

Ash, mf (%) 3.5±0.2 7.3±2.2 5.8

As received moisture (%) 3.6±0.2 3.0 ±0.2 0.85

Equilibrium moisture (%) 6.4±0.1 4.0±0.5 4.8 ±0.3

Vitrinite / huminite (%) 61.4 69.3 ±3.2 83

Liptinite (%) 31.4 7.1 ±2.3 0

Inertinite (%) 0.6 14.6±4.0 17

Mineral matter (%) 6.6 8.7±1.4 n.a.

C (% by weight) 62.2 74.7 88.5

H (% by weight) 4.68 4.47 2.54

N (% by weight) 1.01 1.46 1.34

S (% by weight) 0.48 1.96 0.96

O diff (% by difference) 31.63 17.41 6.66

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

189 | P a g e

Figure 50: FTIR-spectra of subbituminous (light grey), high-volatile bituminous (dark

grey) and anthracite (black) coals. The characteristic wave numbers for the hydroxyl

groups are at 3360 cm-1 (left circle), for the carbonyl-groups 1800-1600 cm-1,and the

carboxyl groups and ketones at 1700cm–1 (right circle). It is obvious that the

subbituminous coal has the highest concentration of functional groups, while the

anthracite has the lowest concentration.

Experimental procedure

All experimental sorption data reported here were obtained using the manometric

method. The set-up used for sorption isotherm measurements and for determining

sorption kinetics is explained in detail by Busch and Gensterblum (2011). Further

experimental details of this method are discussed by Gensterblum et al. (2009, 2010),

reporting on two different inter-laboratory comparison studies on activated carbon and

a set of three different coals.

Free phase gas densities of CH4 and CO2 as a function of pressure and temperature were

calculated using the GERG multi-component equation of state (EOS) (Kunz et al., 2007)

that provides also mixture densities for CO2/water vapour and CH4/water vapour.

Experimental conditions The isotherms on moisturised coals were measured on aliquots from the same batch.

Samples once exposed to CO2 were not re-used in successive measurements because it is

believed that CO2 may change the microporous network of coal and/or dissolve in the

organic structure while extracting organic compounds from coal (Goodman, 2005a;

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

190 | P a g e

Goodman et al., 2006; Goodman, 2005b; Larsen, 2004). One aim of this study was to

elucidate the influence of these organic compounds on the excess sorption capacities.

Samples used for sorption tests in the dry state were dried at 105°C under vacuum over

night. After transferring the sample to the sample cell it was dried again at 105°C under

vacuum for 90 minutes to remove water potentially adsorbed during the transfer

process.

Evaluation procedure The three-parameter excess sorption function based on the Langmuir model (Equation

43) and taking explicitly into account the density of the adsorbed phase (ads), provided

an adequate representation of the measured excess sorption data. The total sorption

capacity and the Langmuir pressure PL(T) as a function of temperature were used as

fitting parameters.

The density of the adsorbed phase was taken as constant with values of 373 and

1027 kg m-3 for CH4 and CO2, respectively, corresponding to the inverse of the “van der

Waals volumes”. This assumption is valid only if the adsorbent-adsorbate energy

(roughly 20 kJ/mol for CH4 and 30 kJ/mol for CO2) is significantly higher than the

thermal energy (RT 2.5 - 2.9 kJ/mol for T = 303 K – 349 K).

Data fitting procedure The observation of maxima in CH4 and CO2 excess sorption isotherms on coals at high

pressures has important implications for the choice of parameterisation of the

experimental data. As stated above, the three-parameter excess sorption function

of equation 43 with the Langmuir function as the “absolute

sorption” term is able to reproduce the qualitative features of the experimental excess

sorption isotherms. The function may, however, fail to yield unique fits for CH4

isotherms if the maximum is not sufficiently high in amplitude. Excellent approximations

of the experimental sorption data can be achieved by two or more slightly different 3-

parameter sets, indicating that different combinations of total sorption capacity,

Langmuir pressure and adsorbed phase density may result from the optimization

procedure. Fits of equal quality may thus be obtained with significantly different

adsorbed phase density values (see Table and Figure A1). Usually the adsorbed phase

density of CH4 cannot be estimated by the graphical method used for instance by

(Humayun and Tomasko, 2000). The results can be improved by simultaneously fitting

several isotherms recorded at different temperatures.

It should be noted that heterogeneous data point distributions (irregular pressure

differences between the different pressure steps resulting in weighting effects) may also

influence the results of the fitting procedure.

In terms of the parameterization of the experimental values two opposing effects are

included in equation 43 with the Langmuir pressure term, dominating the low-pressure

range (0 - 8 MPa) of the isotherm, and the bulk vs. adsorbed phase density ratio

controlling the high-pressure portion (>8 MPa). In the intermediate to high pressure

range (5 MPa - maximum pressure) the fitting quality is sensitive to both fitting

parameters, the total sorption capacity and the sorbed phase density (ads). To

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

191 | P a g e

bypass this limitation we assume that is temperature-independent and only PL is

used as fitting parameter (Tables 20-22). During the fitting procedure it became evident

that the temperature invariant “van der Waals densities” were too low to achieve a good

approximation to the experimental data (equation 5: n>0.02 mmol/g limit). In this

case, the adsorbed phase density was also used as a fitting parameter. Furthermore, for

a few isotherms it was necessary to adjust the total amount of sorption sites slightly

(Table 20 and Table 21), to achieve appropriate fitting results. We considered the error

sum limiting criterium to be fulfilled when the deviation between measured and fitted

values was less than 0.02mmol/g. This limit is justified by the accuracy limit of the

experimental set-up (Gensterblum et al., 2010b; Gensterblum et al., 2009b).

In order to improve the quality of the fitting procedure for the Langmuir pressure, the

first step involves the calculation of a weighted error sum (equation 44), with emphasis

on the low-pressure range. As a general quality criterium for the fitting procedure the

error sum according to equation 45 was calculated. The degrees of freedom were

reduced until the unweighted error sum (equation 45) fell below the the n=0.02

mmol/g limit.

Equation 44

Equation 45

Results and Discussion

Due to the exothermic nature of the process, PL increases with increasing temperature.

This means that higher pressures are necessary to reach the same degree of coverage

(sorption capacity n(p)) at higher temperatures. Based on the assumption that a

temperature increase (35°C up to 75°C) will not change the surface, pore or chemical

structure of the coal, we assume that the sorption capacity should be independent of

temperature, i.e. the total amount of sorption sites (sorption capacity) should remain

constant. However, due to the presence of moisture, a shift in the portion of sorption

sites occupied by water can be assumed. The saturated moisture content, or sorption

capacity of water, will decrease only slightly with increasing temperature, because the

sorption enthalpy of water is much higher than the sorption enthalpies of carbon

dioxide and methane (Busch and Gensterblum, 2011). As a simplification, we assume

that the total amount of sorption sites remains approximately constant with

temperature; however, the proportion of sorption sites occupied by CH4 and CO2,

respectively, at a given temperature and pressure will decrease slightly (due to the

different differential enthalpy for CH4 and CO2).

In the present study 36 sorption isotherms for CO2 and CH4 were recorded at dry and

moist conditions and temperatures between 318 and 349 K for CO2 and between 303

and 349 K for CH4. Sorption capacities are reported on a dry ash-free basis. Figure 51

through 54 show the CH4 and CO2 isotherms on dry and moist coals with the fitted

excess sorption function (equation 43).

As documented in earlier studies, almost all isotherms for dry coals have significantly

higher excess sorption capacities than for moist coals. Furthermore all CO2 isotherms

show higher excess sorption capacities than their corresponding CH4 isotherms on dry

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

192 | P a g e

and moist coals (Figure 51 - 54). The thermodynamic treatment and a concept of the

molecular adsorption and competition process between H2O and CO2 or CH4 is published

elsewhere.

The following results and discussion focus on the saturation properties based on the

Langmuir parameters and sorption properties as a function of pressure (surface

coverage).

Isotherm characteristics

All isotherms correspond to the IUPAC type I up to pressures around 8 MPa but

subsequently pass through a maximum and decrease. The effect is stronger at measuring

temperatures close to the critical point of the particular gas (Pc_CO2=7.37MPa;

Tc_CO2=304K; Pc_CH4=4.6MPa; Tc_CH4=190.4K). Therefore, the maximum is much more

pronounced for CO2 than for CH4 due to the higher reduced temperature Tr = T/Tcrit of

1.67 for CH4 as compared to 1.05 for CO2 at 318K. This phenomenon is due to the

decreasing density difference between the free gas/supercritical phase and the

adsorbed phase. This phenomenon was observed previously e.g. (Busch et al., 2007; Pini

et al., 2010).

The CH4 and CO2 sorption capacity at 5 MPa decreases with increasing temperature due

to the fact that adsorption is an exothermic process (Figure 51 - 54). This is documented

in figure 51 through 54 for the subbituminous, high-volatile bituminous and anthracite

coals, respectively. However, for CO2 the temperature trend of the excess sorption

isotherm reverses at high pressures (~15 MPa) (Figure 53 and 54). At 15 MPa the

highest excess sorption occurs at the highest temperature where the density of the

supercritical CO2 is lowest (and the free vs. sorbed phase density ratio is highest). This

crossover was documented and discussed previously by Li et al. 2010.

With increasing temperature (318 K, 334 K and 350 K) the CO2 excess sorption

isotherms show a shift of the excess sorption maximum towards higher pressures (8, 10

and 12 MPa, respectively). For dry samples this maximum in excess sorption capacity

occurs at lower pressures than for moist samples. As a consequence of the exothermic

nature of the sorption process, all isosters (equal occupancy ratios ) shift to

higher pressures with increasing temperature.

However, for CH4 excess sorption isotherms this decrease at pressures higher than 10

MPa is not observed in all measurements. It is found for dry samples and for the 303K

isotherms on the moist high-volatile bituminous coal. This is an indicator for a lower

adsorbed phase density for these particular isotherms.

Adsorbed phase density Adsorbed phase density values were obtained from the least-square fit of the excess

sorption function to the sorption isotherms. The results of the approximations are listed

in Table 20 through Table 22.

For dry low-rank, subbituminous coals the adsorbed phase densities of CO2 and CH4, are

comparatively high. This is attributed to the presence of oxygen-containing functional

groups causing higher packing densities for CO2 and CH4 (Liu and Wilcox, 2012a). These

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

193 | P a g e

oxygen-containing functional groups are preferentially occupied by water in moist coal

due to its polarity. For CO2 and CH4, which have no permanent dipole moment, adsorbed

phase densities are considered to be lower (Liu and Wilcox, 2012a). Therefore, lower

adsorbed phase densities and lower sorption capacities are observed experimentally for

gases on low rank coals in the presence of water. The increases in adsorbed phase

density for high rank coals can be attributed to the overlapping Lennard-Jones

potentials in the ultra-micropores (strong pore wall – wall interactions) and the increase

in polarizability per carbon atom as the polyaromatic ring systems become larger (Behar

and Vandenbroucke, 1987). The higher polarizability is due to the delocalization of the π

electrons of the aromatic rings (Schuyer et al., 1953).

The observed increase of the adsorbed phase density with increasing temperature is

counter-intuitive, because, as in the case of a free gas, the density would be expected to

decrease with increasing temperature. Similar findings have, however, been reported in

the literature previously (Day et al., 2008c; Li et al., 2010; Pini et al., 2010). The

adsorbed phase density contains all volumetric differences between helium and CH4 or

CO2 and volumetric effects like swelling during the measurements. Whereas, the

swelling of the coal matrix, in the presents of CH4 or CO2 will lead to a lower adsorbed

phase density. However, with increasing temperature the amount of swelling will

decrease and this will lead to a higher adsorbed phase density value.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

194 | P a g e

Table 20: Langmuir fitting parameters for CH4 and CO2 excess sorption isotherm on dry

and moist sub-bituminous coal.

SUB-BITUMINOUS COAL

PL

(MPa)

n

(mmol/g)

ρads

(kg/m³)

n

(mmol/g)

dry, CH4

308K 2.17 1.18 1100 0.012

319K 4.94 1.02 600 0.004

334K 6.48 1.02 600 0.006

moist, CH4

303K 7.10 0.77 600 0.004

319K 10.47 0.77 373 0.006

334K 13.57 0.77 373 0.003

dry, CO2

318.5K 2.80 2.34 1539 0.023

334.3K 3.87 2.34 1465 0.011

350.0K 7.11 2.34 1263 0.007

moist, CO2

318.1K 3.68 1.61 894 0.019

334.0K 7.06 1.61 1357 0.022

350.1K 10.22 1.61 1638 0.013

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

195 | P a g e

Table 21: Langmuir fitting parameters for CH4 and CO2 excess sorption isotherm on dry

and moist high-volatile bituminous coal.

HIGH-VOLATILE BITUMINOUS COAL

PL

(MPa)

(mmol/g)

ρads

(kg/m³)

n

(mmol/g)

dry, CH4

308.0K 1.88 0.879 631 0.006

318.0K 2.94 0.879 519 0.007

334.0K 4.13 0.879 434 0.01

moist, CH4

303K 5.00 0.84 407 0.005

318.6K 6.80 0.88 900 0.005

334.2K 9.46 0.88 900 0.005

dry, CO2

318.7K 1.84 1.85 1038 0.007

334.1K 2.90 1.85 1160 0.010

350.0K 3.92 1.85 1294 0.012

moist, CO2

318.3K 2.80 1.56 1061 0.008

333.9K 5.03 1.56 1168 0.018

349.7K 7.38 1.56 1364 0.014

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

196 | P a g e

Table 22: Langmuir fitting parameters for CH4 and CO2 excess sorption isotherm on dry

and moist anthracite.

ANTHRACITE

PL

(MPa)

(mmol/g)

ρads

(kg/m³)

n

(mmol/g)

dry, CH4

308.5K 0.89 1.52 515 0.011

318.7K 1.20 1.52 488 0.013

349.5K 2.17 1.52 398 0.016

moist, CH4

303.3K* 2.28 1.55 99999* 0.015

(0.119)*

318.2K 3.14 1.55 505 0.014

334.6K 4.53 1.55 464 0.015

dry, CO2

318.3K 0.39 2.04 1801 0.025

334.2K 0.68 2.04 1901 0.009

350.0K 1.13 2.04 1632 0.012

moist, CO2

318.2K 1.39 1.94 1295 0.022

334.6K 2.09 1.94 1135 0.02

349.7K 3.21 1.94 1152 0.02

* This isotherm is not in the line with the other measurements especially at high pressures. Therefore we

only consider the pressure range up to 4 MPa for the fitting here.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

197 | P a g e

Figure 51: CH4 adsorption isotherms for moisture-equilibrated coals (particle size fraction ~0.5 to 1mm) at 303K, 319K and 334K (1mmol/g

=24.5 Std m³/t).The triangles represents isotherms measured on the subbituminous coal. Circles stands for isotherms measured on high-

volatile bituminous coal and diamonds for isotherms measured on anthracites. The isotherm temperature increases with increasing filling

of the symbols.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

0 5 10 15 20

CH

4e

xce

ss s

orp

tio

n (

mm

ol/

g DA

F)

Pressure (MPa)

Subbit. - 303K HvB - 303K Anthracite - 303KSubbit. - 319K HvB - 319K Anthracite - 319KSubbit. - 334K HvB - 334K Anthracite - 334K

moist

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

198 | P a g e

Figure 52: CH4 adsorption isotherms for dry coals (particle size fraction ~0.5 to 1mm) at 303K, 318K and 333K (1mmol/g =24.5 Std m³/t).

Triangles represent isotherms measured on the Subbituminous A coal. Circles are isotherms measured on high-volatile bituminous coal and

diamonds isotherms measured on anthracites. Shading of symbols increases with isotherm temperature.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

0 5 10 15 20

CH

4e

xce

ss s

orp

tio

n (

mm

ol/

g DA

F)

Pressure (MPa)

Subbit. - 308K HvB - 308K Anthracite - 308K

Subbit. - 319K HvB - 319K Anthracite - 319K

Subbit. - 334K HvB - 334K Anthracite - 349K

dry

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

199 | P a g e

Figure 53: CO2 excess sorption isotherms for moist coals as a function of gas pressure (particle size fraction ~0.5 to 1mm) at 318K, 334K and

349K (1mmol/g =24.5 Std m³/t). The triangles represents isotherms measured on the subbituminous coal. Circles stands for isotherms

measured on high-volatile bituminous coal and diamonds for isotherms measured on anthracites. Shading of symbols increases with

isotherm temperature

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

0 5 10 15 20

CO

2ex

cess

so

rpti

on

(mm

ol/

g DA

F)

Pressure (MPa)

Subbit. - 318K HvB - 318K Anthracite - 318K

Subbit. - 334K HvB - 334K Anthracite - 334K

Subbit. - 350K HvB - 350K Anthracite - 350K

moist

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

200 | P a g e

Figure 54: CO2 excess sorption isotherms for dry coals as a function of gas pressure (particle size fraction ~0.5 to 1mm) at 318K, 334K and 349K

(1mmol/g =24.5 Std m³/t). The triangles represents isotherms measured on the subbituminous coal. Circles stands for isotherms measured

on high-volatile bituminous coal and diamonds for isotherms measured on anthracites. The isotherm temperature increases with

increasing filling of the symbols.

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

2.00

0 5 10 15 20

CO

2ex

cess

so

rpti

on

(mm

ol/

g DA

F)

Pressure (MPa)

Subbit. - 318K HvB - 318K Anthracite - 318KSubbit. - 334K HvB - 334K Anthracite - 334KSubbit. - 350K HvB - 350K Anthracite - 350K

dry

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

201 | P a g e

Evolution of sorption properties with coal rank

The elemental composition and the pore structure of coals change substantially during

thermal maturation (Prinz and Littke, 2005a; Prinz et al., 2004; Sakurovs et al., 2012;

van Krevelen, 1993), and this is equally the case for their sorption properties.

In Figure 55, the Langmuir pressures for CO2 and CH4 are shown as a function of vitrinite

reflectance of the coal samples. Generally, the Langmuir pressure is an indicator for the

affinity between adsorbent and adsorbate, with high Langmuir pressures indicating low

affinity. As evidenced by this study, Langmuir pressures are higher for moist than for

dry coals and decrease with coal rank. This is evidenced by the increase of the initial

slope of the isotherms with increasing coal rank. Thus, with increasing maturity the

affinity of the coal for the gases increases and the same degree of “surface coverage” (q)

is achieved at lower pressures. Taking the ratios of the Langmuir pressures in Table 20-

22 and the relationship

,

it can be seen that, the CO2 pressure required to achieve a certain surface

coverage (isostere) amounts to 57%, 63% and 33% of the corresponding CH4 pressure

for subbituminous, high-volatile bituminous and anthracite coals respectively in

the dry state (Figure 55B). A similar trend is observed for sorption of CO2 and CH4 on

moisture-equilibrated coals. Here the isosteric CO2 pressures are 35%, 41% and 44% of

the corresponding CH4 pressures for the same maturity sequence (Figure 55B).

The total sorption capacity ndry(p) can be interpreted as the total number of sorption

sites available for the particular gas on a specific sample. In this study, the CO2 and CH4

sorption capacity for dry coals follow, as reported previously, a parabolic trend with a

minimum at ~1% vitrinite reflectance (Chapter 1 and 4). This parabolic trend is more

distinct for CO2 than for CH4. For moisturized coals a linear increase in CO2 and CH4

sorption capacity with increasing coal maturity has been observed. Generally a linear

sorption capacity trend with increasing maturity for the moist samples and a more

parabolic trend for dry coal samples are observed here and are in agreement with

previous findings (Chapter 1 and 4).

The CO2 Langmuir sorption capacity is higher than for CH4 over the total maturity

range. Further, the difference between the different Langmuir sorption amounts n, as

observed for moist and dry low rank coals, becomes small for the high-rank anthracite.

For the dry subbituminous coal the Langmuir sorption capacity is higher than for the

corresponding moist coal while this difference decreases with increasing maturity

(Figure 56).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

202 | P a g e

Figure 55: Part A) CH4 (triangle) and CO2 (square) Langmuir pressures as a function of

coal maturity on moist and dry coal samples at 318K (Chapter 5). In part B) the

Langmuir pressure ratios of CO2 divided by CH4 (light grey, circles) at dry and for

moist coal samples (dark grey, circles) has been calculated. Langmuir pressure ratios

of CH4 and CO2 isotherms calculated of dry and moist coals (CH4, triangles and CO2,

squares) at dry and for moist coal samples.

1

2

0

2

4

6

8

10

12

0 0.5 1 1.5 2 2.5 3 3.5

CH

4, C

O2

Lan

gmu

ir p

ress

ure

(MP

a)

Vitrinite reflectance (%)

CH4 - dry

CH4 - moist

CO2 - dry

CO2 - moist

A)

0

0.5

1

1.5

2

2.5

3

3.5

4

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 0.5 1 1.5 2 2.5 3 3.5

mo

ist/

dry

Lan

gmu

ir p

ress

ure

rati

o (

MP

a)

CO

2/C

H4

Lan

gmu

ir p

ress

ure

rati

o (

MP

a)

Vitrinite reflectance (%)

dry

moist

CH4

CO2

B)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

203 | P a g e

Figure 56: CH4 (triangle) and CO2 (square) Langmuir sorption amount as a function of coal

maturity of moist and dry coals (Chapter 5) Langmuir capacity moist/dry ratio for CH4

(triangles) and CO2 (squares) at 319K as a function of maturity on moist and dry coal

samples. CO2/CH4 Langmuir capacity ratio for dry coal samples (light grey circles) and

for moisturised samples (dark grey circles) at 319K as a function of maturity.

0

0.5

1

1.5

2

2.5

0 0.5 1 1.5 2 2.5 3 3.5

Lan

gmu

ir s

orp

tio

n c

apac

ity

(mm

ol/

g DA

F)

Vitrinite reflectance (%)

CH4 - dry

CH4 - moist

CO2 - dry

CO2 - moist

A)

0

0.5

1

1.5

2

2.5

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

0 0.5 1 1.5 2 2.5 3 3.5

Lan

gmu

ir s

orp

tio

n c

apac

ity

(CO

2/C

H4)

Lan

gmu

ir s

orp

tio

n c

apac

ity

(mo

ist/

dry

)

Vitrinite reflectance (%)

CH4

CO2

dry

moist

B)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

204 | P a g e

For methane we clearly observe a decreasing influence of moisture with increasing

maturity. The influence of water on the CO2 sorption capacity is high for the low-rank

coal and slightly less for the high-volatile bituminous coal. The anthracite shows only a

comparatively high influence of water on the sorption capacity of CO2 and CH4 at low

pressures.

The difference between dry and moist sorption capacity for both, CO2 and CH4 increases

with increasing oxygen content of the coals (Figure 57). Joubert and Bienstock (1974)

observed similar correlations between the reduction of sorption capacity by the

presence of sorbed water and the oxygen content (Joubert et al., 1974). Furthermore the

oxygen content correlates qualitatively with the amount of hydrophilic and carboxylic

functional groups (FTIR analysis). The oxygen content can be correlated with the

amount of oxygen-containing functional groups (Busch and Gensterblum, 2011; Ozdemir

and Schroeder, 2009; Svábová et al., 2012).

In general the amount of functional groups (OH) decreases and the carbon content

increases with increasing coal rank (van Krevelen, 1993). Therefore the functional

groups act as primary sorption (high energy) sites for water or, if no water is present,

for CO2 and CH4. The pore radius contribution is of lesser importance at low rank coals

due to the presence of functional groups. Increasing the surface density of oxygen-

containing functional groups generally increases CO2 adsorption capacity and lowers the

pore filling pressure (Tenney and Lastoskie, 2006). However, for anthracites the pore

size distribution is dominant because most of the functional groups are eliminated

during coalification.

Only very small changes in the sorption capacity ratio between moist and dry conditions

are observed upon a change in isotherm temperature. Therefore, we conclude the

influence of water on the CO2 and CH4 sorption capacity at different temperatures is

approximately constant over the temperature range used. This is probably due to high

sorption enthalpies of water molecules (up to 63 kJ/mol) for sorption sites attributed to

functional groups (Busch and Gensterblum, 2011; Darcey, 1958). Therefore the

occupancy of sorption sites by water remains approximately constant within the small

temperature range used in this study.

Generally the CO2/CH4 excess sorption ratios are higher for moist coals than for dry

coals. The CO2/CH4 ratios for the dry and moist state are highest for the subbituminous

coal, followed by the high-volatile bituminous coal and the anthracite.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

205 | P a g e

Figure 57: CH4 and CO2 sorption capacity difference between dry and moist coals as a

function of oxygen content.

Evolution of sorption capacity with surface coverage

Influence of moisture on the sorption capacity The influence of moisture on the CO2 and CH4 excess sorption capacity of the three coals

was analysed in a first approach by calculating the dry vs. moist sorption capacity ratios:

(i=CH4 or CO2)

The results for the individual coal samples at 318 K are plotted in Figure 58. This

diagram documents the increase of the sorption capacity ratio of a dry and moisturised

sample changes with increasing surface coverage.

In most cases, for both CH4 and CO2, the moist/dry excess sorption capacity ratio

increases with increasing pressure and asymptotically approaches the 1:1 ratio. This is

equivalent to an increasing difference in surface coverage with increasing pressure for

the two species. For CH4 the moist/dry sorption capacity ratio shows only a slight

increase with rank from 0.4 to 0.55 at low surface coverage (pressures) (Figure 58). CO2

and CH4 moist/dry sorption capacity ratios at low surface coverage are identical for the

anthracite up to 8 MPa (Figure 58). The influence of water on sorption capacity is

generally higher at low (ratio 0.49-0.63 at 1 MPa) than at high surface coverage (ratio of

0.70 up to 0.85 at 8 MPa) except for the subbituminous coal. This low-rank coal shows a

slight increase in the moist/dry sorption capacity ratio for CH4 (0.39 up to 0.5) and for

CO2 a constant capacity ratio of 0.55 followed by a drop down to 0.15 at 8 MPa. This

nCH4 = 0.0108cOxygen - 0.1179R² = 0.9737

nCO2 = 0.0242 cOxygen - 0.0827R² = 0.9894

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 5 10 15 20 25 30 35 40

CH

4, C

O2

Lan

gmu

ir s

orp

tio

n c

apac

ity

(nd

ry-n

mo

ist)

(m

mo

l g-1

)

Oxygen content (%)

CH4

CO2

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

206 | P a g e

drop is associated with a low CO2 adsorbed phase density (894kg m-3) for the moist as

compared to a high value (1539kg m-3) for the dry sample.

Figure 58: Moist/dry excess sorption capacity ratios for CH4 (grey) and CO2 (black).

The generally low sorption capacity ratios (0.38 up to 0.58) for the subbituminous A

coal in Figure 58 provide indicate that

i) competition between H2O and CH4 or CO2 for sorption sites on the subbituminous

coal is not limited to low surface coverage.

ii) the CO2 adsorbed phase density is lower when water is pre-adsorbed. For CH4 such a

conclusion is not possible because the free gas density is much lower than the

adsorbed phase density

In general, it is obvious that the main competition between H2O and CO2 or CH4 will

occur at low surface coverage and predominantly for low-rank coals. Subbituminous A

coals have only a small proportion of micropores (Prinz and Littke, 2005a). However,

micropores are less prone to water uptake than functional groups. At high pressures the

moist/dry sorption capacity ratios for CH4 converge to 0.50, 0.86 and 0.95, respectively,

in the maturity sequence. The trends of the ratios in Figure 58 show that the influence of

water on the sorption capacity initially decreases with increasing pressure. At higher

surface coverage the influence of water on the CH4 sorption capacity decreases for the

high-volatile bituminous coal and the anthracite.

The first sorption sites occupied are those, which are easily polarisable, and these have

the highest sorption enthalpies (Ruthven, 1984). These sites are preferably occupied by

water, the dipole moment of which favours a relatively strong interaction between the

0

0.2

0.4

0.6

0.8

1

0 5 10 15 20

CH

4,

CO

2e

xce

ss s

orp

tio

n c

apac

ity

rati

oD

AF

(m

ois

t/d

ry)

pressure (MPa)

318K

subbituminous A

high-volatile bituminous

anthracite

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

207 | P a g e

water molecule and the coal surface. These high-energy sorption sites are commonly

attributed to functional groups (Joubert et al., 1973b). Molecular dynamic simulations

indicate that carboxylic groups (-COOH) followed by hydroxyl groups (-OH) are

preferentially occupied by CO2 (Liu and Wilcox, 2012a; Tenney and Lastoskie, 2006).

