THESIS

Présentée a

L’UNIVERSITÉ DE PAU ET DES PAYS DE L’ADOUR

ÉCOLE DOCTORALE DES SCIENCES EXACTES ET LEURS APPLICATIONS

par Olga Patricia Ortiz Cancino

POUR OBTENER LE GRADE DE

DOCTEUR
SPECIALITÉ: Génie Pétrolier

ETUDE EXPERIMENTALE DE L’ADSORPTION DU METHANE DANS DES GAZ

DE SCHISTE COLOMBIENS ET DE LA SEPARATION METHANE/DIOXYDE
CARBONE.
Soutenance prévue le:
Après avis de:

Devant la Commission d’examen formée de:

Rapporteurs : Jose Luis LEGIDO Professeur – UNIV DE VIGO

Jimmy CASTILLO Professeur – UCV

Rodica CHIRIAC Ingénieur de Recherche- UNIV DE LYON

Cécile HORT Maître de Conférences HDR – UPPA

Directeur scientifique David BESSIERES Professeur -UPPA

1
2
3
4
 CONTENT

INTRODUCTION………………………………………………………………………………………… 21
INTRODUCTION GÉNÉRALE………………………………………………….……………………27

Chapter 1. FUNDAMENTAL ASPECTS OF ADSORPTION

1.1 GENERAL DEFINITIONS AND TERMINOLOGY.............................................................. 37

1.2 TYPES OF ADSORPTION ISOTHERMS ......................................................................... 38
1.2.1 Physisorption Isotherms ......................................................................................... 38
1.2.2 Chemisorption Isotherms........................................................................................ 40
1.3 GENERAL ASPECTS ON THE THERMODYNAMIC OF ADSORPTION AT THE GAS-SOLID
INTERFACE......................................................................................................................... 41
1.3.1 Single gas adsorption up to 1 bar............................................................................. 41
1.3.2 Adsorption for pressures above 1 bar ...................................................................... 44
1.4 HEATS OF ADSORPTION ........................................................................................... 45
1.4.1 The Integral and Differential Heats of Adsorption..................................................... 45
1.4.2 The Isothermal Heat of Adsorption.......................................................................... 47
1.4.3 The Adiabatic Heat of Adsorption ............................................................................ 47
1.4.4 The Isosteric Heat of Adsorption ............................................................................. 47
1.5 EXPERIMENTAL METHODOLOGY OF ADSORPTION AT THE GAS-SOLID INTERFACE ....... 49
1.5.1. Introduction and Basic Aspects............................................................................... 49
1.5.2 Determination of the Amount Adsorbed .................................................................. 50
1.5.3 Gas Adsorption Calorimetry .................................................................................... 52

Chapter 2. EXPERIMENTAL TECHNIQUES

2.1 CALORIMETRIC-MANOMETRIC DEVICE........................................................................... 57

2.1.1 Calorimeter description .......................................................................................... 57
2.1.2 Combined manometric-calorimetric device using the C80 Calorimeter ...................... 60
2.1.3 Measuring principle of the adsorption isotherm....................................................... 63
2.1.4 Determination of Adsorption Heat .......................................................................... 64
2.2 SPECIFIC DEVICE FOR THE STUDY OF GAS MIXTURE ADSORPTION .................................... 66
2.2.1 General description ................................................................................................ 66
2.2.2 Measuring principle................................................................................................ 68
2.3 DEVICES FOR SHALE CHARACTERIZATION ..................................................................... 66

5
 CONTENT

Chapter 3. STUDY OF METHANE ADSORPTION CAPACITY ON COLOMBIAN

SHALES

3.1 INTRODUCTION ............................................................................................................ 75

3.2 SAMPLE CHARACTERIZATION......................................................................................... 80
3.2.1. X-Ray diffraction analysis ....................................................................................... 80
3.2.2. Rock Eval Analysis.................................................................................................. 81
3.2.3 Specific surface area (SSA) ...................................................................................... 84
3.3 METHANE ADSORPTION MEASUREMENTS ..................................................................... 85
3.3.1 Sample preparation ................................................................................................ 85
3.3.2 Manometric adsorption setup. ................................................................................ 86
3.3.3 Calculation of excess adsorption ............................................................................. 86
3.3.4 Parameterization of excess adsorption isotherms..................................................... 89
3.4 VARIATION OF THE CH4 ADSORPTION CAPACITY ............................................................. 91
3.4.1 As function of organic matter.................................................................................. 91
3.4.2 As function of clay content...................................................................................... 92
3.4.3 As function of Thermal maturity .............................................................................. 94
3.4.4 As function of Specific Surface Area (BET area)......................................................... 96
3.5 CONCLUSIONS .............................................................................................................. 96

