cbm and co2-ecbm related sorption processes in coal
TRANSCRIPT
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
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
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
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
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
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
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
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