Carboxylic and hydroxyl groups are also preferential adsorption sites for water

molecules. Oxygen-containing functional groups increase the CO2 density in graphitic slit

pores (Liu and Wilcox, 2012a). Furthermore, the oxygen-containing functional groups

induce a dense packing pattern and therefore the pore space is used by CO2 more

efficiently when carboxylic groups (-COOH) and hydrophilic groups (-OH) are present

(Liu and Wilcox, 2012a).

CO2/CH4 sorption capacity ratio The CO2/CH4 sorption ratio for dry and moist coals as a function of gas pressure was

used to assess the relative sorptive affinity of the two gas species (Figures 59 and 60).

This ratio is defined as:

(i=dry or moist)

Figures 59 and 60 illustrate the differences between excess- and absolute sorption in

terms of sorption capacity. Both Figures show the differences between observations at

low and high surface coverage. However, coal rank has a more dominant influence on

the extent of the CO2/CH4 sorption ratio (5.1, 4.1 and 2.7 for moist and 3.8, 3.1 and 2.6

for dry coals in order of increasing rank) than temperature. For the following discussion

we arbitrarily define the “low surface coverage” pressure range from 0 up to 8 MPa and

the “high surface coverage" range from 8 MPa up to 20MPa.

The observation of a decreasing CO2/CH4 excess sorption ratio at high pressure can be

explained by the increasing density ratio of (supercritical) gas and adsorbed phase,

which is more pronounced for CO2 than for CH4 because of the lower reduced

temperature (T/Tc = 1.05 for CO2 and T/Tc=1.66 for CH4) (Figure 60). This implies that

the influence of the adsorbed phase density decreases with increasing reduced

temperature. It is not surprising that when calculating the sorption capacity ratios based

on absolute sorption capacities no significant difference at high pressures is observed

(Figure 60).

Influence of coal rank Figures 59 and figure 60 illustrate the influence of coal rank and pressure (surface

coverage) on the CO2/CH4 sorption capacity ratio. This ratio is generally highest at low

pressures and decreases with increasing surface coverage. Furthermore it decreases

with increasing maturity (Figure 60). The differences in sorption capacity ratio between

the different coal maturities decrease drastically with increasing surface coverage. We

conclude therefore that

i) the observed differences are controlled by the high energy sorption sites and/or

ii) at high surface coverage the sorption processes are dominated by the adsorbate-

adsorbate interaction (between adsorbed molecules) (Gensterblum et al., 2013b) and by

weak adsorbent-adsorbate interactions such as carbon-CH4 and carbon-CO2

interactions.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

208 | P a g e

As already introduced, for an ideal sorbate the total amount of sorption sites (n) can

be considered temperature-independent. This means that at higher temperatures,

higher pressures are required to achieve the same (isosteric) surface coverage. For

moisture-equilibrated coals the amount of sorption sites occupied by water molecules

will decrease with increasing temperature due to the exothermal sorption of water on

the coal surface. This decrease will be negligible in comparison to CO2 and CH4 because

the heat of adsorption for water (up to 63 kJ/mol) is much higher than that for CO2 and

CH4 (max. 20kJ/mol) (Darcey, 1958) .

Influence of moisture on the CO2/CH4 sorption capacity ratio The CO2/CH4 sorption capacity ratio is higher (especially at a low surface coverage) and

the drop with increasing surface coverage. These differences between dry and moist

samples at low surface coverage decrease with increasing rank (Figure 59). For the

subbituminous coal the reduction in sorption capacity ratio at low surface coverage

decreases from 5.1 for the moist sample to 3.6 for the dry sample. A slightly smaller

reduction from 3.8 down to 2.9 is observed for the high-volatile bituminous coal. Finally,

for the anthracite the smallest reduction occurs (2.5 down to 2.1). Thus, we can conclude

that also the influence of water on the individual sorption capacities decreases with

increasing coal rank, while the difference in CO2/CH4 sorption capacity ratio between

the dry and moist samples get smaller with increasing pressure. This could be due to

firstly high molecular specific affinities of the high energetic sorption sites (low surface

coverage) and secondly due to low molecular specific affinities due to the less

favourable ‘van der Waals’-interactions between H2O, CH4 or CO2 and the carbon

molecules of the coal surface at high surface coverage. This decrease in the amount of

functional groups with increasing rank is caused by the decarboxylation and

dehydration during the coalification.

The decrease of the CO2/CH4 sorption capacity ratio caused by the pre-adsorbed water,

implying that the CH4 sorption sites are more readily occupied by water than the CO2

sorption sites. However, at high pressures the CO2/CH4 ratios are almost equal. At high

pressures (high surface coverage) the primary sorption sites (functional groups) are

already occupied and the sorption behaviour of CH4 or CO2 is dominated by interactions

with carbon atoms at the coal surface. These carbon-related sorption sites are equally

unattractive for water and for CH4 or CO2. Therefore the sorption capacity ratios become

smaller with increasing surface coverage. The major affinity difference between CO2 and

CH4 is observed for the primary (high energetic) sorption sites. For low-rank coals these

sites are attributed to oxygen-containing functional groups (Joubert et al., 1973b;

Joubert et al., 1974) and for high rank coals (anthracites) these sites are attributed to

micro and ultra-micro pores with large aromatic rings.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

209 | P a g e

Figure 59: CO2/CH4 excess sorption capacity ratio at 318K on dry (black) and moist (grey)

coal samples as a function of gas pressure. The ratio is calculated from the fitted excess

sorption functions

Figure 60: Absolute CO2/CH4 sorption capacity ratio at 318K on dry (black) and moist

(grey) coal samples. The calculated ratio is based on the fitted excess sorption

functions.

0

1

2

3

4

5

6

0 5 10 15 20

exc

ess

so

rpti

on

cap

acit

y ra

tio

D

AF

(C

O2/

CH

4)

pressure (MPa)

318Kdry & moist

subbituminous A

high-volatile bituminous

anthracite

0

1

2

3

4

5

6

0 5 10 15 20

abso

lute

so

rpti

on

cap

acit

y ra

tio

D

AF

(C

O2/C

H4)

pressure (MPa)

318Kdry & moist

subbituminous A

high-volatile bituminous

anthracite

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

210 | P a g e

Temperature

The influence of temperature (318K vs. 334K) on the CO2/CH4 sorption capacity ratio is

negligible for dry coals at low pressures. At higher pressures (20 MPa) and for dry coals

the CO2/CH4 sorption capacity ratios are higher for the 334K isotherms than for the

318K isotherms. At this pressure the excess sorption isotherms are controlled by the

density ratio between the free and the adsorbed phase. Therefore this observation

reflects a lower density ratio between gas- and adsorbed phase at higher temperatures.

The influence of temperature (318K vs. 334K) on the CO2/CH4 sorption capacity ratio is

higher when moisture is present especially for the subbituminous coal. The moist

subbituminous coal sample shows the highest difference (140%) at low pressures

between the CO2/CH4 sorption capacity ratios for 318 K and 334 K.

However for the anthracite the difference between the sorption capacity ratios at 318

and 334K is generally small. The CO2/CH4 excess sorption capacity ratio of 2.9 calculated

from the measured isotherms at 334K is slightly higher than the ratio of 2.7 at 318K.

Conclusion

Based on the results of sorption isotherm measurements with methane and carbon

dioxide on three coals of different rank at different temperatures and moisture

conditions the following conclusions could be reached (Table 23):

The excess sorption capacity of coals is always higher for CO2 than for CH4. At

least a 2:1 ratio for CO2 to CH4.

The CO2 and CH4 sorption capacity of dry coals shows the expected parabolic-like

shape with a minimum approximately around 1% vitrinite reflectance. This

parabolic trend is more pronounced for CO2 than for CH4. For moisturized coal

samples a linear increasing trend in CO2 and CH4 sorption capacity with

increasing coal maturity has been observed. These trends support previous

findings summarized in Busch and Gensterblum 2011.

Moisture content has a much higher impact on gas sorption capacity than

temperature or moderate variations in coal rank (in the temperature ranges

considered here and maturity variations in range of tenth of percent of vitrinite

reflectance).

When moisture is present, lower sorption capacities are generally observed. This

reduction in CO2 and CH4 sorption capacity becomes smaller with increasing coal

rank. The reduction in sorption capacity correlates linearly with the oxygen

content, which in turn correlates with the amount of hydrophilic and carboxylic

functional groups as observed by FTIR analysis.

The affinity of CH4 or CO2 to the coal is much higher in the absence of pre-

adsorbed water. This is evidenced by Langmuir pressures being 33-58% lower

for dry coals than for moisturised coals.

The preferential sorption of CO2 over CH4 is the highest at low rank coals and low

pressures. This is related to the availability of functional groups (for low rank

coals) or long aromatic rings and micro and ultra-micro pores (high rank coals).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

211 | P a g e

The major affinity difference between CO2 and CH4 can be observed for the high

energetic sorption sites. For low rank coals are the sites attributed to oxygen-

containing functional groups.

The CO2/CH4 excess sorption capacity ratio is higher (especially at a low surface

coverage) and its decrease with increasing surface coverage is much more

pronounced when water is present. Therefore the CH4 sorption sites are more

efficiently competitively occupied by water than the CO2 sorption sites.

We observe linear increasing Langmuir sorption capacities with coal rank for

moist coals. The sorption capacity of dry coals shows the expected parabolic (“u-

shaped”) trend with a minimum approximately around 1.1 to 1.3% vitrinite

reflectance.

The contribution of pore size distribution is of lower importance for low rank

coals due to the presence of functional groups. However for high mature coals

like anthracites the pore radius distribution is dominant because most of the

functional groups are decomposed during the coalification process.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

212 | P a g e

Table 23: Overview of methane and carbon dioxide sorption behavior on coals investigated in this study. Columns 3 through 6 summarize

the dependence of moist vs. dry and CO2 vs. CH4 sorption capacities on surface coverage based on Figs. 58, 60 and 61. The dependence of

the Langmuir sorption parameters (sorption capacity and affinity or Langmuir pressure) on moisture state and gas type is outlined in rows

7 trough 9.

Coal type Structural

characteristics

Surface coverage dependencies

Sorption

properties

Langmuir isotherm properties

moist/dry CO2/CH4 Moist/dry CO2/CH4

0<θ<0.3 Θ<0.7 0<θ<0.3 Θ<0.7

Affected by

pre- adsorbed

water

preferential adsorption

Sub-

bituminous

meso PSD

high amount of

oxyFG

highly

affected

by

pread.w.

highly

affected

by

pread.w.

very high

preferential

adsorption

high

preferential

adsorption

sorption

capacity

CH4: highly

CO2: highly

CO2 >> CH4

CO2 has a 2.3 (dry) and 2.1(moist)

times higher than CH4 (Θ=1)

affinity CH4: highly

CO2: moderate

CO2 >> CH4

CO2 has a 1.8 times higher than

CH4

High-

volatile

bituminous

meso- to micro

PSD

small amount of

oxyFG

highly

affected

by

pread.w.

minor to

not

affected

by

pread.w.

high

preferential

adsorption

moderate

preferential

adsorption

sorption

capacity

CH4: minor

CO2: moderate

CO2 >> CH4

CO2 has a 2.1 (dry) and 1.8 (moist)

times higher than CH4 (Θ=1)

affinity CH4: highly

CO2: moderate

CO2 > CH4

CO2 has a 1.6 times higher than

CH4

anthracite

micro to ultra-

micro pore PSD

no oxyFG

large aromatic

rings

moderate

affected

by

pread.w.

not

affected

by

pread.w.

moderate

preferential

adsorption

minor

preferential

adsorption

sorption

capacity

CH4: not

CO2: minor

CO2 > CH4

CO2 has a 1.3 (dry and moist)

times higher than CH4 (Θ=1)

affinity CH4: minor

CO2: minor

CO2 > CH4

CO2 has a 3 times higher than CH4

pread.w. = preadsorbed water, PSD = pore size distribution , oxyFG= oxygen containing functional groups

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

213 | P a g e

Appendix

Table A1: Therefore three slightly different 3-parameter fitting sets show an excellent

approximation to the experimental data of (only) one isotherm. The fitting residual is

below the experimental error.

SUB-BITUMINOUS COAL

PL

(MPa) n

(mmol/g)

ρads

(kg/m³)

n

(mmol/g)

moist, CH4

319K 12.39 0.846 373 0.004

319K 8.64 0.643 600 0.002

319K 7.15 0.564 900 0.001

Figure A2: Therefore three slightly different 3-parameter fitting sets show an excellent

approximation to the experimental data (using only one isotherm). This indicates

different interpretation for total sorption capacity, Langmuir pressure and adsorbed

phase density.

0

0.1

0.2

0.3

0.4

0.5

0.6

0 2 4 6 8 10 12 14 16 18

exc

ess

so

rpti

on

(mm

ol/

gDA

F)

pressure (MPa)

318.65

318.65

318.65

abs.sorp ads.phase dens. 900kg/m³

abs.sorp ads.phase dens. 600kg/m³

abs.sorp ads.phase dens. 373kg/m³

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

214 | P a g e

Symbols Units Physical

θ [-] Surface coverage

n [mol kg-1] Total specific sorption capacity, Langmuir sorption

amount of substance or Langmuir volume

nsorbed [mol kg-1] Amount of sorbed gas per kg coal

Δn [mol kg-1] Square error sum

Hp

0 [mmol gcoal -1 MPa-

1 ]

Henry coefficient

PL [MPa] Langmuir pressure

P0 [MPa] Pre-exponential factor for the temperature dependence

of the Langmuir pressure

dHK [J mol-1]

Sorption differential enthalpy of the dynamic equilibrium

for ad- and desorption

(often defined unclearly as isosteric heat of adsorption)

dHN [J mol-1 ] Sorption differential enthalpy of the sorption sites

dS [J mol-1 K-1] Sorption entropy

T [K] Temperature

p [MPa] Gas pressure

R =

8.31451

[J mol-1 K-1] Gas constant

ρadsorbed [kg m-3] Density of the adsorbed phase

ρfree [kg m-3] Gas density calculated by GERG

VRr [%] Vitrinite reflectance (mean random)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

215 | P a g e

References

Behar, F. and Vandenbroucke, M., 1987. Chemical modelling of kerogens. Organic

Geochemistry, 11(1): 15-24.

Busch, A. and Gensterblum, Y., 2011. CBM and CO2-ECBM related sorption processes in coal:

A review. International Journal of Coal Geology, 87: 49-71.

Busch, A., Gensterblum, Y. and B.M., K., 2003. Methane and CO2 sorption and desorption

measurements on dry Argonne Premium Coals: Pure components and mixtures.

International Journal of Coal Geology, 55: 205-224.

Busch, A., Gensterblum, Y. and Krooss, B.M., 2007. High-pressure sorption of nitrogen,

carbon dioxide, and their mixtures on Argonne Premium Coals. Energy and Fuels,

21(3): 1640-1645.

Busch, A., Gensterblum, Y., Krooss, B.M. and Siemons, N., 2006. Investigation of high-

pressure selective adsorption/desorption behaviour of CO2 and CH4 on coals: An

experimental study. International Journal of Coal Geology, 66(1-2): 53-68.

Coppens, L., 1936. Ann. Mines Belg., 37: 173.

Darcey, J.R., Clunie, J.C.,Thomas, D.G., 1958. The adsorption of water by Saran charcoal

Transactions of the Faraday Society, 54: 250-256.

Day, S., Duffy, G., Sakurovs, R. and Weir, S., 2008a. Effect of coal properties on CO2 sorption

capacity under supercritical conditions. International Journal of Greenhouse Gas

Control, 2(3): 342-352.

Day, S., Sakurovs, R. and Weir, S., 2008b. Supercritical gas sorption on moist coals.

International Journal of Coal Geology, 74(3-4): 203-214.

Gasparik, M. et al., 2012. High-pressure methane sorption isotherms of black shales from the

Netherlands. Energy and Fuels, 26(8): 4995-5004.

Gensterblum, Y., Merkel, A., Busch, A. and Krooss, B.M., 2013. Molecular concept and

experimental evidence of H2O, CH4 & CO2 sorption on organic material. Fuel.

Gensterblum, Y. et al., 2010. European inter-laboratory comparison of high pressure CO2

sorption isotherms II: Natural coals. International Journal of Coal Geology, 84(2): 115-

124.

Gensterblum, Y. et al., 2009a. European inter-laboratory comparison of high pressure CO2

sorption isotherms. I: Activated carbon. Carbon, 47(13): 2958-2969.

Gensterblum, Y. et al., 2009b. European inter-laboratory comparison of high pressure CO2

sorption isotherms. I: Activated carbon. Carbon, 47: 2958-2969.

Given, P.H., Marzec, A., Barton, W.A., Lynch, L.J., Gerstein, B.C., 1986. The concept of a

mobile or molecular phase within the macromolecular network of coals: A debate.

Fuel, 65: 155-163.

Glymour, C., 1980. Theory and Evidence. Princeton University Press.

Goodman, A.L., Campus, L.M., Schroeder, L.M., 2005a. Direct Evidence of Carbon Dioxide

Sorption on Argonne Premium Coals Using Attenuated Total Reflectance-Fourier

Transform Infrared Spectroscopy. Energy and Fuel, 19: 471-476.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

216

Goodman, A.L., Favors, R.N. and Larsen, J.W., 2006. Argonne Coal Structure Rearrangement

Caused by Sorption of CO2. Energy & Fuels, 20(6): 2537-2543.

Goodman, A.L., Favrs, R.N., Hill, M.M., Larsen, J.W., 2005b. Structure changes in Pittsburgh

No. 8 coal causewd by sorption of CO2 gas. Energy and Fuels, 19: 1759-1760.

Gutierrez-Rodriguez, J.A., Purcell, R.J. Jr.,Aplan F.F., 1984. Estimating the Hydrophobicity of

coal. Colloids and Surfaces, 12: 1-25.

Heuchel, M., Davies, G.M., Buss, E., Seaton, N.A., 1999. Adsorption of Carbon Dioxide and

Methane and Their Mixtures on an Activated Carbon: Simulation and Experiment.

Langmuir, 15: 8695-8705.

Humayun, R. and Tomasko, D.L., 2000. High-Resolution Adsorption Isotherms of Supercritical

Carbon Dioxide on Activated Carbon. AlCHE Journal, 46(10): 2065-2075.

Joubert, J.I., Grein, C.T. and Bienstock, D., 1973. Sorption of methane in moist coal. Fuel,

52(3): 181-185.

Joubert, J.I.G., T. & Bienstock, D., 1974. Effect of moisture on the methane capacity of

American coals. Fuel, 53: 186-191.

Kunz, O., Klimeck, R., Wagner, W. and Jaeschke, M., 2007. The GERG-2004 wide-range

reference equation of state for natural gases. GERG Technical Monograph 15

Fortschritts-Berichte Verein Deutscher Ingenieue (VDI), 6. VDI-Verlag, Düsseldorf.

Larsen, J.W., 2004. The effects of dissolved CO2 on coal structure and properties. Int J. Coal

Geology, 57: 63– 70.

Li, D. et al., 2010. High-pressure sorption isotherms and sorption kinetics of CH4 and CO2 on

coals. Fuel, 89(3): 569-580.

Liu, Y. and Wilcox, J., 2012a. Effects of surface heterogeneity on the adsorption of CO 2 in

microporous carbons. Environmental Science and Technology, 46(3): 1940-1947.

Liu, Y. and Wilcox, J., 2012b. Molecular simulation of CO2 adsorption in micro- and

mesoporous carbons with surface heterogeneity. International Journal of Coal

Geology, 104(0): 83-95.

Lynch, L.J. and Webster, D.S., 1982. Effect of thermal treatment on the interaction of brown

coal and water: a nuclear magnetic resonance study. Fuel, 61(3): 271-275.

Mu, R. and Malhotra, V.M., 1991. A new approach to elucidate coal-water interactions by an

in situ transmission FT-i.r. technique. Fuel, 70(10): 1233-1235.

Murata, K., El-Merraoui, M. and Kaneko, K., 2001. A new determination method of absolute

adsorption isotherm of supercritical gases under high pressure with a special

relevance to density-functional theory study. The Journal of Chemical Physics, 114(9):

4196-4205.

Nishino, J., 2001. Adsorption of water vapor and carbon dioxide at carboxylic functional

groups on the surface of coal. Fuel, 80: 757-764.

Ozdemir, E. and Schroeder, K., 2009. Effect of Moisture on Adsorption Isotherms and

Adsorption Capacities of CO2 on Coals. Energy & Fuels, 23(5): 2821–2831.

Pini, R., Ottiger, S., Burlini, L., Storti, G. and Mazzotti, M., 2010. Sorption of carbon dioxide,

methane and nitrogen in dry coals at high pressure and moderate temperature.

International Journal of Greenhouse Gas Control, 4(1): 90-101.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

217

Prinz, D. and Littke, R., 2005. Development of the micro- and ultramicroporous structure of

coals with rank as deduced from the accessibility to water. Fuel, 84(12-13): 1645-

1652.

Prinz, D., Pyckhout-Hintzen, W. and Littke, R., 2004. Development of the meso- and

macroporous structure of coals with rank as analyzed with small angle neutron

scattering and adsorption experiments. Fuel, 83: 547-556.

Ruthven, D.M., 1984. Principles of adsorption and adsorption processes. Wiley.

Sakurovs, R., Day, S. and Weir, S., 2010. Relationships between the critical properties of

gases and their high pressure sorption behavior on coals. Energy and Fuels, 24(3):

1781-1787.

Sakurovs, R., Day, S. and Weir, S., 2013. Relationships between the sorption behaviour of

methane, carbon dioxide, nitrogen and ethane on coals. Fuel, 97(0): 725-729.

Sakurovs, R. et al., 2012. Pore size distribution and accessible pore size distribution in

bituminous coals. International Journal of Coal Geology, 100(0): 51-64.

Schuyer, J., Blom, L. and Vankrevelen, D.W., 1953. The molar refraction of condensed

aromatic compounds. Transactions of the Faraday Society, 49(12): 1391-1401.

Suárez, N., Laredo, E. and Nava, R., 1993. Characterization of four hydrophilic sites in

bituminous coal by ionic thermal current measurements. Fuel, 72(1): 13-18.

Svábová, M., Weishauptová, Z. and Pribyl, O., 2012. The effect of moisture on the sorption

process of CO2 on coal. Fuel, 92(1): 187-196.

Tenney, C.M. and Lastoskie, C.M., 2006. Molecular simulation of carbon dioxide adsorption

in chemically and structurally heterogeneous porous carbons. Environmental

Progress, 25(4): 343-354.

van Krevelen, D.W., 1993. Coal: Typology-Physics-Chemistry-Constitution. Elsevier, 1002 pp.

Weast, R.C., 1972. Handbook of Chemistry and Physics (53rd Edn.). Chemical Rubber Co.,

Cleveland.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

218 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

219 | P a g e

True wisdom comes to each of us when we realize how little we understand about life,

ourselves, and the world around us.

(Sokrates)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

220 | P a g e

6. Gas saturation and CO2 enhancement

potential of coalbed methane

reservoirs as a function of depth

AAPG, 2013 (in press)

Yves Gensterblum,

Alexej Merkel, Bernhard M. Krooss and Ralf Littke

RWTH-Aachen University,

Institute of Geology and Geochemistry of Petroleum and Coal, Lochnerstr. 4-20,

D-52056 Aachen, Germany

Andreas Busch

Shell Global Solutions International,

Kessler Park 1, 2288GS Rijswijk, The Netherlands

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

221 | P a g e

Abstract

The influence of moisture, temperature, coal rank and differential enthalpy on the

methane (CH4) and carbon dioxide (CO2) sorption capacity of coals of different rank has

been investigated by using high-pressure sorption isotherms at 303, 318, 333K (CH4)

and 318, 333, 348K (CO2), respectively. The variation of sorption capacity was studied as

a function of burial depth of coal seams by using the corresponding Langmuir

parameters in combination with a geothermal gradient of 0.03 K/m and a normal

hydrostatic pressure gradient. Taking the gas content corresponding to 100% gas

saturation at maximum burial depth as a reference value the theoretical CH4 saturation

after uplift of the coal seam was computed as a function of depth. According to these

calculations the change in sorption capacity caused by changing p, T conditions during

uplift will lead consistently to oversaturation. Therefore the frequently observed

undersaturation of coal seams is most likely related to dismigration (losses into adjacent

formations and atmosphere). Finally we attempt to identify sweet spots for CO2-

enhanced coal bed methane production. CO2-ECBM is expected to become less effective

with increasing depth because the CO2 to CH4 sorption capacity ratio decreases with

increasing temperature and pressure. Furthermore CO2-ECBM efficiency will decrease

with increasing maturity due to the highest sorption capacity ratio and affinity

difference between CO2 and CH4 for low mature coals.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

222 | P a g e

Introduction

World gas consumption has steadily increased over the past decades leading to a

depletion of conventional gas reserves. The end of nuclear power generation in Germany

in 2022 will lead to an increase in natural gas consumption especially if CO2 reduction

will be part of the European energy strategy. Due to the fact that natural gas has the

smallest carbon footprint of the conventional energy sources (like oil, coal, etc.) (Cathles

et al., 2011), its importance will increase (Birol, 2012). In the future, gas supply can only

be maintained if additional gas resources are utilized. In this context, gas production

from unconventional reservoirs has substantially increased in recent years, especially in

the USA and Australia. In Europe, exploration of these unconventional reservoirs,

including tight gas, coalbed methane and gas-shales, has greatly expanded over the last

decade (Littke et al., 2011).

Depending on the structural and tectonic setting of the sedimentary basin, coal seams of

similar depositional origin may be encountered at different depth levels, i.e. under

different temperature and fluid pressure conditions. Usually this involves also minor or

major variations in thermal maturity (coal rank). The variations in (sorptive) storage

capacity as a function of temperature can be estimated by using the thermodynamic

parameters of sorption enthalpy and entropy. Furthermore the gas saturation due to

field-variations in the pressure and temperature conditions can be estimated. This paper

presents some fundamental considerations and observations, mainly based on

laboratory experiments, to outline how sorption capacity, methane saturation and

recovery potential of coal seams can be estimated as a function of present-day and past

burial depth and temperature.

Hydrocarbons in coal seam gas are derived from either thermal breakdown of kerogen

or microbial generation processes (fermentation and CO2 reduction). The gas in place

(GIP) content [mol] of coal seams is a crucial parameter determining the economics of

gas resources. The CBM GIP estimations take into account: (1) a sorbed phase consisting

of gas molecules adhering to the surface of the coal matrix and having a density that

approaches that of a liquid, nsorbed [mol kg-1] is the amount of adsorbed gas (2) free gas

within the pores and natural fractures (nporosity [mol kg-1]) and (3) gas dissolved in

formation water in the coalbed (nformation water [mol kg-1]), which can be considered

negligible for methane:

(46)

Here is the coal seam volume and the bulk density of the coal (

Table 30). Another important parameter is the present-day gas saturation, defined as

the ratio of the gas content of the coal seam (per unit mass or volume) and the maximum

sorption capacity. Gas saturation indicates how much gas can be produced and how high

the recovery factor will be.

(47)

Where ntotal denotes the total amount of gas (desorbable, free gas and dissolved in

formation waters) stored in a volume element of coal at a certain stage during burial

history, and nabsolute the maximum amount of gas that can theoretically be stored in a

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

223 | P a g e

volume element of coal under the in-situ pressure and temperature conditions (Hol et

al., 2013). However, the determination of the methane saturation is often prone to

errors- and difficult to determine since it cannot directly be derived from well logs

(Diamond and Schatzel, 1998; Mares et al., 2009; Mavor, 1995). Gas content is usually

determined by canister desorption, while the sorption capacity as a function of pressure

and temperature is determined by high-pressure sorption experiments.

Figure 61: Calculated CH4 storage capacity in the gas-saturated porosity of a coal seam

(assumed coal density 1.4 g/cm³) using the EOS for methane (Setzmann and Wagner,

1991) and the maximum dissolved CH4 content in the formation water under the

assumption of a water-filled pore system. A thermal gradient of 0.03 K/m and

hydrostatic pressure gradient of 0.01MPa/m was used for the calculations (1 mmol/g

= 24.5 Std. m³/ton). The porosity was kept constant with depth and we did not take

into account the compaction of the coal during burial.

The amount of free compressed gas nporosity [mol kg-1], which is stored in the open

porosity and cleats of a coal seam is usually determined during drilling and mainly

depends on water saturation. However, the laboratory estimation of porosity (using low

pressure mercury injection in combination with helium pycnometry) is usually done

without external confining stress; porosity values determined by mercury injection on

coals tend to be unreliable; therefore, due to pore elasticity, cleat volume

compressibility and different penetration properties, porosity values are generally

overestimated for coals.

Recent studies indicate, however, that rather than the composition and rank of the

organic matter, water content has a dominant or even overriding effect on the gas

storage capacity of coals (Chapter 1, 4 and 5). The dramatic influence of water content

on methane storage capacity of coals is illustrated and explained in Chapter 4 and 5. To

account for this effect and arrive at more realistic values for sorption capacities,

sorption measurements are increasingly performed on moisture-equilibrated samples

(Clarkson and Bustin, 2000; Crosdale et al., 2008; Day et al., 2008c; Goodman et al.,

0

500

1000

1500

2000

2500

3000

3500

4000

0.00001 0.0001 0.001 0.01 0.1 1

dep

th (m

)

CH4 stored in pore system (mmol/g)

compressed methane (1% free porosity)

compressed methane (4% free porosity)

methane - solubility (4% porosity - brine 50mg/L)

methane - solubility (1% porosity - brine 50mg/L)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

224 | P a g e

2007b) and a summary is provided by Busch and Gensterblum (2011). The methane

sorption capacity of moisture-equilibrated coals is only in the order of 40-60% of the

sorption capacity of the dry samples (Busch and Gensterblum, 2011). Usually the

sorption capacity of the “as received” coals varies between these two values.

Figure 61 illustrates the theoretical gas content in the accessible porosity (free of water)

as a function of depth using the Setzmann and Wagner EOS for CH4 (Setzmann and

Wagner, 1991). The assumed coal matrix density is 1.4 g cm-3. A thermal gradient of 0.03

K/m and a hydrostatic gradient of 0.01 MPa/m have been assumed. The content of

dissolved gas in the formation water (nformation water, mmol g-1) is very low (Duan et al.

1992) however, for CO2 it is not (Duan and Sun, 2003) and is a function of brine salinity,

pressure and temperature. The solubility of CH4 decreases with increasing salinity of the

formation water (O'Sullivan and Smith, 1969).

In order to determine the desorbable gas content needed for GIP determination, the lost

gas content (nlost gas, mmol g-1) is the fraction of gas desorbed while bringing the core

sample to the surface during drilling and sealing it into desorption canisters (lost gas

time) (Smith, 1984b). Therefore nlost gas is an empirically assumed value. It is commonly

determined by backward extrapolating the measured desorbed gas data points collected

at reservoir temperature over the lost gas time interval to time zero (Metcalfe et al.,

1991). Determining the lost gas volume requires accurate determination of the desorbed

gas content (ndesorbed). The desorbed gas is commonly measured by determining the

amount of desorbed gas ndesorbed during canister desorption tests. The residual gas

content nresidual [mmol g-1] remains adsorbed in the coal sample at the end of canister

desorption measurements. It is commonly measured by pulverising the sample and

measuring the released gas content. This results in the total desorbable gas content

[mmol g-1]:

(48)

It is important to be aware that not all of the desorbed or residual gas has to be

methane; usually some smaller amounts of CO2, N2 or higher carbon number

hydrocarbon gases are part of the desorbed gas. Therefore, additional gas composition

analysis is required. The residual gas content is also crucial for the gas recovery

estimation, because due to the strong “sorptive trapping” the residual gas content

indicates the amount of gas that is either situated in closed porosity or can only be

produced by pressure drawdown to very low reservoir pressures. Therefore the

residual gas content is able to provide an estimation of the ‘unproducible’ gas content.

By substituting equation 50 into 49 we obtain the predicted in-situ gas saturation.

(49)

Sorbed gas is the main storage mechanism for CBM because of the low solubility of CH4

in water and the generally low porosity of coal seams (Figure 61). There are several

methods to determine the sorbed gas content (Diamond and Schatzel, 1998). It has been

documented by numerous studies that the sorption properties of coals may vary

systematically with rank and/or maceral composition (e.g. Busch and Gensterblum

2011). Due to the observation that the sorption capacity also strongly depends on the

moisture content (Crosdale et al., 2008; Day et al., 2008c), which ranges between ”as

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

225 | P a g e

received” and (moisture) saturated, determining the sorption capacity on such coals

becomes more complicated (Goodman et al., 2007b) (Figure 62). An oversaturated coal

seam would be highly beneficial; albeit, they are rather scarce (Figure 63).

Oversaturated coal seams develop when the amount of generated gas is higher than the

loss of gas due to changes in sorption capacity as a result of basin uplift, migration, and

diffusion. Therefore, the coal seam is assumed to be of low-permeability and should be

sealed by a tight caprock to maintain high CH4 concentrations over geological time

scales. In general the geological reasons for oversaturated coal seams could be a special

geological reservoir situation in combination with very low permeable caprocks and or

high microbial gas generation (Clayton, 1998; Faiz, 2004; Krooss et al., 1988).

Alternatively and likely as well could be the experimental determination of the sorption

capacity or desorbed gas content. For instance an underestimation of the CH4 sorption

capacity at reservoir pressure, temperature and moisture conditions or an

overestimation of the lost gas (determined by exponential extrapolation of canister

desorption data) would lead to an apparent oversaturation. In addition, the adjusted

moisture content of the sorption isotherm determined in the laboratory is higher than

that within the coal seam (Diamond and Schatzel, 1998; Mares et al., 2009; Zhao et al.),

hence lower sorption capacities compared to the in-situ capacities are determined.

The reasons for methane undersaturated coal seams are i) less gas from neighbouring or

the same coal seam was generated than the coal could possibly adsorb or ii) a certain

amount of gas is lost because of burial and uplift or dissolution of CH4 in formation

brine.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

226 | P a g e

Figure 62: Schematic illustration of CH4 sorption isotherms, amount of producible gas n

and the effect of CH4 affinity (Langmuir pressure pL is a function sorption enthalpy dH)

for two different coal ranks. The figure illustrates the importance of a Langmuir

pressure as an indicator (reciprocal) for the affinity between adsorbent and adsorbate.

Furthermore, the Langmuir pressure is a function of the sorption enthalpy (Equation

56 and following equation). The parameter set for this figure is taken from Table 26

and Table 28. The interpolated CH4 Langmuir pressures at a temperature of 327 K

which corresponds to a depth of 1300 m are 12.3 MPa and 8.3 MPa for the sub-

bituminous A and high-volatile bituminous coal, respectively. The highlighted pressure

levels are PR for initial reservoir pressure prior to production and PA for abandonment

pressure, which is the reservoir pressure at the end of production. The total amount of

producible gas for the sub-bituminous A coal is 33% lower than for the high-volatile

bituminous coal. Assuming the same total sorption capacity for both coals have the

producible gas is still 18% lower for the sub-bituminous A compared to the high-

volatile bituminous coal. Therefore we conclude high Langmuir pressures are less

favourable for the amount of producible gas.

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0 2 4 6 8 10 12 14

abso

lute

so

rpti

on

cap

acit

y (m

mo

l/g)

CH4 partial pressure (MPa)

sub-bituminous

high-volatile bituminous

PL (dH1) > PL(dH2)

n(pL(dH1))

n(pL(dH2))

PAPR

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

227 | P a g e

Scott (2002) explains this as an interplay of coalification, gas generation, the

hydrogeological situation and burial history (Scott, 2002). Experimental errors and their

impact should also be considered when defining over- or undersaturation. Experimental

gas sorption isotherms are very sensitive to the moisture content of the coal (Busch and

Gensterblum, 2011). Due to this fact, moisture content is an important parameter

affecting sorption capacity determinations. Furthermore the Langmuir pressure

decreases with increasing moisture content. Figure 62 illustrates that with increasing

Langmuir pressure, productivity improves, because the amount of gas, which will be

desorbed, is higher for the same production pressure.

Figure 63: Schematic illustration of the methane sorption isotherm, the degree of

saturation and the amount of producible gas (modified after (Bustin and Bustin, 2008)).

The two dots mark the amount of gas stored at reservoir pressure. The highlighted

pressure levels are the initial reservoir pressure PR, the initial desorption pressure PD

(also called critical desorption pressure (Ayers, 2002)), and the abandonment pressure

PA.

The most favoured explanation for coal seams being undersaturated with methane is

uplift. This argumentation however, mainly depends on the sorption enthalpy and the

temperature gradient in the respective basin (Figure 64). At maximum burial depth and

burial temperature (Tmax) more gas can be generated than the coal can possibly adsorb

(Clayton, 1998; Jüntgen, 1987; Kotarba and Rice, 2001; Rice et al., 1993). It is assumed

that the coal at maximum burial depth is either fully saturated or oversaturated. During

uplift and exhumation, hydrostatic pressure and temperature decrease. Assuming a

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

228 | P a g e

closed container, the amount of gas sorbed in the micropores and present in the pore

system remains constant. Under certain geological conditions additional gas may be

generated by microbial activities. It is commonly assumed that the sorption capacity will

increase during uplift, because sorption is an exothermic process (Figure 64, solid line).

This leads to an undersaturated coal seam, if the total amount of gas in the coal

(compressed + sorbed) remains unchanged during uplift. However, in order to

understand gas saturation during uplift both, Langmuir pressure PL and heat of

adsorption have to be considered. Langmuir pressure can be regarded as a measure for

the general affinity between methane and coal: If PL is low the affinity is high since large

amounts of methane are adsorbed at relatively low pressures and vice versa. The heat of

adsorption on the other hand describes the affinity of methane sorption on coal with

changes in temperature. Therefore, due to the decreasing hydrostatic pore pressure and

the presence of water, the sorption capacity as a function of depth is more complex

(Figure 64, dashed line).

Figure 64: Schematic illustration of the CH4 sorption capacity (at constant n) with depth

and influence of differential enthalpy on the Langmuir pressure caused by a dry (dH=-

22kJ mol-1; ln(K)= 9.6 and n = 0.99 mmol g-1) and moist sub-bituminous coal (dH=-17.9

kJ mol-1; ln(K)= 9.09 and n = 0.9 mmol g-1) (Gensterblum et al. 2012).

Beside these considerations for equilibrated geological systems, we need to consider the

coalification process as a function of burial history (Equation 50). Obviously, the

sorption capacity is a function of coal maturity or rank, expressed here as vitrinite

reflectance VRr (Busch and Gensterblum, 2011). Coalification (maturation) is regarded

exclusively a function of temperature T and coalification time t.

(50)

0

500

1000

1500

2000

2500

3000

0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80

de

pth

(m

)

CH4 sorption capacity (mmol/g)

Sub-bituminous - dry

Sub-bituminous - moist

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

229 | P a g e

The correction (coalification) function k, which is specific for each basin, must be applied

to determine the actual coalification time. This factor is equal to the ratio of the time

integral of the burial history and the product of exposure time to gas generated and

maximum temperature which is a function of burial depth, although stress has also some

influence, especially at high levels of maturation in tectonically active areas (Littke et al.,

2012). These models for vitrinite maturation (coalification) have limitations and can be

divided into six categories: (i) models that describe vitrinite maturation explicitly as a

function of temperature (Barker and Pawlewicz, 1986; Price, 1983), (ii) models that

incorporate time as a rule of thumb (Bostick, 1983; Lopatin, 1971; Walpes, 1980), (iii)

models that describe vitrinite maturation as an Arrhenius first order chemical reaction

with a single activation energy (Antia, 1986; Barker and Goldstein, 1990; Middleton and

Falvey, 1983), (iv) models that describe vitrinite maturation as an Arrhenius first-order

chemical reaction with a single activation energy while the activation energy is itself a

function of temperature (Waples et al., 1992a; Waples et al., 1992b), (v) models that

describe vitrinite maturation with parallel Arrhenius first-order chemical reactions with

a Gaussian distribution of activation energies (Hvoslef et al., 1988), and (vi) models

using an Arrhenius first-order parallel-reaction approach with a distribution (not

stringent Gaussian) of activation energies (Burnham and Sweeney, 1989; Krooss, 1988;

Krooss et al., 1995b; Sweeney and Burnham, 1990). Besides methane, some coal basins

contain significant amounts of CO2, which can even be the dominant gas component like

in parts of the Silesia Basin (Kotarba and Rice, 2001) or the Sydney Basin (Faiz, 2007;

Faiz, 2006). The source of CO2 in coalbed gases can be attributed to (i) decarboxylation

reactions of kerogen and soluble organic matter during burial heating of the coal

(Boudou et al., 2008), (ii) mineral reactions, like dissolution of carbonates, thermal

decomposition or metamorphic reactions, (iii) bacterial oxidation of organic matter

(Suess and Whiticar, 1989), and (iv) migration from deep-seated sources like magma

chambers or from the upper crust (Kotarba and Rice, 2001).

In this study we present some fundamental concepts how sorption capacity and gas

saturation change with depth and coal maturity. Furthermore, we show and discuss

some observations that may be useful to improve sweet-spot identification related to

coalbed methane and CO2-enhanced coalbed methane production. The identification of

high gas in place contents is one major aspect. A second aspect of no lesser importance is

reservoir permeability, which is commonly attributed to the cleat and fracture system of

the coal seam (Laubach, 1998). The relationship between cleat formation and in-situ

stress fields during burial and uplift is beyond the scope of this paper and will be treated

elsewhere.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

230 | P a g e

Samples and sample preparation

In order to investigate the effect of coal rank on the thermodynamics of gas sorption

(enthalpy and entropy), three coal samples of different maturity were used in this study:

two German coal samples of intermediate and high rank from the Ruhr Basin, Germany

and one low-rank coal from the Surat Basin, Queensland, Australia.

Geological setting

The Prosper Haniel and Ibbenbüren coal samples were obtained from two German coal

mines in the Ruhr Basin and are of Upper Carboniferous (Pennsylvanian) age. These

coals formed in a tropical paralic environment (Littke et al., 1988; Littke et al., 1990).

Because of the complex burial history of the Ruhr Basin, the coals show a wide range of

rank from high-volatile bituminous (hvb) coal to anthracite. The Westphalian B seams of

the Prosper-Haniel mine contain hvb coal. Within the Prosper Haniel mine, located near

Bottrop, coal seams are mined down to a depth of 1200 m. The sample used in this study

was obtained from a depth of 915 m and has a random vitrinite reflectance VRr of 0.93%

(Table 24 and Table 25).

Table 24: Coal samples used in this study

Location Basin Lithostratigraphic

Units Age Seam Rank

QLD

Australia

Surat

Basin

Taroom Coal

measures Middle Jurassic Condamine

sub-bituminous

A

Prosper

Haniel

mine

Germany

Ruhr

basin Westphalia B Pennsylvanian K

high-volatile

bituminous

Ibbenbüren

Germany

Ruhr

basin Pennsylvanian 74 anthracite

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

231 | P a g e

Table 25: Ultimate and proximate analysis of coal samples used in this study

(* Equilibrium moisture content at 97% humidity; + dry basis)

QLD Australia

(Sub-bituminous

A)

Prosper Haniel

mine

Germany

(High-volatile

bituminous)

Ibbenbüren

Germany

(anthracite)

Volatile mater: + (%) 43.5±0.4 31.7±1.2 6.0

Fixed carbon: + (%) 49.5±0.1 58.0±1.7 88.2

Ash content: + (%) 3.5±0.2 7.3±2.2 5.8

As received (%)

moisture content 3.6±0.2 3.0 ±0.2 0.85

Equilibrium moisture*

(%) 6.4±0.1 4.0±0.5 4.8 ±0.3

Vitrinite or huminite

(%) 61.4 69.3 ±3.2 83

Liptinite (%) 31.4 7.1 ±2.3 0

Inertinite (%) 0.6 14.6±4.0 17

Mineral matter (%) 6.6 8.7±1.4 n.a.

Vitrinite refl. VRr (%) 0.5 0.93 3.3

The Ibbenbüren mine is located 80 km further north at the southern edge of the Lower

Saxony Basin (LSB) and is also of Upper Carboniferous (Pennsylvanian) age, but slightly

younger then the one from Prosper Haniel. It represents an isolated uplifted block,

which was exposed to an intense secondary coalification during the Cretaceous. The

Ibbenbüren coal is of anthracitic rank. The complex history of burial and uplift is still a

matter of debate (Senglaub et al., 2005; Senglaub et al., 2006). Basin modelling studies

suggest that the coal seams have undergone a maximum burial depth of 4500 to 5000 m

during late Carboniferous and early Permian time (Pennsylvanian) (Senglaub et al.,

2005). Anthracite is mined in Ibbenbüren at a depth down to 1650 m below surface. Gas

contents of the anthracite are generally high but vary among coal seams showing also

lateral variations. Mining strategies have been optimized to allow for gas drainage of

notoriously gas-rich seams by underground boreholes from over- and underlying

workings. The Ibbenbüren anthracite sample was taken from seam 74 at 1200 m depth

and has a vitrinite reflectance of 3.3%.

Coalbed methane production from the Surat basin, Australia started in 2007. The

Walloon Subgroup is well developed across the eastern Surat Basin and crops out along

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

232 | P a g e

the NE margin of the basin (Scott et al., 2007). The coals from the Wallon Subgroup are

of middle Jurassic age. The sub-bituminous A coal sample investigated (VRr=0.5%) is

taken from a cored well and belongs to the Condamine Seam, which is part of the

Taroom coal measures.

Sample preparation

For measurements on powdered coal material, crushed samples were divided and

aliquots were ground to pass a sieve size of 2 mm. The coal samples were subdivided

into four different particle-diameter fractions (0.177–0.354, 0.354–0.5, 0.5–1.0 mm and

greater than 1.0 mm). The sieve fraction below 0.177 mm was discarded, because it is

considered to be enriched in mineral fragments compared to the larger size fractions.

For sorption measurements, the sieve fraction of 0.5 to 1 mm was used. The coal

samples were moisture-equilibrated in a desiccator over periods of 5 to 14 days at a

relative humidity of 97% provided by a potassium sulphate solution (K2SO4), ASTM

procedure 1412-99. The sample material was then transferred immediately into the

adsorption cell. An aliquot was used for determination of the moisture content. A

summary, including available data for proximate and ultimate analysis, is provided in

Table 25.

The equilibrium moisture contents for the high-volatile bituminous coals and the

anthracite are relatively high compared to data reported by Busch and Gensterblum

(2012). However, the values are averages of 9 to 11 measurements and the standard

deviations indicate their accuracy and reproducibility. Therefore they are considered

reliable.

Experimental and conceptual procedure

All experimental sorption data reported here were obtained using the manometric

method. The set-up used for sorption isotherm measurements and for determining

sorption kinetics is explained in detail by Busch and Gensterblum (2011). Further

experimental details on this method were reported by Gensterblum et al. (2009, 2010).

Free phase gas densities of CH4 and CO2 as a function of pressure and temperature were

calculated using the GERG multi-component equation of state (EOS) (Kunz et al., 2007)

that provides also mixture densities for CO2 and water and CH4 and water.

For quality checks and to test consistency of the experimental results, volume

calibrations of the experimental set-up were performed with different gases (He, CH4

and CO2) by successive expansion steps and using the GERG EOS for the corresponding

gas.

Parameterisation of experimental sorption data The Langmuir equation is frequently used as a first approach to describe gas adsorption

on solid surfaces as a function of pressure and at constant temperature (sorption

isotherms) (Langmuir, 1916).

(51)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

233 | P a g e

Here n(p) [mmol g-1] is the amount of substance adsorbed at pressure p [MPa], n [mmol

g-1] is the amount of substance adsorbed at infinite pressure and PL [MPa] is the

Langmuir pressure (pressure at which half of the sorption sites are occupied). The

underlying assumptions (monomolecular coverage, energetically identical sorption sites

etc.) are discussed in Langmuir (1918) and textbooks on sorption science.

The Langmuir equation was derived and is predominantly used for low-pressure

sorption tests where the volume occupied by the adsorbed phase can usually be

neglected. This is justified because the density of the adsorbed phase is much higher

than the density of the free gas phase in equilibrium with it. In high-pressure sorption

experiments the density difference between the free gas and the adsorbed phase

becomes successively smaller and the volume of the adsorbed phase has measurable

effects in the experiments: in the manometric and volumetric methods it reduces the

void volume of the measuring cell and in the gravimetric method it generates additional

buoyancy. By performing a mass and volume balance explicitly taking into account the

volume of the adsorbed phase (which is proportional to the absolute amount of

substance adsorbed) at a given pressure, the following equation for the experimentally

measurable excess sorption can be derived:

(52)

Here is the density of the free gas phase, which is usually accurately known for

given pressure and temperature conditions, and is the density of the adsorbed

phase. The excess sorption is the parameter that is determined

experimentally (Sircar, 1999). It constitutes the net effect of removing gas molecules

from the free phase and depositing them on the surface of the sorbent (formation of an

adsorbed phase) and simultaneously increasing the volume of the “condensed phase”

(sorbent + adsorbate) by an unknown amount. In Equation 54, n(p) denotes the

“absolute sorption”. The absolute adsorption has the property of increasing

monotonously with increasing gas pressure and approaching a saturation when p→.

It is evident that for low gas pressures, as approaches zero, the difference between

excess sorption and absolute sorption vanishes.

Taking the Langmuir equation as a model equation for absolute sorption, combination of

Equations 51 and 52 yields:

(53)

In Equation 53 the Langmuir function may be replaced by any other reasonable

“absolute sorption” function.

The absolute sorption cannot be determined directly from sorption experiments

because the density of the adsorbed phase ( ) is not known, consequently this

parameter has to be estimated. It was found experimentally that the adsorbed phase

density for methane is close to the inverse of the van der Waals volume (Sakurovs et al.,

2010) of the adsorbate (Humayun and Tomasko, 2000). However many experimental

studies use 1027 and 373 kg m−3 for CO2 and CH4 as reasonably good approximations,

respectively (Gensterblum et al., 2010b; Gensterblum et al., 2009b).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

234 | P a g e

Temperature-dependence of gas sorption isotherms While the sorption isotherms describe the pressure-dependence of gas sorption on

solids, the temperature dependence has to be accounted for as a relevant parameter in

geological systems. In general only the Langmuir pressure (PL) is considered to be

temperature-dependent (Equation 53).

(54)

where n(T) and PL(T) are sorption capacity and Langmuir pressure of the gas at

temperature T and at a constant moisture content.

For dry coals the amount of sorption sites is constant over wide temperature ranges.

However for moisturised coals the situation is more complicated because the sorption

sites are partly occupied by water molecules. The amount of sorbed water molecules at a

defined partial pressure depends also on the temperature.

Due to the fact that sorption enthalpies for water on coals are much larger than those for

methane (HK and Hn) the total concentration of methane sorption sites (total sorption

capacity) n are is kept constant (HN=0) for the temperature range of this study

(Equation 56).

(55)

One of the objectives of this work was to derive, based on experimentally measured

sorption isotherms, a function describing the variation of the sorption capacity of

methane and CO2 on (moisture-equilibrated) coals as a function of pressure and

temperature. Therefore we use 7 degrees of freedom (n, PL,T=303K, PL,T=319K, PL,T=334K,

ρT=303K, ρT=319K, ρT=334K) to describe the CH4 and CO2 Isotherms at three different

temperatures. The quality of the fit by using equation 53 and the restrictions of equation

54, 55 is shown in the Appendix (Figure A1 and A2).

In-situ gas sorption capacity of coal seams

Gas sorption capacity of coal seams is primarily controlled by the in-situ moisture,

pressure and temperature conditions. The relationships derived for the temperature

dependence of the parameters of the sorption isotherms can be used to compute the

sorption capacity of the coal for individual gases as a function of depth.

If a relationship for the change of sorption capacity with coal rank is established and

incorporated into the calculation scheme, too, then the dynamic evolution of the

sorption characteristics of a coal seam throughout its burial history can be

reconstructed (Bustin and Bustin, 2008; Hildenbrand et al., 2006b; Juch, 2004; Kronimus

et al., 2008; Weniger et al., 2010b).

Static case (present-day sorption capacity of a coal seam of known

coal rank and depth)

A geothermal gradient of 0.03 K/m with a surface temperature of T0=288K (15°C), and a

hydrostatic pressure gradient of 0.01 MPa/m were chosen to represent temperature and

pressure conditions for a hypothetical CBM reservoir as a function of depth. Methane

sorption isotherms were measured at 303, 318, 333, and 348 K corresponding to

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

235 | P a g e

reservoir depths of 500, 1000, 1500, and 2000 m, respectively. Sorption capacities for

greater depths were extrapolated.

Dynamic case (evolution of sorption capacity with burial history)

The Langmuir parameters and the CH4 and CO2 sorption enthalpies determined for

moisture-equilibrated coals were taken from Gensterblum et al. 2013 (Table 26 and

Table 27). This reference also provides more details about the fitting procedure as well

as aspects regarding the evaluation and assumptions made.

The sorption capacity as a function of depth n(z) is also a function of the corresponding

temperature T(z) and the hydrostatic pressure profile p(z) at a specific location in the

basin. For simplicity a time-invariant linear geothermal gradient and a hydrostatic

pressure gradient were assumed in this calculation (Equation 58).

(56)

Where and are the geothermal and the hydrostatic pressure gradients, respectively

and T0 and P0 are surface temperature and pressure, respectively. Integration equation

56 into the Langmuir equation leads to Equation 58.

(57)

As argued above, the total Langmuir sorption capacity (n) for methane is considered

independent of temperature especially for the small temperature range used in this

study. Equation 60 has been simplified accordingly. Furthermore, since the experimental

temperature range is limited (303-349 K) it is important to be aware that extrapolation

to higher pressures and temperatures is associated with uncertainties.

(58)

The geothermal gradient was varied between 0.02 and 0.04 K/m; the surface

temperature T0 was chosen as 288K. The hydrostatic gradient was kept constant at 1

MPa per 100 m. Formation fluid density was kept constant at 1000 kg/m³. It must be

considered that at greater depths, water density decreases with increasing temperature

but usually increases with increasing salinity. These changes are, however considered to

be minor and have been disregarded in this study.

Equation 61 is an empirical formula after Barker and Pawlewicz (1994) describing the

correlation (coefficient r²=0.7 and sample n>600) between vitrinite reflectance VRr and

the maximum temperature Tmax that humic organic matter has been exposed to (Barker

and Pawlewicz, 1994; Schoenherr et al., 2007).

(59)

The kinetics of the hydrocarbon gas generation during coalification have been not

considered explicitly (Burnham, 1979a; Burnham, 1979b; Burnham and Sweeney, 1989;

Cramer et al., 1998; Krooss, 1988; Krooss et al., 1995a)

Consequently, following equation 58 and 61 the maximum temperature was 362 K, 403

K and 504 K for the sub-bituminous A, the high-volatile bituminous and the anthracite

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

236 | P a g e

coal, respectively. Assuming a geothermal gradient of 0.03 K/m, this leads to maximum

burial depths of ~2500 m, 3900 m and 7300 m, respectively. Clearly, a representative

temperature gradient for a specific basin is important and a sensitive parameter for any

saturation considerations.

In order to calculate maximum CH4 saturations as a function of basin uplift starting at

maximum burial depth we use the simplified conceptual model shown in Figure 65. The

first step in the algorithm is to compute the maximum coalification temperature from

present-day coal maturity (VRr) (Figure 65, step 1). Assuming a thermal gradient

(=(dT/dz) e.g. 0.03 K/m) the maximum burial depth can be calculated (Figure 65, step

2). Assuming a normal hydrostatic pressure gradient of 0.01 MPa/m the methane

sorption capacity can be calculated according to the pressure and temperature

conditions at maximum burial depth (Figure 65, step 3). At maximum burial depth, the

generation rate of thermogenic methane reaches or has reached its maximum. It is

generally assumed that the total amount of gas generated is 2-4 times the maximum

sorption capacity (Hunt and Steele, 1991; Jüntgen and Klein, 1974; Tang et al., 1996).

Following their results CH4 saturation of the coal can be expected to be 100% at

maximum depth (Figure 65, step 4). The excess of generated gas will lead to an

overpressure and or will dismigrate into neighbouring formations. Finally the uplift

starts to the present-day depth (Figure 65, step 5). Gas saturation during uplift is

calculated as the ratio of the sorption capacity at a given present-day depth (with

corresponding temperature and fluid pressure) and the methane sorption capacity at

the pressure and temperature conditions at maximum burial depth (2500 m, 3900 m

and 7300 m respectively) for the low, medium and high rank coal considered here. In

reality the maximum burial depth for the anthracite is considered lower due to the

higher temperature gradient (heat flow) at the time of maximum burial, i.e. roughly

5000 m (Senglaub et al., 2005).

Gas recovery potential of coal seams

Equation 52 can be used to calculate the recovery potential as a function of burial depth.

By subtracting the residual gas amount at abandonment pressure pA from the CH4

sorption capacity at critical desorption pressure pD, the estimated amount of gas

recovered can be obtained (see Figure 62). The abandonment pressure pA is commonly

controlled by the installed gas peripheral equipment of the CBM field.

(60)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

237 | P a g e

Figure 65: Conceptual design of the evolution of the in-situ gas saturation of coal seams as a function of maturity and thermal gradient. Gas

saturation during uplift is calculated as the ratio of the sorption capacity at a given present-day depth (with corresponding temperature and

fluid pressure) and the methane sorption capacity at the pressure and temperature conditions at maximum burial depth. The first step in the

algorithm is to compute the maximum coalification temperature from present-day coal maturity (VRr) (step 1). Assuming a thermal gradient

(=(dT/dz) e.g. 0.03 K/m) the maximum burial depth can be calculated (step 2). Assuming a normal hydrostatic pressure gradient of 0.01

MPa/m the methane sorption capacity can be calculated according to the pressure and temperature conditions at maximum burial depth

(step 3). At maximum burial depth, the generation rate of thermogenic methane reaches or has reached its maximum. It is generally assumed

that the total amount of gas generated is 2-4 times the maximum sorption capacity (Hunt and Steele, 1991; Jüntgen and Klein, 1974; Tang et

al., 1996). Following their results CH4 saturation of the coal can be expected to be 100% at maximum depth (step 4). Finally the uplift starts

to the present-day depth (step 5).

Tmax(VRr)

• Coalification

• Barker & Pawlewicz 1994

dmax(Tmax,dT/dz)

• Burial maximum depth

nmax(p,Tmax)

• maximum sorbedamount of methane

SCH4(z=dmax)

•Saturation at maximum depth equal to 100%

SCH4 (z)

•Saturation development during uplift

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

238 | P a g e

Results and geological implications

The isotherms of all samples show a clear increase in CH4 Langmuir sorption capacity n

as a function of rank (Table 26 for CH4). Table 26shows the fitted Langmuir parameters

for CH4 sorption on three (moisture-equilibrated) coals at three different temperatures

303, 319 and 334K, respectively. We observe that the Langmuir pressure KL decreases

with an increase in coal rank.

The listed Langmuir pressure PL, total CH4 sorption capacity n and the adsorbed phase

density ρads values are the parameters used in equation 55. The n values denote the

average deviation between measured data points and the best fit function. We were not

able to achieve acceptable fitting results with a constant adsorbed phase density ρads.

Therefore the experimental data were fitted by adjusting PL and ρads for each of the three

isotherms and taking n as a common temperature-independent parameter for each

coal. The methane 303K isotherm for the anthracite did not reach an equilibrium

(plateau) value and therefore the sorbed phase density resulting from the fitting

procedure is high.

Table 27 lists the CO2 Langmuir parameters used to calculate the temperature-

dependence of the CO2 Langmuir pressure for the three coals. The isosteric heats of

adsorption for CO2 listed in Table 27 are higher than the respective CH4 values and in the

same range as values reported in previous studies (17-29 kJ/mol) (Nodzenski, 1998;

Ozdemir and Schroeder, 2009).

Table 6 lists the isosteric heats of CH4 sorption for the three different coal samples.

These are in good agreement with the reported values of previous publications (18-20

kJ/mol) (Nodzenski, 1998; Yang and Saunders, 1985).

Sorption enthalpies for CH4 and CO2 are listed in Table 26and Table 27, respectively.

They were calculated from the fitted Langmuir pressures KL for different temperature

with a common Langmuir volume n following equation 55.

Figure 66 and Figure 67 show the calculated CH4 and CO2 sorption capacity as a function

of burial depth for the specified geothermal and hydrostatic pressure gradients. For all

samples these curves exhibit a maximum followed by a slight decrease with increasing

depth (Figure 66). This is because the effect of temperature (decrease in sorption

capacity with increasing temperature) exceeds the pressure effect (increasing sorption

capacity with increasing pressure). This maximum is more pronounced for high

geothermal gradients (Figure 66). Further the difference in CH4 and CO2 sorption

capacities between maximum and values at 4 km depth increases with increasing

geothermal gradient, but also with increasing coal maturity, from sub-bituminous A to

anthracite. The difference in sorption capacity between the different coal rank decreases

with increasing depth. In general these trends are much more sensitive to the

geothermal gradient than to the hydrostatic pressure gradient.

The maximum of the CH4 sorption capacity depth trend is located at shallower depth

than the maximum of the CO2 trend (Figure 67A). It is more pronounced and shifts to

shallower depth as the thermal gradient increases. The maximum in the sorption

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

239 | P a g e

capacity trend occurs at greatest depth for the high-volatile bituminous coal (Figure 66).

The maxima for the sub-bituminous A and anthracite coals occur at similar depth levels.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

240 | P a g e

Figure 66: CH4 (black) and CO2 (grey) sorption capacities as function of depth and thermal

gradient for the three coal samples investigated here. The sorption capacity as a

0

500

1000

1500

2000

2500

3000

3500

4000

0 0.5 1 1.5 2

de

pth

(m)

estimated CO2,CH4 sorption capacity (mmol/g)

sub-bituminous A0.02 K/m

0.03 K/m

0.04 K/m

CH4CO2

A)

0

500

1000

1500

2000

2500

3000

3500

4000

0 0.5 1 1.5 2

dep

th (m

)

estimated CH4,CO2 sorption capacity (mmol/g)

high-volatile bituminous

0.02 K/m

0.03 K/m

0.04 K/m

CH4 CO2

B)

0

500

1000

1500

2000

2500

3000

3500

4000

0 0.5 1 1.5 2

dep

th (m

)

estimated CH4,CO2 sorption capacity (mmol/g)

Anthracite

0.02 K/m

0.03 K/m

0.04 K/m

CH4 CO2

C)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

241 | P a g e

function of depth is plotted to the maximum burial depth according to Barker and

Pawlewicz (1994). (1mmol/g =24.5 Std. m³/ton).

Figure 67: A) CH4 (black) and CO2 (grey) sorption capacities as function of depth assuming

a thermal gradient of 0.03K/m for sub-bituminous A, high-volatile bituminous and

anthracite coal deposits. B) Illustrates the CH4 sorption capacity in proportion to the

compressed gas in 5% matrix porosity. The porosity of 5% is assumed to be constant

0

1000

2000

3000

4000

0 0.5 1 1.5 2

de

pth

(m

)

calculated CH4, CO2 sorption capacity (mmol/g)

sub-bituminous A

high-volatile bituminous

anthracite

A)

0

1000

2000

3000

4000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

de

pth

(m)

CH4 sorption capacity / compressed gas ratio

sub-bituminous A

high-volatile bituminous

anthracite

B)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

242 | P a g e

with depth (no poro-elasticity or compaction is considered) (1mmol/g =24.5 Std.

m³/ton).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

243 | P a g e

This can be explained by decreasing sorption enthalpies with increasing rank as

summarised in Table 28 and Table 29.

Table 27 shows the fitted CO2 Langmuir parameters of the three coal samples at the

three temperatures, 319, 334 and 349 K respectively. A dependence of the Langmuir

volume n on coal rank cannot be observed. The anthracite has the highest CO2 sorption

capacity, followed by the sub-bituminous A and the high-volatile bituminous coal.

Figure 67B illustrates the proportion of the sorption capacity to free (compressed) gas

within the porosity as a function of depth. It is interesting to note that the ratio between

sorbed and compressed gas decreases with increasing depth. This indicates that with

increasing depth the porosity determination becomes more important for the gas-in-

place assessment.

Figure 68: CH4 recovery potential as a function of depth. Sorption enthalpies are used from

Table 28 and Table 29 assuming a constant abandonment pressure of 1 MPa, normal

hydrostatic pressure gradient and geothermal gradient of 0.03K/m.

Using equation 62, the potential recovery capacity for CH4 and CO2 has been calculated

as a function of depth (Figure 68). These calculations show a stronger decrease in CO2

sorption capacity with depth for moist low rank coals at approximately 1000 m than for

high rank coal. However, the maximum recovery potential decreases only slightly for

moist high rank coals. This implies that CO2-ECBM will become less effective at depths of

more than ~1000 m.

0

1000

2000

3000

4000

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

dep

th (m

)

max. CO2 storage- and CH4 recovery capacity (mmol/g)

methane (sub-bituminous A)

methane (high-volatile bituminous)

methane (anthracite)

carbon dioxide (sub-bituminous A)

carbon dioxide (anthracite)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

244 | P a g e

Table 26: Fitted Langmuir parameters for methane excess sorption isotherms on sub-

bituminous A, high-volatile bituminous and an anthracite at three different

temperatures (see text).

Sub-bituminous A PL

(MPa)

n

(mol/kg)

ρads

(kg/m³)

n

(mol/kg)

303K 7.10 0.774 600 0.004

319K 10.47 0.774 373 0.006

334K 13.57 0.774 373 0.003

High-volatile

bituminous

303K 4.98 0.88 406 0.008

319K 6.78 0.88 885 0.003

334K 9.26 0.88 800 0.004

Anthracite

303K 3.34 1.88 946566* 0.015

319K 4.34 1.85 505 0.014

334K 6.63 1.85 464 0.015

Table 27: Fitted Langmuir parameters for CO2 excess sorption isotherms on sub-

bituminous A, high- volatile bituminous and anthracite at three different

temperatures.

Sub-bituminous A PL

(MPa)

n

(mol/kg

)

ρads

(kg/m³)

n

(mol/k

g)

318K 3.84 1.65 885 0.017

334K 7.31 1.65 1333 0.022

349K 10.60 1.65 1611 0.013

High-volatile

bituminous

318K 3.03 1.63 1189 0.018

334K 5.59 1.66 1124 0.017

349K 8.0 1.67 1247 0.013

Anthracite

318K 1.39 1.94 1270 0.020

334K 2.09 1.94 1167 0.023

349K 3.21 1.94 1466 0.004

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

245 | P a g e

Table 28: CH4 sorption enthalpy for moist coals of different rank based on Langmuir

pressure (see Appendix, Figure A1).

CH4 dH

(kJ/mol)

ln (PL)

Sub-bituminous A -17.9 ± 1.6 9.09 ± 0.6

High-volatile bituminous -16.7 ± 0.5 8.26 ± 0.2

Anthracite -18.9 ± 3.1 8.66 ± 1.2

Table 29: Isosteric heats of CO2 sorption on moisturized coals based on Langmuir pressure

(see Appendix, Figure A2).

dH

(kJ/mol) ln PL

Sub-bituminous A coal -29.2 ± 2.8 12.59 ± 1.0

High-volatile bituminous -27.7 ± 3.0 11.54 ± 1.1

Anthracite -25.6 ± 1.1 10.02 ± 0.4

Figure 69: CO2 to CH4 sorption capacity ratio as a function of depth for the three coal

samples studied, assuming a thermal gradient of 0.03K/m and a hydrostatic pressure

gradient of 0.01MPa/m.

0

1000

2000

3000

4000

0 1 2 3 4 5

dep

th (m

)

CO2/CH4 sorption capacity ratio

sub-bituminous A

high-volatile bituminous

anthracite

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

246 | P a g e

All coal samples show lower sorption capacities for CH4 than for CO2 and the CO2 to CH4

sorption capacity ratio varies from 5 at shallow depths to <3.5 at great depth for the sub-

bituminous A and high-volatile bituminous coal respectively (Figure 69). The sorption

capacity of the anthracite coal shows a different depth trend. The change in relative

sorption capacity with depth is smaller and the ratio increases with depth. The

minimum of the sorption capacity ratio at 1300 m depth roughly, has a ratio of 1.35 and

increases up to ratio of 1.75. Minimal to no differences are observed with different

thermal gradients for the high-volatile bituminous, the sub-bituminous A and anthracite

coal. Only the anthracite shows an increasing ratio with increasing depth (Figure 69).

This trend is more pronounced at higher thermal gradients.

The sorption enthalpies for methane and CO2 obtained from the fitted Langmuir

pressures (Table 27) are in line with the published data for Silesian coals (-20 kJ/mol for

methane; -27.0 and -29.4 kJ/mol for CO2; (Taraba, 2011)). As observed for the two

Silesian coal samples studied by Taraba (2011) the CO2 sorption enthalpies measured in

this study (Table 29) decrease with increasing rank.

Considering oversaturation or overpressured reservoirs it should be mentioned that on

the geological time scales needed for this major uplift, diffusion and or pressure

diffusion are relevant transport mechanisms. For the following considerations, we

assume that at the beginning of uplift, i.e. at maximum burial depth, the coal seam is fully

gas saturated and equilibrated (cf. Figure 65). This gas content is used as the reference

value to calculate the relative gas saturation during uplift. The results of these

computations are shown in Figure 70 for three different coal ranks. Model A and B

illustrate the change in saturation with increasing geothermal gradient (Figure 70). The

minimal CH4 saturation decreases with increasing thermal gradient. Furthermore, this

change in saturation is highest for the high-volatile bituminous coal. The calculated

maximum burial depth of 7300 m by Barker et al. 1986 for the anthracite is not in line

with a maximum burial depth of 4500 to 5000 m from a basin modelling study by

Senglaub et al. (2006). Additionally, it should be kept in mind that an assumed maximum

burial depth of 7300 m requires an extensive extrapolation of the measured sorption

data to high pressures and temperatures.

Discussion

Adsorption is an exothermic process and adsorption constants are equilibrium

constants that obey van't Hoff's equation. Due to the exothermic nature of the process,

PL increases with increasing temperature. This means that higher pressures are

necessary to reach the same degree of coverage (sorption capacity n(p)) at higher

temperatures. Based on the assumption that a temperature increase (308K up to 349K <

Tmax) will not change the surface, pore or chemical structure of the coal, we assume that

the sorption capacity n should be independent of temperature, i.e. the total amount of

sorption sites should remain constant with temperature. However, due to the presence

of moisture, a shift in the portion of sorption sites occupied by water can be assumed.

The saturated moisture content, or sorption capacity of water, will decrease only slightly

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

247 | P a g e

with increasing temperature, because the sorption enthalpy of water is much higher

than the sorption enthalpies of carbon dioxide or methane (Busch and Gensterblum,

2011). As simplification, we assume that the total amount of sorption sites remains

approximately constant with temperature; however, the proportion of sorption sites for

CH4 and CO2 at a given temperature and pressure will decrease slightly (the different

differential enthalpy for CH4 and CO2).

Due to the opposing effects of increasing pressure and temperature with depth, the

value of the sorption enthalpy will determine to what extent the increase in surface

coverage with increasing hydrostatic pressure will be compensated by the reduction of

the surface coverage due to increasing temperature along the geothermal gradient.

Generally, sorption capacity is reduced to a greater extent with increasing depth at

higher temperature gradients. This is in line with previous observations (Bustin and

Bustin, 2008; Hildenbrand et al., 2006b). For this computation of sorption capacity

nsorbed [mol kg-1] with depth (under given pressure and temperature gradients),

Hildenbrand et al (2006) used methane sorption isotherms measured for dry coals,

while Bustin and Bustin (2008) used methane sorption isotherms on moist coals with

vitrinite reflectance values of 0.4, 0.5 and 1.7%.

While gas sorption capacities of moisture-equilibrated coals are generally lower than for

dry coals, the decrease in sorption capacity with depth calculated in the present study is

smaller than the trends obtained by Hildenbrand et al. (2006). This is essentially due to

the use of isotherms for moisture-equilibrated coals in this work as opposed to data for

dry coals used by Hildenbrand et al. (2006).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

248 | P a g e

Figure 70: Calculated CH4 saturation trends after uplift from the maximum burial depth

(burial history) following the concept illustrated in figure 5. Model A: assuming a

lower geothermal gradient of 2K/100m and the corresponding depth of 3700 m for the

sub-bituminous, 5800 m for the high-volatile bituminous and 10900 m for the

anthracite (Barker and Pawlewicz 1994). Model B: assuming a geothermal gradient of

3K/100m and the corresponding depth 2500 m, 3900 m and 7300 m (Barker and

Pawlewicz 1994) for the three coal samples assuming no loss of CH4 and a hydrostatic

pressure gradient of 1MPa/100m.

0

500

1000

1500

2000

2500

3000

3500

4000

50% 75% 100% 125% 150%d

epth

(m)

CH4 saturation

thermal gradient 2K/100m

sub-bituminous A

high-volatile bituminous

anthracite

Model A

98%93%

undersaturation oversaturation

79%

0

500

1000

1500

2000

2500

3000

3500

4000

50% 75% 100% 125% 150%

de

pth

(m

)

CH4 Saturation

sub-bituminous A

high-volatile bituminous

anthracite

90% 98%

Model B

thermal gradient 3K/100m

61%

undersaturation oversaturation

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

249 | P a g e

Influence of water on gas sorption capacity The sorption of methane or carbon dioxide in moist coal samples is largely controlled by

competition with water molecules. In general, the interaction forces between coal and

water (and consequently, the differential sorption enthalpy) are much stronger than for

methane or carbon dioxide, especially for low water surface coverage. It is reasonable to

assume that the amount of sorption sites occupied by water molecules will change with

temperature. For moisturised coals several publications report a change in the Langmuir

amount of sorbed substance with temperature (Crosdale et al., 2008; Moore and

Crosdale, 2006; Ozdemir, 2009).

Figure 70shows the calculated gas saturations relative to the saturation at maximum

burial depth for three uplifted coal seams. Based on the trends illustrated in Figure 66,

we can expect a high saturation level during uplift for the sub-bituminous and high-

volatile bituminous coals. The highest saturation for any thermal gradient is expected

for the high-volatile bituminous followed by sub-bituminous coal. The calculated

maximum burial depth of 7300 m by Barker et al. (1986) for the Ibbenbüren anthracite

is not in line with the basin modelling study by Senglaub et al. (2006). Lower CH4

saturation is expected if uplift started from a more reasonable lower maximum burial

depth of only 4000 m to 5000 m (Senglaub et al., 2006), due to the stronger decrease in

sorption capacity at identical temperatures (T=504K) but lower pressures (p=40-

50MPa). Gas contents in the German anthracite mine in Ibbenbüren and gas outbursts

that occurred in the past indicate that current CH4 saturation levels in the anthracite

seams are higher (Figure 70B). Therefore it is believed that the saturation data in Figure

70for the anthracite needs further investigations. Lower maximum coalification

temperature or a lower thermal gradient would be a possible explanation (Figure 70A).

It is likely that coal deposits are generally undersaturated with respect to CO2 because of

the high CO2 solubility in formation brine (Figure 61). The results of this study suggest

that, due to the rapid decrease in CO2 sorption capacity with depth, carbon capture and

storage (CCS) in coal becomes unattractive at high geothermal gradient and less

favourable for depths greater than ~2 km. As illustrated in Figure 63 and Figure 64, an

increase in Langmuir pressure will increase the productivity, because the higher

amounts of gas desorbing at the same pressure drop (pD-pA). In Table 26 and Table 27,

we show that Langmuir pressure decreases with increasing coal rank. Therefore, we can

conclude that lower rank coal deposits degas more readily than high rank coal deposits

(Figure 67). Furthermore, Figure 67 illustrates that CO2-ECBM is more efficient at

depths shallower than 1000 m, regardless of the sorption enthalpies. For depths greater

than 1000 m, CO2-ECBM becomes less efficient for low rank coals. It should be

mentioned, that transport processes or effects resulting from coal swelling or changes in

permeability - stress relationships are not considered in these conclusions.

Saturation values determined from dry or as-received coal samples suggest much lower

saturations after uplift and the sorption capacity will show a more pronounced decrease

with increasing depth and increasing thermal gradient. Due to lower Langmuir

pressures pL and higher sorption enthalpies for the CH4 sorption on dry or as-received

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

250 | P a g e

coals, the maximum recovery potential will be also different. We assume that

undersaturated coal seams are closer to reality.

Conclusions

Among the three possible gas storage mechanisms in coals (free compressed gas in the

open porosity, dissolved gas in the formation water and sorbed gas), sorption is the

dominant mechanisms for CBM reservoirs (Figure 61). For example, given a porosity of

5%, the amount of compressed methane is 2.5 to 10 times lower than the sorption

capacity for methane and carbon dioxide. The largest difference is observed for

anthracite coal. The ratio of compressed over sorbed gas shows a decreasing trend with

increasing depth (Figure 66B). Furthermore the lowest contribution to the total storage

capacity is attributed to methane solubility in water or brine. Carbon dioxide solubility

in water is about 10 times higher than for methane (Figure 61). Nonetheless, the mass of

dissolved carbon dioxide is still roughly 100 times lower than the carbon dioxide

sorption capacity of moisture-equilibrated coal. Therefore the amount of gas dissolved

in water or brine is negligible for methane and in most instances also for CO2 in coal.

The main parameters controlling ultimate gas recovery from coal deposits are the

natural cleat permeability and the total gas content. Sorption capacity contributes the

most to the storage potential of gas in coal seams. It depends largely on the type and

rank of the coal and the subsurface conditions. Sorption capacity is influenced by

pressure and temperature, moisture content, composition of the organic material,

mineral content, and coal rank. Sorption capacity and sorption enthalpy are strongly

influenced by coal properties, such as macerals composition, total carbon content, coal

rank, pore structure, pore size distribution and mineral matter (Busch and Gensterblum

2011). Sorption enthalpy controls the temperature-dependence of the sorption

equilibrium constant and is therefore an important factor when considering storage

capacity and gas saturation based on the burial and uplift history of a CBM deposit.

Measurements performed on dry coals and at different water contents reveal that even

very low water contents lead to a strong reduction of CH4 sorption capacity.

The evaluation of the isotherms measured in this study reveals that for coals of sub-

bituminous A rank, common in many producing CBM reservoirs, and pressure and

temperature gradients typical for most sedimentary basins, the sorption capacity of

coals decreases only slightly during uplift.

The geothermal gradient is much more important than the hydrostatic pressure

gradient for the CH4 sorption capacity prediction as a function of depth. For CBM as well

as CO2-ECBM activities a low geothermal gradient is favourable.

Considering only the sorption measurements and recovery calculations, the feasibility of

CO2 storage and CH4 recovery for CO2-ECBM will increase with increasing rank and will

decrease with depth. The feasibility of CO2-ECBM aiming at carbon storage depends on

the thermal gradient of the deposit and will decrease with increasing depth for thermal

gradients lower than 0.03 K/m.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

251 | P a g e

CO2-ECBM is likely to become less effective with increasing depth because the CO2 to CH4

ratio decreases with depth, hence CO2-ECBM is expected to be less effective with

increasing maturity due to the observation that the highest selectivity is observed for

low mature coals.

We conclude from this study that low rank coal deposits require a lower pressure draw-

down than high rank coal deposits to desorb an equivalent amount of methane. In high

rank coal seams it is more likely to observe under-saturation than in low rank deposits.

Based on the results of this study, the frequently observed under-saturation of coal

seams in the field (with thermal gradients not higher than 0.03K/m) is more likely

related to loss of methane from the coal seam (dismigration) than changes in the

sorption capacity, especially for the maturity level of sub-bituminous A coals. Thermal

gradients higher than 3K per 100m or high heat flow events in burial history could also

cause under-saturation. Discovering an over-saturated coal seam is not likely; but if so, it

can only be explained by the decrease in sorption capacity at shallow depths (below

1000 m) or additional gas generation, e.g. by microbial activities, combined with

retention due to efficient sealing.

These findings indicate that reliable information on the in-situ moisture content of coal

seams is crucial for the estimation of reservoir capacities and resource potential. One

shortcoming is the still limited amount of published sorption isotherms on moisture-

equilibrated coals.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

252 | P a g e

Appendix

Figure A1 and A2 illustrate the quality of the fitting (according to equation 55 and 56).

Furthermore the figures show that it is acceptable to extrapolate sorption capacity

values down to temperature of a depth of 4km.

Figure A1: The CH4 Langmuir pressure for moisture-equilibrated (at 97% relative

humidity) coals of different rank as a function of the inverse temperature to calculate

the heat of adsorption (adsorption enthalpy) (Gensterblum, 2013; Myers, 2004).

Figure A2: The CO2 Langmuir pressure for moisture-equilibrated coals of different rank as

a function of the inverse temperature to calculate the heat of adsorption (adsorption

enthalpy) (Gensterblum, 2013; Myers, 2004).

1.00

1.20

1.40

1.60

1.80

2.00

2.20

2.40

2.60

2.80

0.0029 0.003 0.0031 0.0032 0.0033 0.0034

ln(P

L(C

H4))

Temperature-1 (K-1)

HvB

Sb

Anth.

0.00

0.50

1.00

1.50

2.00

2.50

3.00

0.0028 0.0029 0.003 0.0031 0.0032

ln(K

L,C

O2)

Temperature-1 (K-1)

SB

hvb

Anth.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

253 | P a g e

Table 30: List of symbols

Symbols Units Physical GIP [mol] Gas in place S - Saturation nsorbed [mol kg-1] In-situ sorbed gas (amount of substance per kg coal) nporosity [mol kg-1] Compressed gas in the pore volume (porosity) (amount of substance

per kg coal) ntotal uptake [mol kg-1] Total sorption capacity nformation water [mol kg-1] Gas dissolved in the formation water nlost gas [mol kg-1] Gas lost during coal sample recovery Ndesorbed [mol kg-1] Amount of desorbed gas from the canister desorption nresidual [mol kg-1] Residual gas after canister desorption npredicted [mol kg-1] Total estimated desorbable gas content (theoretical value) nabsolute [mol kg-1] Total uptake capacity of a coal sample

dHK [kJ mol-1] Sorption differential enthalpy of the dynamic equilibrium for ad- and desorption

dHn [kJ mol-1 ] Sorption differential enthalpy of the sorption sites dH1 or dH2 [kJ mol-1 ] Sorption differential enthalpy of the sorption sites of samples 1 and 2 Vseam [m3] Net volume of the coal seam ρbulk coal [kg m-3] Bulk density of the coal ρadsorbed [kg m-3] Density of the adsorbed phase ρfree [kg m-3] Gas density calculated by GERG

n [mol kg-1] Total specific sorption capacity

n(z) [mol kg-1] Specific sorption capacity as a function of depth n(p) [mol kg-1] Amount of substance adsorbed at pressure p p [MPa] Gas pressure PL [MPa] Langmuir pressure P0 [MPa] Pre-exponential factor for the temperature dependence of the

Langmuir pressure PR [MPa] Initial reservoir pressure

PD [MPa] Initial desorption pressure (also called critical desorption pressure), below this pressure the CH4 starts to desorb

PA [MPa] Abandonment pressure of CBM reservoir p0 [MPa] Atmospheric pressure p(z) [MPa] Formation water pressure as a function of depth z [m] Depth

[MPa m-1] Pressure gradient

σmax [MPa] Maximum stress of burial history T [K] Temperature Tmax [K] Maximum temperature of burial history T0 [K] Surface temperature T(z) [K] Temperature as a function of depth

[K m-1] Geothermal gradient and equal to the differential expression dT/dz

R [J mol-1 K-1] Gas constant R = 8.31451 VRr [%] Vitrinite reflectance (mean random) Rmax [%] Maximum vitrinite reflectance K(…) Vitrinite maturation function

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

254 | P a g e

References

Alcaniz-Mongue, J., Linares-Solano, A., Rand, B., 2001. Mechanism of Adsorption of Water in Carbon Micropores As Revealed by a Study or Activated Carbon Fibers. J. Phys. Chem., 106(No. 12): 3209 - 3216.

Allardice, D.J. et al., 2003. The characterisation of different forms of water in low rank coals and some hydrothermally dried products[small star, filled]. Fuel, 82(6): 661-667.

Allardice, D.J. and Evans, D.G., 1971. The-brown coal/water system: Part 2. Water sorption isotherms on bed-moist Yallourn brown coal. Fuel, 50(3): 236-253.

Amarasekera, G., Scarlett, M.J. and Mainwaring, D.E., 1995. Micropore size distributions and specific interactions in coals. Fuel, 74(1): 115-118.

Antia, D.D.J., 1986. Kinetic method for modeling vitrinite reflectance. Geology, 14(7): 606-608.

ASTM, D1412-93. Standard Test Method of eqilibrium moisture of Coal at 96 to 97 percent relative Humidity and 30°C.

Avgul, N.N. and Kiselev, A.V., 1970. Chemistry and Physics of Carbon, 6, New York. Ayers, W.B., 2002. Coalbed gas systems, resources, and production and a review of

contrasting cases from the San Juan and powder river basins. AAPG Bulletin, 86(11): 1853-1890.

Bae, J.S. and Bhatia, S.K., 2006. High-Pressure Adsorption of Methane and Carbon Dioxide on Coal. Energy and Fuels, 20(6): 2599-2607.

Barker, C.E. and Goldstein, R.H., 1990. Fluid-inclusion technique for determining maximum temperature in calcite and its comparison to the vitrinite reflectance geothermometer. Geology, 18(10): 1003-1006.

Barker, C.E. and Pawlewicz, M.J., 1986. Lecture Notes in Earth Sciences. The correlation of vitrinite reflectance with maximum temperature in humic organic matter., 5. Springer-Verlag, Berlin Heidelberg, 15 pp.

Barker, C.E. and Pawlewicz, M.J., 1994. Calculation of vitrinite reflectance from thermal histories and peak temperatures. A comparison of methods. In: P.K. Mukhopadhyay and W.G. Dow (Editors), ACS Symposium series 570. Vitrinite reflectance as a maturity parameter: applications and limitations, pp. 216–229.

Battistutta, E., van Hemert, P., Lutynski, M., Bruining, H. and Wolf, K.-H., 2010. Swelling and sorption experiments on methane, nitrogen and carbon dioxide on dry Selar Cornish coal. International Journal of Coal Geology, In Press, Accepted Manuscript.

Beamish, B.B.C., P.J., 1998. Instantaneous outbursts in underground coal mines: An overview and association with coal type. Int J. Coal Geology, 35(27-55).

Behar, F. and Vandenbroucke, M., 1987. Chemical modelling of kerogens. Organic Geochemistry, 11(1): 15-24.

Bhatia, S.K., 1987. Modelling the pore structure of coal. AICHE Journal, 33(10): 1707-1718.

Billemont, P., Coasne, B. and De Weireld, G., 2011. An experimental and molecular simulation study of the adsorption of carbon dioxide and methane in nanoporous carbons in the presence of water. Langmuir, 27(3): 1015-1024.

Billemont, P., Coasne, B. and De Weireld, G., 2013. Adsorption of carbon dioxide, methane, and their mixtures in porous carbons: Effect of surface chemistry, water content, and pore disorder. Langmuir, 29(10): 3328-3338.

Birol, F., 2012. World Energy Outlook 2012, International Energy Agency (IEA), Paris.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

255 | P a g e

Bostick, N.H., 1983. Vitrinite reflectance and temperature-gradient models applied at a site in piceance basin, Colorado AAPG Bulletin-American Association of Petroleum Geologists, 67(3): 427-428.

Boudou, J.-P. et al., 2008. Organic nitrogen chemistry during low-grade metamorphism. Geochimica et Cosmochimica Acta, 72(4): 1199-1221.

Bratek, K., Bratek, W., Gerus-Piasecka, I., Jasienko, S., Wilk, P., 2002. Properties and structure of different rank anthracites. Fuel, 81: 97-108.

Brennan, J.K., Bandosz, T.J., Thomson, K.T. and Gubbins, K.E., 2001. Water in porous carbons. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 187-188: 539-568.

Brunauer, S., Emmett, P.H., Teller, E, 1938. Adsorption of gases in multimolecular layers. Journal of the American Chemical Society 60: 309-319.

Burnham, A.K., 1979a. Reaction kinetics between CO2 and oil-shale residual carbon. 1. Effect of heating rate on reactivity. Fuel, 58(4): 285-292.

Burnham, A.K., 1979b. Reaction kinetics between CO2 and oil-shale residual carbon. 2. Partial-pressure and catalytic-mineral effects. Fuel, 58(10): 713-718.

Burnham, A.K. and Sweeney, J.J., 1989. A chemical kinetic model of vitrinite maturation and reflectance. Geochimica et Cosmochimica Acta, 53(10): 2649-2657.

Busch, A. et al., 2008. Carbon dioxide storage potential of shales. International Journal of Greenhouse Gas Control, 2(3): 297-308.

Busch, A. and Gensterblum, Y., 2011. CBM and CO2-ECBM related sorption processes in coal: A review. International Journal of Coal Geology, 87: 49-71.

Busch, A., Gensterblum, Y. and B.M., K., 2003a. Methane and CO2 sorption and desorption measurements on dry Argonne Premium Coals: Pure components and mixtures. International Journal of Coal Geology, 55: 205-224.

Busch, A., Gensterblum, Y. and Krooss, B.M., 2003b. Methane and CO2 sorption and desorption measurements on dry Argonne Premium Coals: Pure components and mixtures. International Journal of Coal Geology, 55: 205-224.

Busch, A., Gensterblum, Y. and Krooss, B.M., 2007. High-pressure sorption of nitrogen, carbon dioxide, and their mixtures on Argonne Premium Coals. Energy and Fuels, 21(3): 1640-1645.

Busch, A., Gensterblum, Y., Krooss, B.M. and Littke, R., 2004. Methane and carbon dioxide adsorption-diffusion experiments on coal: upscaling and modeling. International Journal of Coal Geology, 60(2-4): 151-168.

Busch, A., Gensterblum, Y., Krooss, B.M. and Siemons, N., 2006. Investigation of high-pressure selective adsorption/desorption behaviour of CO2 and CH4 on coals: An experimental study. International Journal of Coal Geology, 66(1-2): 53-68.

Busch, A. et al., 2003c. Investigation of preferential sorption behaviour of CO2 and CH4 on coals by high-pressure adsorption/desorption experiments with gas mixtures, Proceedings of the 2003 Coalbed Methane Symposium, Tuscalosa, pp. 14.

Busch, A., Gensterblum, Y., Krooss, B.M., 2007. High-Pressure Sorption of Nitrogen, Carbon Dioxide, and their Mixtures on Argonne Premium Coals. Energy and Fuels, 21: 1640-1645.

Busch, A., Krooss, B.M., Gensterblum, Y., van Bergen, F. and Pagnier, H.J.M., 2003d. High-pressure adsorption of methane, carbon dioxide and their mixtures on coals with a special focus on the preferential sorption behaviour. Presented at Geofluids IV, May 12-16, Utrecht., J. Geochem. Expl., 78-79: 671-674.

Bustin, A.M.M. and Bustin, R.M., 2008. Coal reservoir saturation: Impact of temperature and pressure. AAPG Bulletin, 92(1): 77-86.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

256 | P a g e

Bustin, R.M., 2004. Acid Gas Sorption by British Columbia Coals: Implications for Permanent Disposal of Acid Gas in Deep Coal Seams and Possible Co-Production of Methane, University of British Columbia, Vancouver.

Bustin, R.M. and Clarkson, C.R., 1998. Geological Controls on coalbed methane reservoir capacity and gas content. Int J. Coal Geology, 38: 3-26.

Carroll, R.E., Pashin, J.C., 2003. Relationship of sorption capacity to coal quality: CO2 sequestration potential of coalbed methane reservoirs in the Black Warrior Basin, International Coalbed Methane Symposium, Tuscaloosa, Al.

Carslaw, H.S. and Jaeger, J.C., 1959. Conduction of heat in solids Oxford University Press, Oxford

Cathles, L., III, Brown, L., Taam, M. and Hunter, A., 2011. A commentary on “The greenhouse-gas footprint of natural gas in shale formations― by R.W. Howarth, R. Santoro, and Anthony Ingraffea. Climatic Change, 113(2): 525-535.

Chaback, J.J., Morgan, W.D., Yee, D., 1996. Sorption of nitrogen, methane, carbon dioxide and their mixtures on bituminous coals at in-situ conditions. Fluid Phase Equilibria, 117: 289-296.

Chalmers, G.R.L. and Marc Bustin, R., 2007. On the effects of petrographic composition on coalbed methane sorption. International Journal of Coal Geology, 69(4): 288-304.

Charrière, D. and Behra, P., 2010. Water sorption on coals. Journal of Colloid and Interface Science, 344(2): 460-467.

Charrière, D., Pokryszka, Z. and Behra, P., 2010. Effect of pressure and temperature on diffusion of CO2 and CH4 into coal from the Lorraine basin (France). International Journal of Coal Geology, 81(4): 373–380.

Ciembroniewicz, A. and Marecka, A., 1993. Kinetics of CO2 sorption for two Polish hard coals. Fuel, 72(3): 405-408.

Clarkson, C.R. and Bustin, R.M., 1999a. The effect of pore structure and gas pressure upon the transport properties of coal: a laboratory and modeling study. 1. Isotherms and pore volume distributions. Fuel, 78(11): 1333-1344.

Clarkson, C.R. and Bustin, R.M., 1999b. The effect of pore structure and gas pressure upon the transport properties of coal: a laboratory and modeling study. 2. Adsorption rate modeling. Fuel, 78(11): 1345-1362.

Clarkson, C.R. and Bustin, R.M., 2000. Binary gas adsorption/desorption isotherms: effect of moisture and coal composition upon carbon dioxide selectivity over methane. International Journal of Coal Geology, 42(4): 241-271.

Clarkson, C.R., Bustin, R.M., 2000. Binary gas adsorption/desorption isotherms: effect of moisture and coal composition upon carbon dioxide selectivity over methane. Int J. Coal Geology, 42: 241-271.

Clayton, J.L., 1998. Geochemistry of coalbed gas – A review. Int J. Coal Geology, 35: 159-173.

Coppens, L., 1936. Ann. Mines Belg., 37: 173. Cramer, B., Krooss, B.M. and Littke, R., 1998. Modelling isotope fractionation during

primary cracking of natural gas: a reaction kinetic approach. Chemical Geology, 149(3–4): 235-250.

Crank, J., 1975. The Mathematics of Diffusion. Oxford University Press, 414 pp. Crosdale, P.J., 1999. Mixed Methane/carbon Dioxide sorption by coal: New evidence in

support of pore-filling models. Int. CBM Symposium, Univ. of Alabama, Tuscaloosa: 359-366.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

257 | P a g e

Crosdale, P.J., Beamish, B.B. and Valix, M., 1998. Coalbed methane sorption related to coal composition. International Journal of Coal Geology, 35(1-4): 147-158.

Crosdale, P.J., Moore, T.A. and Mares, T.E., 2008. Influence of moisture content and temperature on methane adsorption isotherm analysis for coals from a low-rank, biogenically-sourced gas reservoir. International Journal of Coal Geology, 76(1-2): 166-174.

Cui, X., Bustin, R. M., Dipple, G., 2004. Selective transport of CO2, CH4, and N2 in coals: insights from modeling of experimental gas adsorption data. Fuel, 83: 293-303.

Darcey, J.R., Clunie, J.C. and Thomas, D.G., 1958. The adsorption of water by Saran charcoal. Transactions of the Faraday Society, 54: 250-256.

Darcey, J.R., Clunie, J.C.,Thomas, D.G., 1958. The adsorption of water by Saran charcoal Transactions of the Faraday Society, 54: 250-256.

Day, S., Duffy, G., Sakurovs, R. and Weir, S., 2008a. Effect of coal properties on CO2 sorption capacity under supercritical conditions. International Journal of Greenhouse Gas Control, 2(3): 342-352.

Day, S., Fry, R. and Sakurovs, R., 2008b. Swelling of Australian coals in supercritical CO2. International Journal of Coal Geology, 74(1): 41-52.

Day, S., Sakurovs, R. and Weir, S., 2008c. Supercritical gas sorption on moist coals. International Journal of Coal Geology, 74(3-4): 203-214.

Day, S.J. et al., 2005. Factors controlling CO2 absorption in Australian coals, Greenhouse Gas Control Technologies 7. Elsevier Science Ltd, Oxford, pp. 2221-2224.

De Weireld, G., Frere, M. and Jadot, R., 1999. Automated determination of high-temperature and high-pressure gas adsorption isotherms using a magnetic suspension balance. Meas. Sci. Technol., 10: 117–126.

DeGance, A.E., Morgan, W. D., Yee, D., 1993. High Pressure Adsorption of Methane, Nitrogen and Carbon Dioxide on Coal Substrates. Fluid Phase Equilibria, 82: 215-224.

Del Bene, J. and Pople, J.A., 1969. Intermolecular energies of small water polymers. Chemical Physics Letters, 4(7): 426-428.

Diamond, W.P. and Schatzel, S.J., 1998. Measuring the gas content of coal: A review. International Journal of Coal Geology, 35(1-4): 311-331.

Do, D.D., 1998. Adsorption Analysis: Equilibria and Kinetics. Imperial College Press. Dreisbach, F., Staudt, R. & Keller, J.U., 1999. High pressure adsorption data of methane,

nitrogen, carbon dioxide and their binary and ternary mixtures on activated carbon. Adsorption, 5: 215-227.

Duan, Z. and Sun, R., 2003. An improved model calculating CO2 solubility in pure water and aqueous NaCl solutions from 273 to 533 K and from 0 to 2000 bar. Chemical Geology, 193: 257-271.

Dubinin, M.M., 1980. Water vapor adsorption and the microporous structures of carbonaceous adsorbents. Carbon, 18(5): 355-364.

Dubinin, M.M. and Serpinsky, V.V., 1981. Isotherm equation for water vapor adsorption by microporous carbonaceous adsorbents. Carbon, 19(5): 402-403.

Dutta, P., Bhowmik, S. and Das, S., 2011. Methane and carbon dioxide sorption on a set of coals from India. International Journal of Coal Geology, 85(3-4): 289-299.

Erickson, D. and Jensen, J.R., 2001. CO2 sequestration in an unminable coalbed- San Juan Basin, Colorado, U.S.A. In: D. Williams, B. Durie, P. McMullan, C. Paulson and A. Smith (Editors), Proceedings of the Fifth International Conference on Greenhouse Control Technologies, CSIRO Publishing, Cairns, Australia, pp. 589–592.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

258 | P a g e

Faiz, M., Saghafi, A., Sherwood, N. and Wang, I., 2007. The influence of petrological properties and burial history on coal seam methane reservoir characterisation, Sydney Basin, Australia. International Journal of Coal Geology, 70(1-3): 193-208.

Faiz, M.M., 2004. Microbial influences on coal seam gas reservoirs - A review, Australasian Institute of Mining and Metallurgy Publication Series, pp. 133-142.

Faiz, M.M., Saghafi, A., Barclay, S. A., Stalker, L., Sherwood, N. R., Whitford, D. J., 2007. Evaluating geological sequestration of CO2 in bituminous coals: The southern Sydney Basin, Australia as a natural analogue. International Journal of Greenhouse Gas Control, 1(2): 223-235.

Faiz, M.M., Saghafi, A., Barclay, S.A., Stalker, L., Sherwood, N.R., Whitford, D.J., 2006. Distribution of Juvenile CO2 in the Sydney Basin, Australia: A Natural Analogue for CO2 Storage in Bituminous Coals, 8th International Conference on Greenhouse Gas Technology (GHGT8), Trondheim, Norway, pp. 726-733.

Fei, Y., Marshall, M., Jackson, W.R., Chaffee, A.L. and Allardice, D.J., 2006. A comparison of adsorption isotherms using different techniques for a range of raw, water- and acid-washed lignites. Fuel, 85(10-11): 1559-1565.

Fei, Y. et al., 2007. The effect of cation content of some raw and ion-exchanged Victorian lignites on their equilibrium moisture content and surface area. Fuel, 86(17-18): 2890-2897.

Fitzgerald, J.E. et al., 2005. Adsorption of methane, nitrogen, carbon dioxide and their mixtures on wet Tiffany coal. Fuel, 84(18): 2351-2363.

Fitzgerald, J.E., Robinson Jr., R.L., Gasem, K.A.M., 2006. Modeling High-Pressure Adsorption of Gas Mixtures on Activated Carbon and Coal Using a Simplified Local-Density Model. Languir, 22: 9610-9618.

Gamson, P., Beamish, B. and Johnson, D., 1996. Coal microstructure and secondary mineralization: their effect on methane recovery. Geological Society, London, Special Publications, 109(1): 165-179.

Gaschnitz, R., 2000. Gasgenese und Gasspeicherung im flözführenden Oberkarbon des Ruhr-Beckens (in German), RWTH-Aachen University, Aachen.

Gasem, K.A.M., Robinson, Jr., L. & Reeves, S.R., 2002a. Adsorption of pure methane, nitroge and carbon dioxide and their mixtures on San Juan Basin coals. Topical Report.

Gasem, K.A.M., Sudibandriyo, M., Fitzgerald, J.E. Pan, Z. & Robinson Jr., R.L., 2002b. Measurement and modelling of gas adsorption on selected coals. Proceesings of the AICHE Spring National Meeting, New Orleans, Lousiana, March 10-16, 2002.

Gasparik, M. et al., 2012. High-pressure methane sorption isotherms of black shales from the Netherlands. Energy and Fuels, 26(8): 4995-5004.

Gauden, P.A., 2005. Does the Dubinin-Serpinsky theory adequately describe water adsorption on adsorbents with high-energy centers? Journal of Colloid and Interface Science, 282(2): 249-260.

Gensterblum, Y., 2013. CBM and CO2-ECBM related sorption processes on coals, RWTH Aachen University, Aachen, 360 pp.

Gensterblum, Y., Merkel, A., Busch, A. and Krooss, B.M., 2013a. High-pressure CH4 and CO2 sorption isotherms as a function of coal maturity and the influence of moisture. International Journal of Coal Geology, submitted.

Gensterblum, Y., Merkel, A., Busch, A. and Krooss, B.M., 2013b. Molecular concept and experimental evidence of H2O, CH4 & CO2 sorption on organic material. Fuel.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

259 | P a g e

Gensterblum, Y. et al., 2010a. European inter-laboratory comparison of high pressure CO2 sorption isotherms. II: natural coals. International Journal of Coal Geology, accepted.

Gensterblum, Y. et al., 2010b. European inter-laboratory comparison of high pressure CO2 sorption isotherms II: Natural coals. International Journal of Coal Geology, 84(2): 115-124.

Gensterblum, Y. et al., 2009a. European inter-laboratory comparison of high pressure CO2 sorption isotherms. I: Activated carbon. Carbon, 47(13): 2958-2969.

Gensterblum, Y. et al., 2009b. European inter-laboratory comparison of high pressure CO2 sorption isotherms. I: Activated carbon. Carbon, 47: 2958-2969.

Given, P.H., Marzec, A., Barton, W.A., Lynch, L.J., Gerstein, B.C., 1986. The concept of a mobile or molecular phase within the macromolecular network of coals: A debate. Fuel, 65: 155-163.

Glymour, C., 1980. Theory and Evidence. Princeton University Press. Goodman, A.L. et al., 2007a. Inter-laboratory comparison II: CO2 isotherms measured on

moisture-equilibrated Argonne premium coals at 55°C and up to 15 MPa. International Journal of Coal Geology, 72(3-4): 153-164.

Goodman, A.L. et al., 2007b. Inter-laboratory comparison II: CO2 isotherms measured on moisture-equilibrated Argonne premium coals at 55 °C and up to 15 MPa. International Journal of Coal Geology, 72(3-4): 153-164.

Goodman, A.L. et al., 2004. An Inter-laboratory Comparison of CO2 Isotherms Measured on Argonne Premium Coal Samples. Energy and Fuels, 18: 1175-1182.

Goodman, A.L., Campus, L.M., Schroeder, L.M., 2005a. Direct Evidence of Carbon Dioxide Sorption on Argonne Premium Coals Using Attenuated Total Reflectance-Fourier Transform Infrared Spectroscopy. Energy and Fuel, 19: 471-476.

Goodman, A.L., Favors, R.N., Hill, M.M. and Larsen, J.W., 2005. Structure Changes in Pittsburgh No. 8 Coal Caused by Sorption of CO2 Gas. Energy & Fuels, 19(4): 1759-1760.

Goodman, A.L., Favors, R.N. and Larsen, J.W., 2006. Argonne Coal Structure Rearrangement Caused by Sorption of CO2. Energy & Fuels, 20(6): 2537-2543.

Goodman, A.L., Favrs, R.N., Hill, M.M., Larsen, J.W., 2005b. Structure changes in Pittsburgh No. 8 coal causewd by sorption of CO2 gas. Energy and Fuels, 19: 1759-1760.

Gregg, S.J.a.S., K.S.W., 1967. Adsorption, surface area and porosity. Academic Press, London and New York, 195 pp.

Gruszkiewicz, M.S. et al., 2009. Adsorption kinetics of CO2, CH4, and their equimolar mixture on coal from the Black Warrior Basin, West-Central Alabama. International Journal of Coal Geology, 77(1-2): 23-33.

Gumma, S. and Talu, O., 2003. Gibbs Dividing Surface and Helium Adsorption. Adsorption, 9(1): 17-28.

Gunter, W.D., Mavor, M.J., Robinson, J.R., 2004. CO2 storage and enhanced methane production: Field testing at Fenn-Big Valley, lberta, Canada, with application, 7th International Conference on Greenhouse Gas Control Technology, Vancouver, Canada.

Gutierrez-Rodriguez, J.A., Purcell, R.J. Jr.,Aplan F.F., 1984. Estimating the Hydrophobicity of coal. Colloids and Surfaces, 12: 1-25.

Hall, F.E.Z., C.; Gasem, K.A.M. & Robinson Jr., R.L., 1994. Adsorption of pure methane, nitrogen and carbon dioxide and their binary mixtures on wet Fruitland coals. SPE, 29194: 329-344.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

260 | P a g e

Hanzawa, Y., Kaneko, K., Yoshizawa, N., Pekala, R.W., Dresselhaus, M.S., 1998. The pore structure determination of carbon aerogels. Adsorption, 4: 187-195.

Harpalani, S., Chen, G., 1997. Influence of gas production induced volumetric strain on permeability of coal. Geotechnical and Geological Engineering, 15: 303-325.

Harpalani, S., Prusty, B.K. and Dutta, P., 2006. Methane/CO2 Sorption Modeling for Coalbed Methane Production and CO2 Sequestration. Energy and Fuels, 20: 1591-1599.

Hartman, C., Pratt, T.J., 2005. A Preliminary Study of the Effect of Moisture on the Carbon Dioxide Storage Capacity in Coal, paper 0533, Proceedings 2005 International Coalbed Methane Symposium, University

of Alabama, Tuscaloosa, pp. 9. Hemza, P., Sivek, M. and Jirásek, J., 2009. Factors influencing the methane content of coal

beds of the Czech part of the Upper Silesian Coal Basin, Czech Republic. International Journal of Coal Geology, 79(1-2): 29-39.

Heuchel, M., Davies, G.M., Buss, E., Seaton, N.A., 1999. Adsorption of Carbon Dioxide and Methane and Their Mixtures on an Activated Carbon: Simulation and Experiment. Langmuir, 15: 8695-8705.

Hildenbrand, A., Krooss, B.M., Busch, A. and Gaschnitz, R., 2006a. Evolution of methane sorption capacity of coal seams as a function of burial history -- a case study from the Campine Basin, NE Belgium. International Journal of Coal Geology, 66(3): 179-203.

Hildenbrand, A., Krooss, B.M., Busch, A. and Gaschnitz, R., 2006b. Evolution of methane sorption capacity of coal seams as a function of burial history - A case study from the Campine Basin, NE Belgium. International Journal of Coal Geology, 66(3): 179-203.

Hirsch, P.B., 1954. X-Ray Scattering from Coals. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, 226(1165): 143-169.

Hol, S., Gensterblum, Y. and Spiers, C.J., 2013. Direct determination of total CO2 uptake by coal: A new technique compared with the manometric method. Fuel, 105: 192-205.

Hol, S., Peach, C.J. and Spiers, C.J., 2010. A new experimental method to determine the CO2 sorption capacity of coal, GHGT 10. Energy Procedia, Amsterdam, the Netherlands.

Hol, S., Peach, C.J. and Spiers, C.J., 2011. A new experimental method to determine the CO2 sorption capacity of coal. Energy Procedia, 4: 3125-3130.

Horikawa, T., Sekida, T., Hayashi, J.i., Katoh, M. and Do, D.D., 2010. A new adsorption-desorption model for water adsorption in porous carbons. Carbon, 49(2): 416-424.

Huang, H., Wang, K., Bodily, D.M. and Hucka, V.J., 1995. Density Measurements of Argonne Premium Coal Samples. Energy & Fuels, 9(1): 20-24.

Humayun, R. and Tomasko, D.L., 2000. High-Resolution Adsorption Isotherms of Supercritical Carbon Dioxide on Activated Carbon. AlCHE Journal, 46(10): 2065-2075.

Hunt, A.M. and Steele, D.J., 1991. Coalbed Methane Development in the Appalachian Basin. Quarterly Review of Methane from Coalseams Technology, 8(4): 10-19.

Hvoslef, S., Larter, Sr. and Leythaeuser, D., 1988. Aspects of generation and migration of hydrocarbons from coal-bearing strata of the Hitra Formation, Haltenbanken area, offshore Norway. Organic Geochemistry, 13(1-3): 525-536.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

261 | P a g e

Iiyama, T., Nishikawa, K., Suszuki, T., Kaneko, K., 1997. Study of the structure of a water molecular assembly in a hydrophobic nanospace at low temperature with in situ X-ray diffraction. Chemical Physics Letters, 274: 152-158.

IUPAC, 2001. Manual of symbols and terminology for physicochemical quantities and units.

Joubert, J.I., Clifford, T. and Bienstock, D., 1973a. Sorption of Methane in moist coal. Fuel, 52: 181-185.

Joubert, J.I., Grein, C.T. and Bienstock, D., 1973b. Sorption of methane in moist coal. Fuel, 52(3): 181-185.

Joubert, J.I., Grein, C.T. and Bienstock, D., 1974. Effect of moisture on the methane capacity of American coals. Fuel, 53: 186-191.

Juch, D., Gaschnitz, R., Thielemann, T., 2004. The influence of geological history on coal mine gas distribution in the Ruhr District - A challenge for future research and recovery. Geologica Belgica, 7: 191-199.

Jüntgen, H., 1987. Research for future in situ conversion of coal. Fuel, 66(4): 443-453. Jüntgen, H. and Klein, J., 1974. Entstehung von Erdgas aus kohligen Sedimenten, 24.

General assembling of the German Society for Petroleum and Coal Science and Technology. German Society for Petroleum and Coal Science and Technology, Hamburg, pp. 52-69.

Karacan, C.Ö., 2003. Heterogeneous Sorption and Swelling in a Confined and Stressed Coal during CO2 Injection. Energy and Fuels, 17: 1595-1608.

Karacan, C.Ö., 2007. Swelling-induced volumetric strains internal to a stressed coal associated with CO2 sorption. International Journal of Coal Geology.

Karge, H., Weitkamp, J. and Ruthven, D., 2008. Fundamentals of Adsorption Equilibrium and Kinetics in Microporous Solids, Adsorption and Diffusion. Molecular Sieves - Science and Technology. Springer Berlin / Heidelberg, pp. 1-43.

Kedzior, S., 2009. Accumulation of coal-bed methane in the south-west part of the Upper Silesian Coal Basin (southern Poland). International Journal of Coal Geology, 80(1): 20-34.

Kelemen, S.R. and Kwiatek, L.M., 2009. Physical properties of selected block Argonne Premium bituminous coal related to CO2, CH4, and N2 adsorption. International Journal of Coal Geology, 77(1-2): 2-9.

Kelemen, S.R., Kwiatek, L.M., Lee, A.G.K., 2006. Swelling and Sorption Response of Selected Argonne Premium Bituminous Coals to CO2, CH4, and N2, International CBM Symposium, Tuscaloosa, Alabama.

Kemball, C., W.G. Frankenburg, V.I.K. and Rideal, E.K., 1950. Entropy of Adsorption, Advances in Catalysis. Academic Press, pp. 233-250.

Kotarba, M.J. and Rice, D.D., 2001. Composition and origin of coalbed gases in the Lower Silesian basin, southwest Poland. Applied Geochemistry, 16(7-8): 895-910.

Kronimus, A. et al., 2008. A preliminary evaluation of the CO2 storage potential in unminable coal seams of the Münster Cretaceous Basin, Germany. International Journal of Greenhouse Gas Control, 2(3): 329-341.

Krooss, B.M., 1988. Experimental investigation of the molecular migration of C1-C6 hydrocarbons: Kinetics of hydrocarbon release from source rocks. Organic Geochemistry, 13(1-3): 513-523.

Krooss, B.M., Leythaeuser, D. and Schaefer, R.G., 1988. Light hydrocarbon diffusion in a caprock. Chemical Geology, 71(1–3): 65-76.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

262 | P a g e

Krooss, B.M. et al., 1995a. Generation of nitrogen and methane from sedimentary organic matter: Implications on the dynamics of natural gas accumulations. Chemical Geology, 126(3–4): 291-318.

Krooss, B.M. et al., 1995b. Generation of nitrogen and methane from sedimentary organic matter: Implications on the dynamics of natural gas accumulations. Chemical Geology, 126(3-4): 291-318.

Krooss, B.M. et al., 2002. High-pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated Pennsylvanian coals. International Journal of Coal Geology, 51: 69-92.

Kunz, O., Klimeck, R., Wagner, W. and Jaeschke, M., 2007. The GERG-2004 wide-range reference equation of state for natural gases. GERG Technical Monograph 15 Fortschritts-Berichte Verein Deutscher Ingenieue (VDI), 6. VDI-Verlag, Düsseldorf.

Lama, R.D., Bodziony, J., 1998. Management of outburst in underground coal mines. Int J. Coal Geology, 35: 83-115.

Lamberson, M.N. and Bustin, R.M., 1993. Coalbed Methane characteristics of Gates formation coals, Northeastern British Columbia: Effect of maceral composition. AAPG Bulletin, 77(12): 2062-2076.

Langmuir, I., 1916. The constitution and fundamental properties of solids and liquids. part i. solids. Journal of the American Chemical Society, 38(11): 2221-2295.

Langmuir, I., 1932. Surface chemistry, Nobelprize Lecture. Swedish Academy of Sciences, Stockholm.

Larsen, J.W., 2004. The effects of dissolved CO2 on coal structure and properties. Int J. Coal Geology, 57: 63– 70.

Laubach, S.E., Marrett, R. A., Olson, J. E., Scott, A. R., 1998. Characteristics and origins of coal cleat: A review. Int J. Coal Geology, 35: 175-207.

Laxminarayana, C. and Crosdale, P.J., 1999a. Role of coal type and rank on methane sorption characteristics of Bowen Basin, Australia coals. International Journal of Coal Geology, 40(4): 309-325.

Laxminarayana, C. and Crosdale, P.J., 1999b. Role of coal type and rank on methane sorption charakteristics of Bowen Basin, Australia coals. Int. J. Coal Geology, 40: 309-325.

Laxminarayana, C. and Crosdale, P.J., 2002. Controls on methane sorption capacity of Indian coals. AAPG Bulletin, 86(2): 201-212.

Levy, J.H., Day, S.J. and Killingley, J.S., 1997. Methane capacities of Bowen Basin coals related to coal properties. Fuel, 76(9): 813-819.

Li, D. et al., 2010. High-pressure sorption isotherms and sorption kinetics of CH4 and CO2 on coals. Fuel, 89(3): 569-580.

Littke, R., Baker, D.R. and Leythaeuser, D., 1988. Microscopic and sedimentologic evidence for the generation and migration of hydrocarbons in Toarcian source rocks of different maturities. Organic Geochemistry, 13(1–3): 549-559.

Littke, R., Krooss, B., Uffmann, A.K., H.M., S. and Horsfield, B., 2011. Unconventional Gas Resources in the Paleozoic of Central Europe Oil & Gas Science and Technology.

Littke, R., Lo ten Haven, H. and Leythaeuser, D., 1990. Paleoecological trends and petroleum potential of Upper Carboniferous coal seams of Western Germany. International Journal of Coal Geology, 16(1–3): 197-200.

Littke, R., Urai, J.L., Uffmann, A.K. and Risvanis, F., 2012. Reflectance of dispersed vitrinite in Palaeozoic rocks with and without cleavage: Implications for burial

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

263 | P a g e

and thermal history modeling in the Devonian of Rursee area, northern Rhenish Massif, Germany. International Journal of Coal Geology, 89(1): 41-50.

Liu, Y. and Wilcox, J., 2012a. Effects of surface heterogeneity on the adsorption of CO 2 in microporous carbons. Environmental Science and Technology, 46(3): 1940-1947.

Liu, Y. and Wilcox, J., 2012b. Molecular simulation of CO2 adsorption in micro- and mesoporous carbons with surface heterogeneity. International Journal of Coal Geology, 104(0): 83-95.

Lopatin, N.V., 1971. Temperature and geologic time as factors in coalification (in Russian). Akad. Nauk SSSR Izv. Ser. Geol., 3: 95-106.

Lu, L., Sahajwalla, V., Kong, C. & Harris, D., 2001. Quantitative X-ray diffraction analysis and its application to various coals. Carbon, 39: 1821–1833.

Lynch, L.J. and Webster, D.S., 1982. Effect of thermal treatment on the interaction of brown coal and water: a nuclear magnetic resonance study. Fuel, 61(3): 271-275.

Maddox, M., Ulberg, D. and Gubbins, K.E., 1995. Molecular simulation of simple fluids and water in porous carbons. Fluid Phase Equilibria, 104: 145-158.

Mahajan, O.P., 1991. CO2 surface area of coals: A 25 year paradox. Carbon, 29: 735-742. Mahajan, O.P., Walker, P.L. Jr., 1970. Water adsorption on coals. Fuel, 50(3): 308-317. Majewska, Z., Ceglarska-Stefanska, G., Majewski, S. and Zietek, J., 2009. Binary gas

sorption/desorption experiments on a bituminous coal: Simultaneous measurements on sorption kinetics, volumetric strain and acoustic emission. International Journal of Coal Geology, 77(1-2): 90-102.

Marecka, A.M., A., 1998. Kinetics of CO2 and CH4 sorption on high rank coal at ambient temperatures. Fuel, 77(14): 1691-1696.

Mares, T.E., Moore, T.A. and Moore, C.R., 2009. Uncertainty of gas saturation estimates in a subbituminous coal seam. International Journal of Coal Geology, 77(3-4): 320-327.

Mastalerz, M., Gluskoter, H. and Rupp, J., 2004. Carbon dioxide and methane sorption in high volatile bituminous coals from Indiana, USA. Int J. Coal Geology, 60: 43-55.

Mavor, M.J., Hartman, C., Pratt, T.J., 2004. Uncertainty in Sorption Isotherm Measurements, 2004 CBM Symposium, Tuscaloosa, Alabama, pp. 14.

Mavor, M.J., Pratt, T.J., Nelson, C.R., 1995. Quantitative Evaluation of Coal Seam Gas Content Estimate Accuracy, SPE Rocky Mountain, Regione/Low-Permeability Reservoir Symposium, Denver, CO, pp. 327-340.

Mavor, M.J.O., L.B. & Pratt, T.J., 1990. Measurement and evaluation of coal sorption isotherm data. SPE, 20728: 157-170.

Mazumder, S., Siemons, N., Wolf, K.-H., 2005. Differential swelling and permeability change of coal in response to CO2 injection for ECBM. SPE, 98475.

Mazumder, S., van Hemert, P., Busch, A., Wolf, K.H.A.A. and Tejera-Cuesta, P., 2006. Flue gas and pure CO2 sorption properties of coal: A comparative study. International Journal of Coal Geology, 67(4): 267-279.

Mazumder, S. and Wolf, K.H., 2008. Differential swelling and permeability change of coal in response to CO2 injection for ECBM. International Journal of Coal Geology, 74(2): 123-138.

McCarty, R.D. and Arp, V.D., 1990. A new wide range equation of state for helium. Advances in Cryogenic Engineering, 35: 1465-1475.

McCutcheon, A.L., Barton, W.A. and Wilson, M.A., 2003. Characterization of Water Adsorbed on Bituminous Coals. Energy & Fuels, 17(1): 107-112.

McCutcheon, A.L., Barton, W.A., Wilson, M.A., 2001. Kinetics of Water Adsorption/Desorption on Bituminous coals. Energy and Fuel, 15: 1387-1395.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

264 | P a g e

Metcalfe, R.S., Yee, D., Seidle, J.P. and Puri, R., 1991. Review of Research Efforts in Coalbed Methane Recovery, SPE Asia-Pacific Conference. Society of Petroleum Engineers, Perth Australia.

Michels, G. and Wouters, H., 1941. Zahlenwerte und Funktionen aus Physik, Chemie, Astromomie Geophysik und Technik, 2. Springer.

Middleton, M.F. and Falvey, D.A., 1983. MATURATION MODELING IN OTWAY BASIN, AUSTRALIA. Aapg Bulletin-American Association of Petroleum Geologists, 67(2): 271-279.

Milligan, J.B., Thomas, K. M., Crelling, J. C., 1997. Solvent Swelling of Maceral Concentrates. Energy and Fuels, 11: 364-371.

Mohammad, S., Fitzgerald, J., Robinson, R.L. and Gasem, K.A.M., 2009. Experimental Uncertainties in Volumetric Methods for Measuring Equilibrium Adsorption. Energy & Fuels, 23(5): 2810-2820.

Mohammed S, J.F., RL Robinson Jr, KAM Gasem, 2009. Experimental uncertainties in volumetric methods for measuring equilibrium adsorption. Energy and Fuels, 23: 2810-2820.

Moore, T.A., 2012. Coalbed methane: A review. International Journal of Coal Geology, 101: 36-81.

Moore, T.A. and Crosdale, P.J., 2006. The effect of moisture and temperature on adsorption isotherms of a low-rank coal: Implications for reservoir modeling, International CBM Symposium. SPE, Tuscaloosa, Alabama.

Mosher, K., He, J., Liu, Y., Rupp, E. and Wilcox, J., 2013. Molecular simulation of methane adsorption in micro- and mesoporous carbons with applications to coal and gas shale systems. International Journal of Coal Geology, 109-110: 36-44.

Mu, R. and Malhotra, V.M., 1991. A new approach to elucidate coal-water interactions by an in situ transmission FT-i.r. technique. Fuel, 70(10): 1233-1235.

Muller, E.A., Rull, L.F., Vega, L.F. and Gubbins, K.E., 1996. Adsorption of Water on Activated Carbons: A Molecular Simulation Study. The Journal of Physical Chemistry, 100(4): 1189-1196.

Murata, K., El-Merraoui, M. and Kaneko, K., 2001. A new determination method of absolute adsorption isotherm of supercritical gases under high pressure with a special relevance to density-functional theory study. The Journal of Chemical Physics, 114(9): 4196-4205.

Myers, A.L., 2004. Thermodynamics of Adsorption. In: T.M. Letcher (Editor), Chemical Thermodynamics for Industry, Cambridge, UK.

Narkiewicz, M.R. and Mathews, J.P., 2009. Visual representation of carbon dioxide adsorption in a low-volatile bituminous coal molecular model. Energy and Fuels, 23(10): 5236-5246.

Nishino, J., 2001. Adsorption of water vapor and carbon dioxide at carboxylic functional groups on the surface of coal. Fuel, 80: 757-764.

Nodzenski, A., 1998. Sorption and desorption of gases (CH4, CO2) on hard coal and active carbon at elevated pressures. Fuel, 77(11): 1243-1246.

O'Sullivan, T. and Smith, N.O., 1969. The Solubility and partial molar volume of nitrogen and methane in water and in aqueous sodium chloride from 50 to 125C and 100 to 600Atm. The Journal of Physical Chemistry, 74(7): 1460-1466.

Ohga, K., Fujioka, M., Yamaguchi, S., Nako, M., 2006. Field tests of CO2 sequestration and ECBM in Japan International CBM Symposium, Tuscaloosa, Al.

Ottiger, S., Pini, R., Storti, G. and Mazzotti, M., 2008. Competitive adsorption equilibria of CO2 and CH4 on a dry coal. Adsorption, 14(4-5): 539-556.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

265 | P a g e

Ottiger, S. et al., 2006. Adsorption of Pure Carbon Dioxide and Methane on Dry Coal from the Sulcis Coal Province (SW Sardinia, Italy). Environmental Progress, 25: 355-364.

Ozdemir, E., 2009. Modeling of coal bed methane (CBM) production and CO2 sequestration in coal seams. International Journal of Coal Geology, 77(1-2): 145-152.

Ozdemir, E., Morsi, B. I., Schroeder, K., 2004. CO2 adsorption capacity of Argonne Premium Coals. Fuel, 83: 1085-1094.

Ozdemir, E., Morsi, B.I., Schroeder, K., 2003. Importance of volume effects to adsorption isotherms of carbon dioxide on coals. Langmuir, 19: 9764-9773.

Ozdemir, E. and Schroeder, K., 2009. Effect of Moisture on Adsorption Isotherms and Adsorption Capacities of CO2 on Coals. Energy & Fuels, 23(5): 2821–2831.

Palmer, I. and Mansoori, J., 1998. How Permeability Depends on Stress and Pore Pressure in Coalbeds: A New Model. SPE Reservoir Evaluation & Engineering, SPE 52607: 539-544.

Pan, Z. and Connell, L.D., 2007. A theoretical model for gas adsorption-induced coal swelling. International Journal of Coal Geology, 69(4): 243-252.

Pan, Z., Connell, L.D., Camilleri, M. and Connelly, L., 2010. Effects of matrix moisture on gas diffusion and flow in coal. Fuel, 89(11): 3207-3217.

Peng, D.Y., Robinson, D.B., 1976. A New Two-Constant Equation of State. Ind. Eng. Chem. Fundam., 15(1): 59-64.

Pini, R., Ottiger, S., Burlini, L., Storti, G. and Mazzotti, M., 2010. Sorption of carbon dioxide, methane and nitrogen in dry coals at high pressure and moderate temperature. International Journal of Greenhouse Gas Control, 4(1): 90-101.

Pini, R., Ottiger, S., Rajendran, A., Storti, G. and Mazzotti, M., 2006. Reliable measurement of near-critical adsorption by gravimetric method. Adsorption, 12(5): 393-403.

Pini, R., Ottiger, S., Burlini, L., Storti, G., Mazzotti, M., Bencini, R., Quattrocchi, F., 2006. Experimental study of CO2 sequestration by ECBM recovery: the case of Sulcis coal, GHGT8, Trondheim, Norway.

Pone, J.D.N., Halleck, P.M. and Mathews, J.P., 2009. Sorption Capacity and Sorption Kinetic Measurements of CO2 and CH4 in Confined and Unconfined Bituminous Coal. Energy & Fuels, 23(9): 4688-4695.

Price, L.C., 1983. Geologic time as a paramter in organic metamorphism and vitrinte reflectance as an absolute paleogeothermometer. Journal of Petroleum Geology, 6(1): 5-38.

Prinz, D., 2004. Die Porenstruktur von Kohlen (in German), RWTH Aachen University, Germany, Aachen, 184 pp.

Prinz, D. and Littke, R., 2005a. Development of the micro- and ultramicroporous structure of coals with rank as deduced from the accessibility to water. Fuel, 84(12-13): 1645-1652.

Prinz, D. and Littke, R., 2005b. Development of the micro- and ultramicroporous structure of coals with rank as deduced from the accessibility to water. Fuel, 84(12-13): 1655-1662.

Prinz, D., Pyckhout-Hintzen, W. and Littke, R., 2004. Development of the meso- and macroporous structure of coals with rank as analyzed with small angle neutron scattering and adsorption experiments. Fuel, 83: 547-556.

Prinz, D., Pyckhout-Hintzen, W., Krooss, B.M. & Littke, R., 2001. Investigations on the Micro- and Mesoporous Structure of Coals of Varying Rank: A Combined Small

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

266 | P a g e

Angle Neutron Scattering (SANS) and Gas Adsorption Experiments Study. Int. CBM Symposium, Tuscaloosa.

Puri, B.R., 1966. Chemisorbed oxygen evolved as carbon dioxide and its influence on surface reactivity of carbons. Carbon, 4(3): 391-400.

Reeves, S., Gonzalez, R., Gasem, K.A.M., Fitzgerald, J.E., Pan, Z., Sudibandriyo, M., Robinson Jr., R.L., 2005. Measurement and Prediction of Single- and Multi-Component Methane, Carbon Dioxide and Nitrogen Isotherms for U.S. Coals, ICBM Symposium Tuscaloosa, Tuscaloosa, Alabama.

Reeves, S., Taillefert, A., Pekot, L. and Clarkson, C., 2003. The Allison Unit CO2 - ECBM Pilot: A Reservoir Modeling Study, Topical Report, U.S. Department of Energy (DE-FC26-0NT40924).

Reucroft, P.J., Patel, H., 1986. Gas-induced swelling in coal. Fuel, 65: 816-820. Rice, D.D., 1993. Composition and origins of coalbed gas. In: B.E. Law, Rice, D.D (Editor),

Hydrocarbons from coal. AAPG studies in Geology, pp. 159-184. Rice, D.D., Law, B.E. and Clayton, J.L., 1993. Coalbed gas - an undeveloped resource. US

Geological Survey Professional Paper, 1570: 389-404. Rubes, M., Kysilka, J., Nachtigall, P. and Bludski, O., 2010. DFT/CC investigation of

physical adsorption on a graphite (0001) surface. Physical Chemistry Chemical Physics, 12(24): 6438-6444.

Rubes, M., Nachtigall, P., Vondrasek, J. and Bludsky, O., 2009. Structure and Stability of the Water-Graphite Complexes. The Journal of Physical Chemistry C, 113(19): 8412-8419.

Ruckenstein, E., Vaidyanathan, S. & Youngquist, G.R., 1971. Sorption by solids with bidisperse pore structure. Chem. Eng. Science, 26: 1305-1318.

Ruthven, D.M., 1984. Principles of adsorption and adsorption processes. Wiley. Ryan, B., Richardson, D., 2002. The potential for CO2 sequestration in British Columbia

coal seams, British Columbia Geological Survey. Saghafi, A., Faiz, M. and Roberts, D., 2007. CO2 storage and gas diffusivity properties of

coals from Sydney Basin, Australia. International Journal of Coal Geology, 70(1-3): 240-254.

Sakurovs, R., Day, S. and Weir, S., 2009. Causes and consequences of errors in determining sorption capacity of coals for carbon dioxide at high pressure. International Journal of Coal Geology, 77(1-2): 16-22.

Sakurovs, R., Day, S. and Weir, S., 2010. Relationships between the critical properties of gases and their high pressure sorption behavior on coals. Energy and Fuels, 24(3): 1781-1787.

Sakurovs, R., Day, S. and Weir, S., 2013. Relationships between the sorption behaviour of methane, carbon dioxide, nitrogen and ethane on coals. Fuel, 97(0): 725-729.

Sakurovs, R., Day, S., Weir, S. and Duffy, G., 2008. Temperature dependence of sorption of gases by coals and charcoals. International Journal of Coal Geology, 73(3-4): 250-258.

Sakurovs, R., Day, S., Weir, S., Duffy, G., 2007. Application of a Modified Dubinin-Radushkevich Equation to Adsorption of Gases by Coals under Supercritical Conditions. Energy and Fuels, 21: 992-997.

Sakurovs, R. et al., 2012. Pore size distribution and accessible pore size distribution in bituminous coals. International Journal of Coal Geology, 100(0): 51-64.

Salame, I.I. and Bandosz, T.J., 1998. Experimental Study of Water Adsorption on Activated Carbons. Langmuir, 15(2): 587-593.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

267 | P a g e

Salame, I.I. and Bandosz, T.J., 1999. Study of Water Adsorption on Activated Carbons with Different Degrees of Surface Oxidation. Journal of Colloid and Interface Science, 210(2): 367-374.

Salame, I.I. and Bandosz, T.J., 2000. Adsorption of Water and Methanol on Micro- and Mesoporous Wood-Based Activated Carbons. Langmuir, 16(12): 5435-5440.

Schoenherr, J., Littke, R., Urai, J.L., Kukla, P.A. and Rawahi, Z., 2007. Polyphase thermal evolution in the Infra-Cambrian Ara Group (South Oman Salt Basin) as deduced by maturity of solid reservoir bitumen. Organic Geochemistry, 38(8): 1293-1318.

Schuyer, J., Blom, L. and Vankrevelen, D.W., 1953. The molar refraction of condensed aromatic compounds. Transactions of the Faraday Society, 49(12): 1391-1401.

Scott, A.R., 2002. Hydrogeologic factors affecting gas content distribution in coal beds. International Journal of Coal Geology, 50(1-4): 363-387.

Scott, S., Anderson, B., Crosdale, P., Dingwall, J. and Leblang, G., 2007. Coal petrology and coal seam gas contents of the Walloon Subgroup - Surat Basin, Queensland, Australia. International Journal of Coal Geology, 70(1-3 SPEC. ISS.): 209-222.

Seewald, H.a.K., J., 1986. Methansorption an Steinkohle und Kennzeichnung der Porenstruktur. Glückauf-Forschungshefte, 47(3): 149-156.

Senglaub, Y., Brix, M.R., Adriasola, A.C. and Littke, R., 2005. New information on the thermal history of the southwestern Lower Saxony Basin, northern Germany, based on fission track analysis. International Journal of Earth Sciences, 94(5-6): 876-896.

Senglaub, Y., Littke, R. and Brix, M.R., 2006. Numerical modelling of burial and temperature history as an approach for an alternative interpretation of the Bramsche anomaly, Lower Saxony Basin. International Journal of Earth Sciences, 95(2): 204-224.

Setzmann, U. and Wagner, W., 1991. A new equation of state and tables of thermodynamic properties for methane covering the range from the melting line to 625 K at pressures up to 1000 MPa. Journal of Physical and Chemical Reference Data, 20: 1061-1155.

Shi, J.-Q., Durucan, S., 2003a. Gas Storage and Flow in Coalbed Reservoirs: Implementation of a Bidisperse Pore Model for Gas Diffusion in Coal Matrix. SPE, 84342: 1-9.

Shi, J.Q.D., S., 2003b. A bidispere pre diffusion model for methane displacement desorption in coal by CO2 injection. Fuel, 82: 1219-1229.

Shieh, J.-J., Chunh, T.S., 1999. Gas permeability, diffusivity, and solubility of Poly (4-Vinylpyridine) film. Journal of Polymer Science, 37: 2851-2861.

Siemons, N., Bruining, H., Castelijns, H. and Wolf, K.-H., 2006. Pressure dependence of the contact angle in a CO2-H2O-coal system. Journal of Colloid and Interface Science, 297(2): 755-761.

Siemons, N. and Busch, A., 2007. Measurement and interpretation of supercritical CO2 sorption on various coals. International Journal of Coal Geology, 69(4): 229-242.

Siemons, N., Wolf, K.-H.A.A. and Bruining, J., 2007. Interpretation of carbon dioxide diffusion behavior in coals. International Journal of Coal Geology, 72(3-4): 315-324.

Sircar, S., 1992. Estimation of isosteric heats of adsorption of single gas and multicomponent gas mixtures. Ind. Eng. Chem. Res., 31: 1813-1819.

Sircar, S., 1999. Gibbsian Surface Excess for Gas Adsorptions Revisited. Ind. Eng. Chem. Res., 38: 3670-3682.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

268 | P a g e

Smith, D.M. and Williams, F.L., 1984. Diffusion models for gas production from coal: Determination of diffusion parameters. Fuel, 63(2): 256-261.

Smith, D.M.W., F.L., 1984a. Diffusion models for gas production from coals-Application to Methane content determination. Fuel, 63: 251-255.

Smith, D.M.W., F.L., 1984b. Diffusional effects in the recovery of methane from coalbeds. SPE, 10821: 529-541.

Span, R. and Wagner, W., 1996. A new equation of state for carbon dioxide covering the fluid region from the triple-point temperature to 1100 K at pressures up to 800 MPa. Journal of Physical and Chemical Reference Data, 25(6): 1509-1596.

Spears AD, C.B., 2002. The composition of size-fractionated pulverised coal and the trace element associations. Fuel, 81: 683-690.

Stach, E. et al., 1982. Stach’s Textbook of Coal Petrology, 3rd edition. Gebruder Borntraeger, Berlin.

Stoeckli, H.F., 1990. Microporous carbons and their characterization: The present state of the art. Carbon, 28(1): 1-6.

Suárez, N., Laredo, E. and Nava, R., 1993. Characterization of four hydrophilic sites in bituminous coal by ionic thermal current measurements. Fuel, 72(1): 13-18.

Sudibandriyo, M., Fitzgerald, J.E., Pan, Z., Robinson, Jr., R.L. & Gasem, K.A.M., 2002. Measurements and modeling of gas adsorption on selected coals. Langmuir, 19: 5323-5331.

Sudibandriyo, M., Pan, Z., Fitzgerald, J.E., Robinson Jr., R.L. and Gasem, K.A.M., 2003. Adsorption of Methane, Nitrogen, Carbon Dioxide, and Their Binary Mixtures on Dry Activated Carbon at 318.2 K and Pressures up to 13.6 MPa. Langmuir, 19: 5323-5331.

Suess, E. and Whiticar, M.J., 1989. Methane-derived CO2 in pore fluids expelled from the Oregon subduction zone. Palaeogeography, Palaeoclimatology, Palaeoecology, 71(1-2): 119-136.

Suuberg, E.M., Otake, Y, Yun, Y & Deevi, S.C., 1993. Role of moisture in coal structure and the effect of drying upon the accessibility of coal structure. Energy & Fuels, 7: 384-392.

Svábová, M., Weishauptová, Z. and Pribyl, O., Water vapour adsorption on coal. Fuel, 90(5): 1892-1899.

Svábová, M., Weishauptová, Z. and Pribyl, O., 2012. The effect of moisture on the sorption process of CO2 on coal. Fuel, 92(1): 187-196.

Sweeney, J.J. and Burnham, A.K., 1990. Evaluation of a simple-model of vitrinite reflectance based on chemical-kinetics. AAPG Bulletin - American Association of Petroleum Geologists, 74(10): 1559-1570.

Tang, Y., Jenden, P.D., Nigrini, A. and Teerman, S.C., 1996. Modeling Early Methane Generation in Coal. Energy & Fuels, 10(3): 659-671.

Taraba, B., 2011. Flow calorimetric insight to competitive sorption of carbon dioxide and methane on coal. Thermochimica Acta, 523(1-2): 250-252.

Tenney, C.M. and Lastoskie, C.M., 2006. Molecular simulation of carbon dioxide adsorption in chemically and structurally heterogeneous porous carbons. Environmental Progress, 25(4): 343-354.

Ulberg, D.E. and Gubbins, K.E., 1995. Water adsorption in microporous graphitic carbons. Molecular Physics: An International Journal at the Interface Between Chemistry and Physics, 84(6): 1139 - 1153.

Unsworth, J.F., Fowler, C.S. and Jones, L.F., 1989. Moisture in coal - 2. Maceral effects on pore structure. Fuel, 68: 18-26.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

269 | P a g e

van Bergen, F., Pagnier, H., Krzystolik, P., 2006. Field experiment of enhanced coalbed methane-CO2 in the upper Silesian basin of Poland. Environmental Geoscience, 13: 201–224.

van Bergen, F., Spiers, C., Floor, G. and Bots, P., 2009. Strain development in unconfined coals exposed to CO2, CH4 and Ar: Effect of moisture. International Journal of Coal Geology, 77(1-2): 43-53.

van Hemert, P., Bruining, H., Rudolph, E.S.J., Wolf, K.-H.A.A. and Maas, J.G., 2009. Improved manometric setup for the accurate determination of supercritical carbon dioxide sorption. Review of Scientific Instruments, 80(3): 035103-11.

van Hemert, P., Rudolph-Floter, S., Wolf, K.-H., A. A. and Bruining, J., 2010. Estimate of Equation of State Uncertainty for Manometric Sorption Experiments: Case Study With Helium and Carbon Dioxide. SPE Journal, 15(1): pp. 146-151.

van Hemert, P., Wolf, K.-H.A.A. and Bruining, J., 2007. The Intrinsic Reliability of Manometric Sorption Apparatus Using Supercritical Carbon Dioxide, SPE Annual Technical Conference and Exhibition, Anaheim, CA, pp. 11.

van Krevelen, D.W., 1993. Coal: Typology-Physics-Chemistry-Constitution. Elsevier, 1002 pp.

Vandamme, M., Brochard, L., Lecampion, B. and Coussy, O., 2010. Adsorption and strain: The CO2-induced swelling of coal. Journal of the Mechanics and Physics of Solids, 58(10): 1489-1505.

Wagner, W. and Pruss, A., 1993. International Equations for the Saturation Properties of Ordinary Water Substance. Revised According to the International Temperature Scale of 1990. Addendum to J. Phys. Chem. Ref. Data, 893 (1987). Journal of Physical and Chemical Reference Data, 22(3): 783-787.

Walker Jr., P.L., Mahajan, O.P., 1993. Pore structure in coals. Energy and Fuels, 7: 559-560.

Walker Jr., P.L., Verma, S.K., Rivera-Utrilla, J., Davis, A., 1988. Densities, porosities and surface areas of coal macerals as measured by their interaction with gases, vapours and liquids. Fuel, 67: 1615-1623.

Walpes, D.W., 1980. Time and temperature in petroleum formation: application of Lopatin´s methode to petroleum exploration. AAPG Bulletin, 64: 916-926.

Waples, D.W., Kamata, H. and Suizu, M., 1992a. The art of maturity modeling: 1. Finding a satisfactory geologic model. AAPG Bulletin - American Association of Petroleum Geologists, 76(1): 31-46.

Waples, D.W., Suizu, M. and Kamata, H., 1992b. The art of maturity modeling: 2. Alternative models and sensitivity analysis. AAPG Bulletin - American Association of Petroleum Geologists, 76(1): 47-66.

Weast, R.C., 1972. Handbook of Chemistry and Physics (53rd Edn.). Chemical Rubber Co., Cleveland.

Weishauptová, Z. and Sýkorová, I., 2011. Dependence of carbon dioxide sorption on the petrographic composition of bituminous coals from the Czech part of the Upper Silesian Basin, Czech Republic. Fuel, 90(1): 312-323.

Weniger, P., Kalkreuth, W., Busch, A. and Krooss, B.M., 2010a. High-pressure methane and carbon dioxide sorption on coal and shale samples from the Paraná Basin, Brazil. International Journal of Coal Geology, 84: 190-205.

Weniger, P., Kalkreuth, W., Busch, A. and Krooss, B.M., 2010b. High-pressure methane and carbon dioxide sorption on coal and shale samples from the Paraná Basin, Brazil. International Journal of Coal Geology.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

270 | P a g e

White, C.M. et al., 2005. Sequestration of Carbon Dioxide in Coal with Enhanced Coalbed Methane Recoverys A Review. Energy and Fuel, 19(3): 659-724.

Wollenweber, J. et al., 2010. Experimental investigation of the CO2 sealing efficiency of caprocks. International Journal of Greenhouse Gas Control, 4(2): 231-241.

Wong, S. et al., 2007. Enhanced coalbed methane and CO2 storage in anthracitic coals--Micro-pilot test at South Qinshui, Shanxi, China. International Journal of Greenhouse Gas Control, 1(2): 215-222.

Xia, X., Litvinov, S. and Muhler, M., 2006. Consistent Approach to Adsorption Thermodynamics on Heterogeneous Surfaces Using Different Empirical Energy Distribution Models. Langmuir, 22(19): 8063-8070.

Yamaguchi, S., Ohga, K., Fujioka, M., Muto, S., 2004. Prospect of CO2 sequestration in the Ishikara coal field, Japan, 7th International Conference on Greenhouse Gas Control Technology, Vancouver, Canada.

Yang, R.T. and Saunders, J.T., 1985. Adsorption of gases on coals and heattreated coals at elevated temperature and pressure: 1. Adsorption from hydrogen and methane as single gases. Fuel, 64(5): 616-620.

Yee, D., Seidle, J.P. and Hanson, W.B., 1993. Gas Sorption on Coal and Measurements of Gas Content. In: B.E. Law, Rice, D.D (Editor), Hydrocarbons from coal. AAPG Studies in Geology, pp. 203-218.

Yi, J., Akkutlu, I.Y. and Deutsch, C.V., 2008. Gas Transport in Bidisperse Coal Particles: Investigation for an Effective Diffusion Coefficient in Coalbeds. Journal of Canadian Petroleum Technology, 47(10): 20-26.

Yu, H., Guo, W., Cheng, J. and Hu, Q., 2008a. Impact of experimental parameters for manometric equipment on CO2 isotherms measured: Comment on "Inter-laboratory comparison II: CO2 isotherms measured on moisture-equilibrated Argonne premium coals at 55°C and up to 15 MPa" by Goodman et al. (2007). International Journal of Coal Geology, 74(3-4): 250-258.

Yu, H., Yuan, J., Guo, W., Cheng, J. and Hu, Q., 2008b. A preliminary laboratory experiment on coalbed methane displacement with carbon dioxide injection. International Journal of Coal Geology, 73(2): 156-166.

Zhang, T., Ellis, G.S., Ruppel, S.C., Milliken, K. and Yang, R., 2012. Effect of organic-matter type and thermal maturity on methane adsorption in shale-gas systems. Organic Geochemistry, 47(0): 120-131.

Zhao, D., Feng, Z. and Zhao, Y., 2011. Laboratory experiment on coalbed-methane desorption influenced by water injection and temperature. Journal of Canadian Petroleum Technology, 50(7-8): 24-33.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

271 | P a g e

Main conclusions and discussion Based on the results of an international inter-laboratory study the methodology for the

determination of high-pressure CO2 sorption isotherms has been improved significantly.

The experimental results of both, manometric (TU-Delft and RWTH Aachen University)

and gravimetric (INERIS and University of

Mons) procedures were found to be in very

good agreement (see Figure). The following

factors are considered critical in minimizing

the uncertainties in the measurement of CO2

excess sorption isotherms on natural coals:

1) In the manometric method the ratio of the

measuring cell void volume (Vm) and the

reference cell volume (Vr) must be optimized: A larger Vm/Vr ratio results in smaller

errors but also in a larger number of small pressure steps to reach the final pressure.

Since errors for each pressure step add up, the total number of pressure steps should

not exceed a value of 20. Calculations show that for CO2 isotherms, a Vm/Vr ratio

between 1 and 10 is a good compromise in terms of pressure-step size and minimizing

error.

2) The sample cell should contain a large amount of sample and a small void volume.

This combination ensures that the amount of

competitively sorbing impurities (e.g. water)

in the gas phase of the sample cell is

minimized in comparison to the sorption

capacity. Furthermore, for measurements on

samples with defined moisture content the

risk of changes in moisture content during the

experiment is greatly reduced. In addition, the

ratio between the initial reference cell pressure and the pressure drop due to sorption

should be as large as possible.

3) When using the manometric procedure for CO2 sorption measuremnts, the pressure

range between 7 and 13 MPa should be avoided while pressurising the reference cell

(see Figure on the right). The potential error due to the strong density increase of CO2 in

this pressure range may affect the mass balance calculations substantially. Furthermore,

even small temperature fluctuations will result in large density variations.

4) The reference cell volume and the void volume of the sample cell should be

determined with great care and accurayc, because even small errors in theses values

have a large impact on the mass balance calculation. It is suggested to perform multiple

series of helium expansion runs and minimize the statistical error by averaging.

0

1

2

3

4

5

6

7

8

9

0 2 4 6 8 10 12 14 16

Pressure / (MPa)

Ex

ce

ss

so

rpti

on

/ (

mm

ol/

g)

FP Mons-1

FP Mons-2

RWTH-1

RWTH-2

TU Delft-1

TU Delft-2

INERIS

Best fit

0

100

200

300

400

500

600

700

800

900

1000

0 5 10 15 20

Pressure / (MPa)

Den

sit

y /

(k

g/m

³)

0

5

10

15

20

25

30

35

Den

sit

y d

iffe

ren

ces

/ (

kg

/m³)

CO2 with residual He

pure CO2

density differences

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

272 | P a g e

5) Temperature measurements should be

performed using high-precision resistance

temperature detectors (RTD). Using

thermocouples may result in errors of up to 1K

associated with the cold junction

compensation. Temperatures should be

measured directly within or close to the sample

and reference cells.

The experimental results obtained in this project are used to arrived at an improved

quantitative understanding of sorption processes in coalbed methane reservoirs as well

as the stimulation potential of CO2. The main parameters controlling ultimate gas

recovery from coal deposits are the natural cleat permeability and the total gas content.

Sorption capacity denotes the maximum amount of gas that can be stored by physical

sorption in coal seams under given subsurface conditions in excess of the amount of free

gas that (theoretically) can be accommodated in the pore space. Sorption capacity is

controlled by pressure and temperature, the current moisture content, the composition

of the organic material, the mineral content, and coal rank. With increasing coal maturity

(from lignite to anthracite) we find a linear increase in sorption capacity for moisture-

equilibrated and a more parabolic trend with a minimum around 1% for CH4 and 1.5%

for CO2 vitrinite reflectance for dry coal samples. The difference in CH4 and CO2 sorption

capacity as a function of moisture content on coal of different maturity has been

investigated in some detail in this study.

Based on these findings and the experimental evidence, a conceptual model has been

developed to explain CO2 and CH4 sorption on coals of different maturity levels (rank).

Low-rank, sub-bituminous coal has high CO2 and CH4 sorption capacities. This

observation is attributed to the presence of oxygen-containing functional groups.

Oxygen-containing functional groups account for selective sorption of (polar and

polarizable) gases and water to coals. In moist coals these oxygen-containing functional

groups are preferentially occupied by water due to its polarity.

0

100

200

300

400

500

600

700

800

900

0 5 10 15 20

Pressure / (MPa)

Den

sit

y /

(k

g/m

³)

0

5

10

15

20

25

30

De

nsit

y d

iffe

ren

ce

s /

(k

g/m

³)

318.6 K free gas density

319.6 K free gas density

density differences

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

273 | P a g e

For CO2 and CH4, which do not have a permanent dipole moment, affinities and

occupancy are expected to be lower. In consequence, lower adsorbed phase densities

and lower capacity are observed for low-rank coals when water is present. The higher

CO2/CH4 ratios at low surface coverage observed in this study, can be explained by a

stronger CO2 ordering as compared to CH4 and a 2.2 times higher site density of oxygen-

containing functional groups as preferential sorption sites for CO2 (Heuchel, 1999). Due

to this CO2 ordering by oxygen-containing functional groups, lower CO2 partial pressures

are required to reach a certain level of occupancy. Due to this ordering and the sorbed

water on the oxygen-containing functional groups, the high-energetic sorption sites are

occupied and therefore higher Langmuir pressures pL are observed for CO2 and CH4

when water is present. The dominance of functional groups to control the sorption

processes on low rank coals is caused by the higher average pore diameters (no pore

wall – wall interaction) for low rank

coals. On a non-polar “carbon related”

sorption sites (such as on anthracite

with a fixed carbon content of 88,2%)

only a low degree of competition

between sorbed water and CO2 or CH4

is observed.

During coalification, associated with

decarboxylation and dehydration, the

surface concentration of oxygen-

containing functional groups decreases (van Krevelen, 1993). This is considered the

cause for the observed minimum in the sorption capacity trend at intermediate coal

rank. Progressive coalification generates new micro- and ultra micropores (Prinz and

Littke, 2005b; Sakurovs et al., 2012). Due to overlapping Lenard Jones potentials in the

ultra micropores (carbon pore wall – wall interactions) an increase in sorption capacity

can be observed irrespective of the presence of water. Therefore the sorption capacity

differences between dry and moist samples at low surface coverage diminish and are

much smaller for the anthracite coal. Furthermore, due to the overlapping Lennard-

Jones potentials in the micropores, lower partial pressures are required to completely

fill these pores. In consequence, Langmuir pressures for CH4 and CO2 are lower for dry

high rank coals such as anthracites.

0

2

4

6

8

10

12

0 0.5 1 1.5 2 2.5 3 3.5

CH

4, C

O2

Lan

gmu

ir p

ress

ure

(MP

a)

Vitrinite reflectance (%)

CH4 - dry

CH4 - moist

CO2 - dry

CO2 - moist

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

274 | P a g e

Based on the observation that at high pressures (equivalent to high surface coverage)

the difference between the samples of different maturity reduces, it is concluded that the

carbon-CO2 and carbon-CH4 interaction dominates the growth of sorption layers. This is

in line with molecular simulations. Restrictions in translation and vibrations of the

sorbed CH4 and CO2 molecules, induced by adsorbed water molecules, cause differences

in sorption behaviour. The observation (see Figure on the right) that type I kerogen (low

O/C atomic ratio) is characterised by low adsorption energy and entropy values is in

agreement with the generally low amount of oxygen-containing functional groups. In

contrast, type III kerogen (high O/C atomic ratio) is dominated by highly energetic

localised sorption sites for CH4 molecules. It can therefore be concluded that the oxygen

containing functional groups are the primary adsorption sites for the different kerogen

types. However, the role of aromatic structures within different kerogen types as well as

the aromatic structures, formed during the coalification process is still unclear. These

observations are commonly attributed to pore size distributions.

The isosteric heat of

adsorption is an important

factor in describing the

temperature-dependence of

sorption and the evolution of

gas saturation based on burial

and uplift history of a CBM

deposit. Measurements

performed on dry coals and at

different water contents reveal

that even very low water

contents lead to a strong reduction of CH4 sorption capacity. A decrease in sorption

capacity by up to 50% has been measured for high volatile bituminous coal.

Applying this observation in a generic CBM case study, the evaluation of the measured

isotherms reveals that for coals of a sub-

bituminous A rank, which is common in

many producing CBM reservoirs, and

pressure and temperature gradients typical

for most sedimentary basins, the sorption

capacity of coals decreases only slightly

during uplift. The geothermal gradient is

much more important than the hydrostatic

pressure gradient for the CH4 sorption

capacity prediction as a function of depth. For CBM as well as CO2-ECBM activities a low

geothermal gradient is favourable. (Figure A shows the calculated saturation for

2K/100m, whereas figure B shows the calculated saturation for 3K/100m).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

275 | P a g e

Low-rank coal deposits require a lower pressure draw-down than high rank coal (lower

Langmuir pressure) deposits to desorb an equivalent amount of methane.

In high-rank coal seams it is more likely to observe undersaturated gas contents than in

low-rank deposits.

Based on the results of the study in Chapter 5, the frequently observed undersaturation

of coal seams in the field (with thermal gradients not higher than 3K/100m) is more

likely related to loss of methane from the coal seam to other formations than to changes

in sorption capacity, especially for high volatile bituminous coals. Geothermal gradients

in excess of 3K per 100m or high heat flow events in burial history could also cause

undersaturation. Discovering an oversaturated coal seam is not likely; but if so, it can

only be explained by the decrease in sorption capacity at shallow depth (below 1000 m),

or due to additional gas generation, e.g. by microbial activities. However, this concept is

a static approach, which did not consider heat flow changes during the burial uplift.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

276 | P a g e

Outlook To investigate the potential and implications of CO2-ECBM several experimental

investigations remain to be performed:

Matrix CO2/CH4 counter-diffusion.

The timing of gas exchange in the coal matrix for CO2-ECBM is determined by CO2

sorption rates and CH4 desorption rates. These processes occur simultaneously,

and very limited information is available to estimate the significance of counter-

diffusion for CO2-ECBM.

Gas sorption under reservoir stress conditions.

Conventional sorption experiments are usually conducted without applying

external stress to the coal sample. Thus subsurface conditions are only realized in

terms of temperature and gas pressure. Experimental effective stress acting on

the coal sample changes sorption properties both in terms of sorption capacity

and in terms of equilibration time (sorption kinetics).

Gas sorption from binary mixtures.

In the literature sorption of CO2/CH4 mixtures on coal is still largely computed

from individual sorption isotherms of each gas and the extended Langmuir

function. This approach does not adequately describe the competitive sorption of

two different gases but is only a weak approximation. Several authors have

pointed out that isotherms measured with gas mixtures are likely to provide

different results.

CH4 permeability as a function of effective stress.

Numerous models and approaches have been proposed to describe the changes

in coal permeability during gas recovery as a function of effective stress. The

number of reliable published experimental data sets, however, is very limited.

Future studies should aim at expanding the amount of published experimental

data to test the validity of existing models.

Rock-mechanical implications during the CO2 sorption on organic material.

Some experimental observations indicate mechanical softening when CO2 is

absorbed by organic material.

Coupling of swelling and absorption.

Investigating the causes for the linear coupling between absorbed amount of CO2

or CH4 and the frequently observed volumetric effects and the influence of

moisture.

Improvement of the accuracy of CO2 equations of state

Capillary processes in coal seams with regard to CBM and CO2-ECBM

Capillary processes for CH4, CO2 and water, such as threshold pressure etc. have

to be investigated

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

277 | P a g e

Permeability change by CO2-swelling induced stress

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

278 | P a g e

Content: Figure and Table list Figures

Figure 1: Illustration showing coal matrix blocks and cleat system of a coal.

Figure 2. Schematic setup for manometric sorption devices. V denotes valves and P denotes pressure

transducers.

Figure 3: Schematic diagram of volumetric gas sorption device. Modified after (Sudibandriyo et al., 2003).

Figure 4: Suspension magnetic balance. EM=electro magnet; PM=permanent magnet; TS=titanium sinker.

Adapted from (Charrière et al., 2010).

Figure 5. Gravimetric sorption device used at CSIRO Energy Technlology, Newcastle, AUS. RC and SC are

reference and sample cells respectively. Adapted from (Day et al., 2005).

Figure 6: Error in calculating CH4 and CO2 sorption isotherm for a temperature increase or decrease by 0.1 K.

Carboniferous coal from Silesia mine (seam 315) of the Upper Silesian Basin, Poland, measured in the

dry state using a manometric method.

Figure 7. Comparison of compressibility factors (Z) for CO2 and CH4 at 318 K and pressures up to 30 MPa.

calculated with different equations of state (EOS): Soave-Redlich-Kwong (SRK), Peng-Robinson (PR) and

(Span and Wagner, 1996) (SpW, CO2) and (Setzmann and Wagner, 1991) (SeW, CH4). Dashed lines

indicate the largest difference between the respective EOS for CO2 (between SpW and SRK; short

dashed line) and CH4 (between SW and PR; long dashed line)

Figure 8: Error in calculating CO2 sorption isotherm for an uncertainty in reference and sample cell volume of

0.5 %. Silesia 315 coal, Upper Silesian Basin, Poland, measured in the dry state.

Figure 9: Error in calculating CH4 sorption isotherm for an uncertainty in reference and sample cell volume of

0.5 %. Silesia 315 coal, Upper Silesian Basin, Poland, measured in the dry state.

Figure 10: Plot of pure CO2 versus CO2+H2O and CH4+H2O densities calculated at 45°C and water vapor

pressures from (Wagner and Pruss, 1993). Density calculations have been performed following the

equation of state by (Kunz et al., 2007).

Figure 11. Schematic diagram of sorbate density in Gibbsian sorbed phase (simplified and modified after

(Sircar, 1999)).

Figure 12: Adsorption of water vapor on oxygenated carbons: (I) heated in vacuum at 200°C; (II) in vacuum at

950°C; (III) in vacuum at 1000°C; (IV) at 1100°C in a hydrogen stream; (V) in hydrogen at 1150°C; (VI) in

hydrogen at 1700°C; and (VII) at 3200°C ((Muller et al., 1996)).

Figure 13: Equilibrium moisture content of coals of varying rank ((Beamish, 1998; Bratek, 2002; Busch, 2007;

Clarkson and Bustin, 2000; Crosdale et al., 1998; Day et al., 2008c; Day et al., 2005; Gasem, 2002a;

Hildenbrand et al., 2006a; Laxminarayana and Crosdale, 1999a; Laxminarayana and Crosdale, 2002;

Mastalerz et al., 2004; Moore and Crosdale, 2006; Ottiger et al., 2006; Ozdemir and Schroeder, 2009;

Pan and Connell, 2007; Prinz et al., 2004; Saghafi et al., 2007)).

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

279 | P a g e

Figure 14: Net heat of water sorption on coal as a function of moisture content ((Day et al., 2008c)). The net

heat of sorption Qnet is equal to the isosteric heat of sorption Qst and subtracting the heat of

condensation. This relationship is only valid for inert sorbents.

Figure 15: Schematic of water uptake on coal.

Figure 16: Sorption sites of each gas at one given fixed surface coverage (fixed p,T) and the intersection for

multi-component sorption isotherms.

Figure 17: Schematic diagram of sorption capacity of coals at different moisture contents and as a function of

temperature.

Figure 18: Sorption capacity at reference pressure of 5 MPa as a function of coal rank for moist and dry coals

((Bae and Bhatia, 2006; Battistutta et al., 2010; Busch et al., 2003b; Bustin, 2004; Chaback, 1996; Day et

al., 2008c; Fitzgerald et al., 2005; Harpalani et al., 2006; Kelemen and Kwiatek, 2009; Li et al., 2010;

Mastalerz et al., 2004; Pini et al., 2010; Reeves, 2005; Ryan, 2002; Sakurovs, 2007; Weniger et al.,

2010a; Yu et al., 2008b)).

Figure 19: Schematic plot of the sorption capacity in the dry and moisturized state as a function of

temperature.

Figure 20: Relationship of different analytical bases to various coal components

Figure 21: Sorption capacity of CH4 and CO2 as function of total organic carbon (TOC) (Weniger et al., 2010a).

Figure 22: CH4 and CO2 sorption capacity of dry coals and the influence of rank. The two lines show the

second-order polynomial fit for CH4 and CO2. The regression factor for methane is in an acceptable

range, however for CO2 it is not. Because of the lack of sorption data of coals at higher coal rank, the

reliability of Figure 22 and Figure 23 has to be proven. So far, the crossover of the trend lines in these

figures has no significant meaning.

Figure 23: CH4 and CO2 sorption capacity on moisture-equilibrated and as received coals as a function of

rank. The two lines show the linear fit for CH4 and CO2. The regression factor for CH4 is in an acceptable

range; however for CO2 the trend is weaker and dominated by the data point at high vitrinite

reflectance.

Figure 24: CO2/CH4 sorption ratio for different moist and dry coal sample at various temperatures as a

function of coal rank. Values were picked between 1 and 5 MPa. Data fit for moist samples only (Bae

and Bhatia, 2006; Battistutta et al., 2010; Busch et al., 2003b; Bustin, 2004; Chaback, 1996; Day et al.,

2008c; Fitzgerald et al., 2005; Harpalani et al., 2006; Kelemen and Kwiatek, 2009; Li et al., 2010;

Mastalerz et al., 2004; Pini et al., 2010; Reeves, 2005; Ryan, 2002; Sakurovs, 2007; Weniger et al.,

2010a; Yu et al., 2008b)

Figure 25: Specific surface areas of meso and macropores as obtained from small angle neutron scattering

(SANS, circles) and low-pressure N2 adsorption experiments (triangles and crosses refer to different

evaluation methods) (after (Prinz et al., 2004).

Figure 26: Specific surface areas for meso/macropores and micro/ultramicropores determined on a suite of

Carboniferous coals from the German Ruhr Basin ((Prinz and Littke, 2005b; Prinz et al., 2004)). Open

symbols represents the mesopore surface area determined with N2 at 77K.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

280 | P a g e

Figure 27: Bidisperse pore model as defined by (Ruckenstein, 1971).

Figure 28: a)The N2 isotherm at 77K (80 data points, 7x10-6

to 0.9965 p/p0) of the F400 in a log plot. b) pore

diameter distribution over the complete range based on NLDFT and c) the important range between 0-

5 nm. d) SEM picture including EDX-Analysis (see table1)

Figure 29: Excess CO2 sorption isotherms on activated carbon Filtrasorb F400 plotted on a linear (a) and a

logarithmic (b) pressure scale.

Figure 30: Excess sorption versus density of free gas phase (using the Span & Wagner equation) for activated

carbon Filtrasorb F400 at 318 K

Figure 31: Experimental excess sorption isotherms (318 K) for Filtrasorb F400 activated carbon measured in

this study with best fit of excess sorption function according to equation (2).

Figure 32: Normalized deviation of experimental CO2 excess sorption isotherms (318 K) for F400 from the

best fit vs. pressure.

Figure 33: Comparison of CO2 excess sorption results for F400 activated carbon of the present study (dashed

line) with literature data.

Figure 34: Comparison of previously published excess sorption isotherms for CO2 on F400 activated carbon

with two measurements performed at RWTH Aachen on samples exposed briefly to air after activation.

Figure 35: CO2 excess sorption isotherms on dry Velenje coal plotted on a linear (a) and a logarithmic (b)

pressure scale.

Figure 36: CO2 excess sorption (dry basis) versus density of free gas phase for lignite sample from the Velenje

mine.

Figure 37: CO2 excess sorption isotherms (dry basis) on Brzeszcze 364 LW plotted on a linear (a) and a

logarithmic (b) pressure scale.

Figure 38: CO2 excess sorption isotherms (dry basis) versus density of free gas phase for bituminous coal

sample Brzeszcze 364.

Figure 39: CO2 excess sorption isotherms (dry basis) for the Selar Cornish semi-anthracite sample plotted on a

linear (a) and a logarithmic (b) pressure scale.

Figure 40: Excess sorption versus density of free gas phase for Selar Cornish semi-anthracite sample.

Figure 41: Master isotherms of the a) Velenje b) Brzeszcze and c) Selar Cornish coals

Figure 42: Normalized deviations of experimental CO2 excess sorption isotherms (318 K) for a) Velenje b)

Breszcze 364 c) Selar Cornish from the best fit vs. pressure

Figure 43: a) Deviation between the CO2 isotherms at 3 and 13 MPa of the different laboratories as a function

of the volatile matter of the samples b) Langmuir pressure K,L and maximum excess sorption capacity

Smax as a function of volatile matter; error bars reflect the standard deviations between the fitting

parameters

Figure 44: a) Pure CO2 density for 318K and 319K. b) Density of pure CO2 and a mixture of CO2 with a residual

gas content of 1 bar He at for 318 K. Equations of State were calculated using the “Software for the

Reference Equation of State for Natural Gases (GERG-2004)” of Ruhr Universität Bochum (Kunz et al.,

2007)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

281 | P a g e

Figure 45: Potential effects of experimental errors on CO2 excess sorption isotherms: a) gas impurities (Kunz

et al., 2007) and uncertainties in temperature measurements; b) errors in reference and void volume

determination

Figure 46: Methane (a) and carbon dioxide (b) sorption capacity at reference pressure of 5 MPa as a function

of coal rank for moist and dry coals. Additional isotherm data are used from previous studies(Busch

and Gensterblum, 2011) 1. (c) CO2/CH4 sorption ratio for different moist and dry coal samples at various

temperatures as a function of coal rank. Values were selected between 1 and 5 MPa. Isotherm data are

taken from Ref. (Busch and Gensterblum, 2011)1. Ratios from this study are taken at 318K on dry and

moist coal samples at 1 MPa (small symbols) and 5 MPa (large symbols).

Figure 47: The difference of CH4 and CO2 Langmuir volume n between dry and moist coals as a function of

oxygen content. (a) shows the absolute and (b) the relative reduction in sorption capacity due to the

presence of water. This indicates that the amount of oxygen is proportional to the reduction in

sorption capacity due to sorbed water. In addition, the reduction is, in relation to the total amount of

sorption sites, the some and independent of the type of gas. This indicates that the reduction in

sorption capacity is only caused by the water-occupied oxygen-containing functional groups.

Figure 48: Langmuir sorption capacity and Langmuir pressure as a function of maturity (vitrinite reflectance).

The Langmuir sorption capacity is calculated according to equation 40.

Figure 49: Isosteric heats of adsorption for CO2 and CH4 at low (part A) and high surface coverage (part B) on

dry and moist coals and different Kerogen types as a function of entropy. Data from this study and Ref.

(Li et al., 2010; Zhang et al.). Figure part A) Surface coverage up to 30% Figure part B) and C) Surface

coverage up to 70%. The dashed line indicates the translation entropy change from the free gas 3D on

the 2D surface (45 and 49 J mol-1

K-1

for CH4 and CO2 respectively). In part B) to illustrate the

thermodynamic parameter of the different coal rank we use increasing grayscale for increasing rank.

Figure 50: FTIR-spectra of subbituminous (light grey), high-volatile bituminous (dark grey) and anthracite

(black) coals. The characteristic wave numbers for the hydroxyl groups are at 3360 cm-1

(left circle), for

the carbonyl-groups 1800-1600 cm-1

,and the carboxyl groups and ketones at 1700cm–1

(right circle). It is

obvious that the subbituminous coal has the highest concentration of functional groups, while the

anthracite has the lowest concentration.

Figure 51: CH4 adsorption isotherms for moisture-equilibrated coals (particle size fraction ~0.5 to 1mm) at

303K, 319K and 334K (1mmol/g =24.5 Std m³/t).The triangles represents isotherms measured on the

subbituminous coal. Circles stands for isotherms measured on high-volatile bituminous coal and

diamonds for isotherms measured on anthracites. The isotherm temperature increases with increasing

filling of the symbols.

Figure 52: CH4 adsorption isotherms for dry coals (particle size fraction ~0.5 to 1mm) at 303K, 318K and 333K

(1mmol/g =24.5 Std m³/t). Triangles represent isotherms measured on the Subbituminous A coal.

Circles are isotherms measured on high-volatile bituminous coal and diamonds isotherms measured on

anthracites. Shading of symbols increases with isotherm temperature.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

282 | P a g e

Figure 53: CO2 excess sorption isotherms for moist coals as a function of gas pressure (particle size fraction

~0.5 to 1mm) at 318K, 334K and 349K (1mmol/g =24.5 Std m³/t). The triangles represents isotherms

measured on the subbituminous coal. Circles stands for isotherms measured on high-volatile

bituminous coal and diamonds for isotherms measured on anthracites. Shading of symbols increases

with isotherm temperature

Figure 54: CO2 excess sorption isotherms for dry coals as a function of gas pressure (particle size fraction ~0.5

to 1mm) at 318K, 334K and 349K (1mmol/g =24.5 Std m³/t). The triangles represents isotherms

measured on the subbituminous coal. Circles stands for isotherms measured on high-volatile

bituminous coal and diamonds for isotherms measured on anthracites. The isotherm temperature

increases with increasing filling of the symbols.

Figure 55: Part A) CH4 (triangle) and CO2 (square) Langmuir pressures as a function of coal maturity on moist

and dry coal samples at 318K (Chapter 5). In part B) the Langmuir pressure ratios of CO2 divided by CH4

(light grey, circles) at dry and for moist coal samples (dark grey, circles) has been calculated. Langmuir

pressure ratios of CH4 and CO2 isotherms calculated of dry and moist coals (CH4, triangles and CO2,

squares) at dry and for moist coal samples.

Figure 56: CH4 (triangle) and CO2 (square) Langmuir sorption amount as a function of coal maturity of moist

and dry coals (Chapter 5) Langmuir capacity moist/dry ratio for CH4 (triangles) and CO2 (squares) at

319K as a function of maturity on moist and dry coal samples. CO2/CH4 Langmuir capacity ratio for dry

coal samples (light grey circles) and for moisturised samples (dark grey circles) at 319K as a function of

maturity.

Figure 57: CH4 and CO2 sorption capacity difference between dry and moist coals as a function of oxygen

content.

Figure 58: Moist/dry excess sorption capacity ratios for CH4 (grey) and CO2 (black).

Figure 59: CO2/CH4 excess sorption capacity ratio at 318K on dry (black) and moist (grey) coal samples as a

function of gas pressure. The ratio is calculated from the fitted excess sorption functions

Figure 60: Absolute CO2/CH4 sorption capacity ratio at 318K on dry (black) and moist (grey) coal samples. The

calculated ratio is based on the fitted excess sorption functions.

Figure 61: Calculated CH4 storage capacity in the gas-saturated porosity of a coal seam (assumed coal density

1.4 g/cm³) using the EOS for methane (Setzmann and Wagner, 1991) and the maximum dissolved CH4

content in the formation water under the assumption of a water-filled pore system. A thermal gradient

of 0.03 K/m and hydrostatic pressure gradient of 0.01MPa/m was used for the calculations (1 mmol/g =

24.5 Std. m³/ton). The porosity was kept constant with depth and we did not take into account the

compaction of the coal during burial.

Figure 62: Schematic illustration of CH4 sorption isotherms, amount of producible gas n and the effect of

CH4 affinity (Langmuir pressure pL is a function sorption enthalpy dH) for two different coal ranks. The

figure illustrates the importance of a Langmuir pressure as an indicator (reciprocal) for the affinity

between adsorbent and adsorbate. Furthermore, the Langmuir pressure is a function of the sorption

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

283 | P a g e

enthalpy (Equation 56 and following equation). The parameter set for this figure is taken from Table 26

and

Figure 63: Schematic illustration of the methane sorption isotherm, the degree of saturation and the amount

of producible gas (modified after (Bustin and Bustin, 2008)). The two dots mark the amount of gas

stored at reservoir pressure. The highlighted pressure levels are the initial reservoir pressure PR, the

initial desorption pressure PD (also called critical desorption pressure (Ayers, 2002)), and the

abandonment pressure PA.

Figure 64: Schematic illustration of the CH4 sorption capacity (at constant n) with depth and influence of

differential enthalpy on the Langmuir pressure caused by a dry (dH=-22kJ mol-1

; ln(K)= 9.6 and n = 0.99

mmol g-1

) and moist sub-bituminous coal (dH=-17.9 kJ mol-1

; ln(K)= 9.09 and n = 0.9 mmol g-1

)

(Gensterblum et al. 2012).

Figure 65: Conceptual design of the evolution of the in-situ gas saturation of coal seams as a function of

maturity and thermal gradient. Gas saturation during uplift is calculated as the ratio of the sorption

capacity at a given present-day depth (with corresponding temperature and fluid pressure) and the

methane sorption capacity at the pressure and temperature conditions at maximum burial depth. The

first step in the algorithm is to compute the maximum coalification temperature from present-day coal

maturity (VRr) (step 1). Assuming a thermal gradient (=(dT/dz) e.g. 0.03 K/m) the maximum burial

depth can be calculated (step 2). Assuming a normal hydrostatic pressure gradient of 0.01 MPa/m the

methane sorption capacity can be calculated according to the pressure and temperature conditions at

maximum burial depth (step 3). At maximum burial depth, the generation rate of thermogenic

methane reaches or has reached its maximum. It is generally assumed that the total amount of gas

generated is 2-4 times the maximum sorption capacity (Hunt and Steele, 1991; Jüntgen and Klein, 1974;

Tang et al., 1996). Following their results CH4 saturation of the coal can be expected to be 100% at

maximum depth (step 4). Finally the uplift starts to the present-day depth (step 5).

Figure 66: CH4 (black) and CO2 (grey) sorption capacities as function of depth and thermal gradient for the

three coal samples investigated here. The sorption capacity as a function of depth is plotted to the

maximum burial depth according to Barker and Pawlewicz (1994). (1mmol/g =24.5 Std. m³/ton).

Figure 67: A) CH4 (black) and CO2 (grey) sorption capacities as function of depth assuming a thermal gradient

of 0.03K/m for sub-bituminous A, high-volatile bituminous and anthracite coal deposits. B) Illustrates

the CH4 sorption capacity in proportion to the compressed gas in 5% matrix porosity. The porosity of

5% is assumed to be constant with depth (no poro-elasticity or compaction is considered) (1mmol/g

=24.5 Std. m³/ton).

Figure 68: CH4 recovery potential as a function of depth. Sorption enthalpies are used from

Figure 69: CO2 to CH4 sorption capacity ratio as a function of depth for the three coal samples studied,

assuming a thermal gradient of 0.03K/m and a hydrostatic pressure gradient of 0.01MPa/m.

Figure 70: Calculated CH4 saturation trends after uplift from the maximum burial depth (burial history)

following the concept illustrated in figure 5. Model A: assuming a lower geothermal gradient of

2K/100m and the corresponding depth of 3700 m for the sub-bituminous, 5800 m for the high-volatile

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

284 | P a g e

bituminous and 10900 m for the anthracite (Barker and Pawlewicz 1994). Model B: assuming a

geothermal gradient of 3K/100m and the corresponding depth 2500 m, 3900 m and 7300 m (Barker and

Pawlewicz 1994) for the three coal samples assuming no loss of CH4 and a hydrostatic pressure gradient

of 1MPa/100m.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

285 | P a g e

Tables

Table 1: Effective water diffusivities on coal ((Charrière and Behra, 2010; McCutcheon et al., 2003)).

Table 2: World-wide coal basins investigated for gas sorption capacity.

Table 3. Sorption kinetic experiments performed by various authors ((Busch et al., 2004; Charrière et al.,

2010; Ciembroniewicz and Marecka, 1993; Clarkson and Bustin, 1999a; Clarkson and Bustin, 1999b; Cui,

2004; Gruszkiewicz et al., 2009; Kelemen and Kwiatek, 2009; Li et al., 2010; Marecka, 1998; Pan et al.,

2010; Pone et al., 2009; Seewald, 1986; Siemons et al., 2007)). h.v.b, m.v.b. and l.v.b are high, medium

and low volatile bituminous coals. Diffusion coefficients derived from unipore or bidisperse models in

chapters 0 and 0.

Table 4: EDX elemental analysis averaged over the total surface

Table 5: F400 Proximate and Ultimate analysis parameters

Table 6: Specifications of experimental devices used in this inter-laboratory study

Table 7: Comparison of experimental results for CO2 excess sorption isotherms on F400 activated carbon.

Table 8: Regression parameters of CO2 excess sorption isotherms on F400 activated carbon obtained as best

3-parameter fits of equation (2) to experimental results of this study. The density of the sorbed phase

could not be determined from the INERIS isotherm because the isotherm was only measured up to 5

MPa.

Table 9: characteristic surface parameters determined low pressure isotherm of N2 at 77K

Table 10: Fit parameters of CO2 excess sorption isotherms on F400 activated carbon from the literature and

RWTH Aachen measurements 3 and 4 (modified drying procedure). Sorbed phase density (sorbed),

maximum absolute sorption capacity nsorbed() and Langmuir coefficient (KL,V) were fitted by regression

of equation (2).

Table 11: Results from ultimate and proximate analyses and petrographic data for the three coals studied

Table 12: Specifications of experimental devices used by different laboratories

Table 13: Comparison of experimental results for CO2 excess sorption isotherms on Velenje coal (according to

equ. 1a and 1b)

Table 14: Comparison of experimental results for CO2 excess sorption isotherms on Brzeszcze 364 coal

sample

Table 15: Comparison of experimental results for CO2 excess sorption isotherms on Selar Cornish anthracite

Table 16: Fitting parameters of the master isotherms in comparison with the average values of the isotherms

from table 13,14,15

Table 17: Maximum CO2 excess sorption values at 318 K and corresponding pressures

Table 18: Intra laboratory reproducibility

Table 19: Ultimate and proximate analysis of coal samples used in this study. For the analyzing carbon,

hydrogen and nitrogen content a Leco CHN-2000 was used and we follow the measurement procedure

DIN 51732. For the sulfur content we use a Leco SC 144 DR and we follow the measurement procedure

DIN 51724.Oxygen was calculated by the difference approach (100%-C%-N%-S%-H%=O%). For the

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

286 | P a g e

determination of the vitrinite reflectance we use 100 points. For the maceral group analysis we count

500 points.

Table 20: Langmuir fitting parameters for CH4 and CO2 excess sorption isotherm on dry and moist sub-

bituminous coal.

Table 21: Langmuir fitting parameters for CH4 and CO2 excess sorption isotherm on dry and moist high-

volatile bituminous coal.

Table 22: Langmuir fitting parameters for CH4 and CO2 excess sorption isotherm on dry and moist anthracite.

Table 23: Overview of methane and carbon dioxide sorption behavior on coals investigated in this study.

Columns 3 through 6 summarize the dependence of moist vs. dry and CO2 vs. CH4 sorption capacities on

surface coverage based on Figs. 58, 60 and 61. The dependence of the Langmuir sorption parameters

(sorption capacity and affinity or Langmuir pressure) on moisture state and gas type is outlined in rows

7 trough 9.

Table 24: Coal samples used in this study

Table 25: Ultimate and proximate analysis of coal samples used in this study (* Equilibrium moisture content

at 97% humidity; + dry basis)

Table 26: Fitted Langmuir parameters for methane excess sorption isotherms on sub-bituminous A, high-

volatile bituminous and an anthracite at three different temperatures (see text) (Gensterblum, 2013) .

Table 27: Fitted Langmuir parameters for CO2 excess sorption isotherms on sub-bituminous A, high- volatile

bituminous and anthracite at three different temperatures (Gensterblum, 2013).

Table 28: CH4 sorption enthalpy for moist coals of different rank based on Langmuir pressure (see Appendix,

Figure A1).

Table 29: Isosteric heats of CO2 sorption on moisturized coals based on Langmuir pressure (see Appendix,

Figure A2).

Table 30: List of symbols

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

287 | P a g e

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

288 | P a g e

The children now love luxury; contempt for authority; they show disrespect for elders

and love chatter in place of exercise. Children are now tyrants, not the servants of their

households. They no longer rise when elders enter the room. They contradict their

parents, chatter before company, gobble up dainties at the table, cross their legs, and

tyrannize their teachers.

Socrates (469–399 B.C.)

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

289 | P a g e

Acknowledgements First of all, I want to thank Prof. Dr. Ralf Littke and Prof. Dr. Guy de Weireld for the

supervision and the useful and extensive discussions about the theoretical aspects of

adsorption process in general and the role of moisture.

In particularly, I would like to thank my mentor and responsible adviser Dr. Bernhard M.

Krooss who has not only motivated me in common talks, he also has woken up my

enthusiasm for the topic. With his tremendous knowledge and his helpful and friendly

contact, which he knew how to pave to me in every situation to show me the way to a

higher level of understanding and the willingness to accompany me on this way.

I would like to acknowledge the Shell funding for this study.

Furthermore, I would like to thank Andreas Busch for organising the funding and his

thematic support. Although his support finding more elegant and precise English

phrases were very much appreciated. I would like to thank him and Bernd for her

trusted friendships.

Many persons supported the experimental work. First of all, my bachelor- and master-

students, Alexej Merkel, Kevin Frank, Christian Hermanni, Hanna Pitzer, Michael

Sartorius, Philipp Storath, Simon Müller and Svetlana Golovin. I would like to thank

especially my student assistant Alexej for his support and high motivation over a long

period of time. Furthermore, the technical support by Stefan Schnorr was very much

appreciated. In addition, I would like to thank Susan Giffin, for her editorial support and

her patience with my English grammar. Finally, I would like to thank all my colleagues

for the nice and friendly working environment.

Meinen Lehrern, Herr Alois Stickelmann, Frau Renate Heinze und Herr Rainer

Wennekamp möchte ich herzlich danken, dass Sie meine Fähigkeiten früh erkannt und

gefördert haben.

Meinen Eltern die mich immer ermutigt und unterstützt haben und ohne die ich es nicht

bis zu einer Promotion geschafft hätte.

Meinen beiden Söhnen Max und Robin, für die einfachste und zugleich schönste

Ablenkung und Entspannung.

Mein besonderer Dank gilt meiner Frau, Michaela. Sie hat mich mit all ihren Kräften bei

dieser Arbeit unterstützt hat.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

290 | P a g e

Wenn der Mensch alles leisten soll,

was man von ihm fordert,

so muß er sich für mehr halten als er ist.

Johann Wolfgang von Goethe (1749 - 1832),

deutscher Dichter der Klassik, Naturwissenschaftler und Staatsmann

Quelle: »Maximen und Reflexionen«,

Nr. 69 nach den Handschriften des Goethe- und Schiller Archivs,

Verlag der Goethe-Gesellschaft, Weimar, 1907

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

291 | P a g e

Curriculum vitae (education)

Education

2009 - 2013 PhD position in “CBM and CO2-ECBM related sorption processes in

coal”

Funded to 100% by Shell Global Solutions International BV,

Rijswijk, The Netherlands

2001 - 2009 Physic studies

Diploma thesis: “Oxygen transport properties of the high

temperature oxygen membrane Ba0.5Sr0.5Cu0.8Fe0.2O3-x for clean

coal power plants”

1998 – 2000 High school degree at Euregio-Kolleg (evening school)

1996 - 1999 Physical-technical assistant education at Research Center Jülich

Grants

2012 Heitfeld travel grant

Neutrondiffraction experiments at the Los Alamos Neutron Science Center

(LANSCE), Los Alamos, USA

2009 DAAD Go8 Australia Travel grant

University of Queensland, Brisbane, Australia

2008 International Office RWTH Aachen University Travel scholarship

Occupational training for 2 month at University of Queensland, Brisbane,

Australia

Research exchange

2008 International Office RWTH Aachen University Travel scholarship

Occupational training for 2 month at University of Queensland, Brisbane,

Australia

2010 DAAD Go8 Australia Travel grant, University of Queensland, Brisbane,

Australia

„New experimental methods for the investigation of gas sorption and

transport processes in coal seams“

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

292 | P a g e

Invited talks

2012 Shell Rijswijk : CO2 -Clay Workshop

“Coupled CO2 sorption and neutron diffraction experiments on clay

minerals”

2011 Arrow Energy Limited, Brisbane, Australia

“CBM related sorption and transport processes in coal”

2011 GFZ Potsdam

“Transport Processes in Coals – Lab. Experiments and Upscaling Issues”

2011 Utrecht University: CO2 -Clay Workshop

“CO2 sorption experiments on clay minerals”

2010 CSIRO Melbourne

“CBM related transport processes in coal”

Scientific voluntary service

Since 2012 Convener of the EGU Session:

“Petrophysics of unconventional hydrocarbon reservoirs”

Since 2010 Peer-reviewer for several international journals:

- International Journal of Coal Geology

- Fuel

- Energy and Fuels

- International Journal of Greenhouse Gas Control

- Industrial & Engineering Chemistry Research

- Journal of Environmental Chemical Engineering

- Marine Petroleum Geology

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

293 | P a g e

Publication list Hirsch index (h-index) is 13 (calculated at the 18th of July 2013using Web of Science)

2014

Gensterblum Y., A. Busch, B.M. Krooss,

Molecular concept and experimental evidence of H2O, CH4 & CO2 sorption on organic

material

Fuel, (2014), 115, pp. 581-588

2013

Gensterblum Y., A. Merkel, A. Busch, B.M. Krooss

High-pressure CH4 and CO2 sorption isotherms as a function of coal maturity and the

influence of moisture

International Journal of Coal Geology, (2013), 118, pp. 45-57

Gensterblum Y., Merkel A., Busch A., Krooss B.M.

Gas saturation and CO2 enhancement potential of coalbed methane reservoirs as a

function of depth

AAPG Bulletin 2013 (in press)

Gasparik, Matus; Ghanizadeh, Amin; Gensterblum, Yves; Krooss, B.M.

"Multi-temperature" method for high-pressure sorption measurements on moist shales

Review of Scientific Instruments 84 (8)

Gasparik, Matus; Ghanizadeh, Amin; Gensterblum, Yves; Weniger, Philipp; Krooss, B.M.

Geological Controls on the Methane Storage Capacity in Organic-Rich Shales

(2013) submitted to International Journal of Coal Geology

Ghanizadeh Amin, M. Gasparik, A. Amann‐Hildenbrand, Gensterblum, Yves, B.M. Krooss

Experimental study of fluid transport processes in the matrix system of the European

organic-rich shales: I. Scandinavian Alum shale

(2013) submitted to Marine and Petroleum Geology (accepted)

Ghanizadeh A., A. Amann-Hildenbrand, M. Gasparik, Gensterblum, Yves, B. M. Krooss, R.

Littke

Experimental study of fluid transport processes in the matrix system of the European

organic-rich shales: II. Posidonia Shale (Lower Toarcian, northern Germany)

International Journal of Coal Geology (2013) (in press) DOI: 10.1016/j.coal.2013.06.009

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

294 | P a g e

Ghanizadeh, M. Gasparik, A. Amann-Hildenbrand, Gensterblum, Yves, B. M. Krooss

Lithological controls on matrix permeability of organic-rich shales: An experimental

study A.

Energy Procedia, (2013), 40, pp. 127 – 136

2012

Gasparik, M., Ghanizadeh, A., Bertier, P., Gensterblum, Y., Bouw, S. & Krooss, B.M.

"High-pressure methane sorption isotherms of black shales from the Netherlands",

Energy and Fuels, (2012), vol. 26, no. 8, pp. 4995-5004.

Hol, S., Gensterblum, Y. and Spiers, C.J.

"Direct determination of total CO2 uptake by coal: A new technique compared with the

manometric method", Fuel, (2013) vol. 105, pp. 192-205.

2011

Busch, Andreas & Gensterblum, Yves,

"CBM and CO2-ECBM related sorption processes in coal: A review",

International Journal of Coal Geology (2011), vol. 87, no. 2, pp. 49-71.

Gensterblum Yves

H2 and CH4 Sorption on Cu-BTC Metal Organic Frameworks at Pressures up to 15 MPa

and temperatures between 273 and 318 K

Journal of Surface Engineered Materials and Advanced Technology, 2011, 1, 23-29

2010

Gensterblum, Yves, van Hemert, P., Billemont, P., Battistutta, E., Busch, A., Krooss, B.M.,

De Weireld, G., Wolf, K.-H.A.A.

“European inter-laboratory comparison of high pressure CO2 sorption isotherms II:

Natural coals”

International Journal of Coal Geology, (2010), 84 (2), pp. 115-124.

Li, D., Liu, Q., Weniger, P., Gensterblum, Y., Busch, A., Krooss, B.M.

“High-pressure sorption isotherms and sorption kinetics of CH4 and CO2 on coals”

Fuel, (2010), 89 (3), pp. 569-580.

2009

Gensterblum, Y., van Hemert, P., Billemont, P., Busch, A., Charriére, D., Li, D., Krooss,

B.M., de Weireld, G., Prinz, D., Wolf, K.-H.A.A.

European inter-laboratory comparison of high pressure CO2 sorption isotherms. I:

Activated carbon

Carbon, (2009) 47 (13), pp. 2958-2969.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

295 | P a g e

2008

Busch, A., Alles, S., Gensterblum, Y., Prinz, D., Dewhurst, D.N., Raven, M.D., Stanjek, H.,

Krooss, B.M.

Carbon dioxide storage potential of shales

(2008) International Journal of Greenhouse Gas Control, 2 (3), pp. 297-308.

2007-2003

Goodman, A.L., Busch, A., Bustin, R.M., Chikatamarla, L., Day, S., Duffy, G.J., Fitzgerald, J.E.,

Gasem, K.A.M., Gensterblum, Y., Hartman, C., Jing, C., Krooss, B.M., Mohammed, S., Pratt,

T., Robinson Jr., R.L., Romanov, V., Sakurovs, R., Schroeder, K., White, C.M.

Inter-laboratory comparison II: CO2 isotherms measured on moisture-equilibrated

Argonne premium coals at 55 °C and up to 15 MPa

(2007) International Journal of Coal Geology, 72 (3-4), pp. 153-164.

Busch, A., Gensterblum, Y., Krooss, B.M.

High-pressure sorption of nitrogen, carbon dioxide, and their mixtures on Argonne

Premium Coals

(2007) Energy and Fuels, 21 (3), pp. 1640-1645.

Busch, A., Gensterblum, Y., Krooss, B.M., Siemons, N.

Investigation of high-pressure selective adsorption/desorption behaviour of CO2 and

CH4 on coals: An experimental study

(2006) International Journal of Coal Geology, 66 (1-2), pp. 53-68.

Krooss, B.M., Friberg, L., Gensterblum, Y., Hollenstein, J., Prinz, D., Littke, R.

Investigation of the pyrolytic liberation of molecular nitrogen from Palaeozoic

sedimentary rocks

(2005) International Journal of Earth Sciences, 94 (5-6), pp. 1023-1038.

Busch, A., Krooss, B.M., Gensterblum, Y.

Carbon dioxide storage in coalbed seams: Adsorption/desorption experiments with

carbon dioxide/methane mixtures within the framework of the EU-RECOPOL project

[CO2-Speicherung in Kohle lo zen: Adsorptions/desorptions-Experimente mit CO2/CH4-

Mischungen im Rahmen des EU-RECOPOL projektes]

(2004) DGMK Tagungsbericht, (2), pp. 473-483.

Busch, A., Gensterblum, Y., Krooss, B.M., Littke, R.

Methane and carbon dioxide adsorption-diffusion experiments on coal: Upscaling and

modelling

(2004) International Journal of Coal Geology, 60 (2-4), pp. 151-168.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

296 | P a g e

Goodman, A.L., Busch, A., Duffy, G.J., Fitzgerald, J.E., Gasem, K.A.M., Gensterblum, Y.,

Krooss, B.M., Levy, J., Ozdemir, E., Pan, Z., Robinson Jr., R.L., Schroeder, K., Sudibandriyo,

M., White, C.M.

An inter-laboratory comparison of CO2 isotherms measured on argonne premium coal

samples

(2004) Energy and Fuels, 18 (4), pp. 1175-1182.

Siemons, N., Busch, A., Bruining, H., Krooss, B., Gensterblum, Y.

Assessing the Kinetics and Capacity of Gas Adsorption in Coals by a Combined

Adsorption/ Diffusion Method

(2003) Proceedings - SPE Annual Technical Conference and Exhibition, pp. 2477-2484.

Busch, A., Gensterblum, Y., Krooss, B.M.

Methane and CO2 sorption and desorption measurements on dry Argonne premium

coals: Pure components and mixtures

(2003) International Journal of Coal Geology, 55 (2-4), pp. 205-224.

Busch, A., Krooss, B.M., Gensterblum, Y., Van Bergen, F., Pagnier, H.J.M.

High-pressure adsorption of methane, carbon dioxide and their mixtures on coals with a

special focus on the preferential sorption behaviour

(2003) Journal of Geochemical Exploration, 78-79, pp. 671-674.

Krooss, B.M., Van Bergen, F., Gensterblum, Y., Siemons, N., Pagnier, H.J.M., David, P.

High-pressure methane and carbon dioxide adsorption on dry and moisture-

equilibrated Pennsylvanian coals

(2002) International Journal of Coal Geology, 51 (2), pp. 69-92.

Gensterblum (2013): CBM and CO2-ECBM related sorption processes in coal

297 | P a g e

Conference talks

2013

Gensterblum Y., A. Merkel, Andreas Busch and Bernhard M. Krooss “CO2 adsorption

isotherm on clay minerals and the CO2 accessibility into the clay interlayer”

General Assembly of the European Geological Union (EGU), Vienna, 2013

2012

Gensterblum Y., A. Merkel, Andreas Busch and Bernhard M. Krooss “A conversion of CO2-

ECBM related lab observations to reservoir requirements” 2012 EGU Vienna,

2012

Gensterblum Y., A. Merkel, Andreas Busch and Bernhard M. Krooss “CO2-ECBM related

coupled physical and mechanical transport processes” 2012 EGU Vienna 2012

Gensterblum Y., A. Merkel, Andreas Busch and Bernhard M. Krooss “CBM and CO2-ECBM

related sorption uptake kinetics and diffusion processes” 1-3th of October 2012

GeoHannover 2012

Gensterblum Y., A. Merkel, Andreas Busch and Bernhard M. Krooss “CO2-ECBM:

Application of laboratory results to reservoirs” 1-3th of October 2012

GeoHannover 2012

Gensterblum Y., A. Merkel, Andreas Busch and Bernhard M. Krooss “CO2-ECBM related

coupled physical and mechanical transport processes” 3-7th December AGU San

Francisco 2012

2011

Gensterblum Y., Bernhard M. Krooss, Duncan Cumming and Andreas Busch,

“Experimental investigation of CBM-related transport processes on Surat Basin

coal” 3rd Asia Pacific Coalbed Methane Symposium 2011, May 3-6, 2011, Brisbane,

Australia

Gensterblum Y., Bernhard M. Krooss, Paul Massarotto, “Results of mechanical, sorption

and transport experiments on a German high volatile bituminous coal” 3rd Asia

Pacific Coalbed Methane Symposium 2011, May 3-6, 2011, Brisbane, Australia

Gensterblum Y., Bernhard M. Krooss, Paul Massarotto, “Results of mechanical, sorption

and transport experiments on a German high volatile bituminous coal“ 03-08.

April 2011 EGU Vienna 2011

2010

Gensterblum Y. Bernhard M. Krooss, Paul Massarotto, “Preliminary results of

mechanical, sorption and transport experiments on a German high volatile

bituminous coal“ Darmstadt, GeoDarmstadt 2010