Chapter 4. CHARACTERIZATION AND ADSORPTION CAPACITY OF SHALE TO

METHANE, CARBON DIOXIDE AND TO AN EQUIMOLAR MIXTURE CH 4/CO2

4.1 INTRODUCTION .......................................................................................................... 105

4.2 SAMPLE CHARACTERIZATION....................................................................................... 105
4.3 ADSORPTION CAPACITY MEASUREMENTS .................................................................... 107
4.3.1 Sample preparation .............................................................................................. 107
4.3.2 Calorimetric-manometric set up............................................................................ 108
4.3.3 Co-adsorption CO2-CH4 setup ................................................................................ 109
4.3.4 Pure components (CO 2 and CH4 ) adsorption........................................................... 109
4.3.5 Differential heat of adsorption .............................................................................. 112
4.3.6 Equimolar mixture adsorption............................................................................... 113
4.4 CONCLUSIONS ............................................................................................................ 115

Chapter 5. NEW ADSORBENTS FOR CH4/CO2 SEPARATION

5.1 INTRODUCTION .......................................................................................................... 121

6
 CONTENT

5.2 SAMPLE CHARACTERIZATION....................................................................................... 122

5.2.1 Native silica samples............................................................................................. 122
5.2.2. Functionalized samples ........................................................................................ 123
5.3 ADSORPTION CAPACITY IN NATIVE SILICA SAMPLES...................................................... 125
5.3.1 Pure components (CO 2 and CH4 )............................................................................ 125
5.3.2 Differential heat of adsorption .............................................................................. 127
5.3.3 Equimolar mixture CH4/CO2 adsorption ................................................................. 128
5.4 ADSORPTION CAPACITY IN FUNCTIONALIZED SILICA NANOPARTICLES............................ 130
5.4.1 Pure components (CO 2 and CH4 )............................................................................ 130
5.4.2 Differential heat of adsorption .............................................................................. 132
5.4.3 Equimolar mixture adsorption............................................................................... 133
5.5 IDEAL ADSORBED SOLUTION THEOR Y .......................................................................... 135
5.6 CONCLUSIONS ............................................................................................................ 138

GENERAL CONCLUSIONS………………………………………………………………………………………….…141

CONCLUSIONS GÉNÉRALES …………………………..……………………………..…………..…………….…147

BIBLIOGRAPHY…………………………………………………..………………………………………………………..…153

ANNEXES…………………………………………………………………………………………………………………….……167

7
8
9
10
 FIGURE LIST

Figure 1.1 Classification of vapour adsorption isotherms (Rouquerol et al., 2014)................. 39

Figure 1.2 The layer model and the Gibbs representation of the surface excess amount.
(Rouquerol et al., 2014) ...................................................................................................... 42
Figure 1.3 Effect of a maximum surface excess (Rouquerol et al., 2014) ............................... 45
Figure 1.4 Isothermal adsorption in an isolated system (Ross et al., 1964) ........................... 46
Figure 1.5 Basic gas adsorption manometry (Rouquerol et al., 2014) ................................... 50
Figure 2.1 Calorimeter C80 (Pino, 2014) ............................................................................. 58
Figure 2.2 Thermocouples distribution (Pino, 2014) ............................................................ 59
Figure 2.3 Calorimetric-manometric equipment (Pino, 2014) .............................................. 61
Figure 2.4 Acquisition and control window from Labview Program ...................................... 62
Figure 2.5 Diagram of experimental equipment for mixtures............................................... 67
Figure 3.1 Prospective Shale Basins in Colombia (EIA, 2013) ................................................ 76
Figure 3.2 Location of Middle Magdalena Valley Basin (in yellow) and sample site (black circle)
(Modified from Barrero et al., 2007) ................................................................................... 77
Figure 3.3 XRD diffractograms of bulk samples ................................................................... 81
Figure 3.4 Tmax - HI Plot ................................................................................................... 83
Figure 3.5 Correlation between BET and TOC ..................................................................... 85
Figure 3.6 Schematic diagram of the HP/HT manometric set-up .......................................... 86
Figure 3.7 Methane adsorption capacity at 50°C................................................................. 88
Figure 3.8 Methane adsorption capacity at 75°C................................................................. 88
Figure 3.9 CH4 adsorption capacity for sample S2B at 50°C and 75°C.................................... 89
Figure 3.10 CH4 adsorption capacity nexc (at 3 MPa and 50°C) as function of TOC ................... 91
Figure 3.11 CH4 adsorption capacity at 50°C as function of TOC. ........................................... 92
Figure 3.12 (a) TOC- normalized Langmuir adsorption capacity (n L) content and (b) TOC-
normalized adsorption capacity at 3 MPa as a function of total clay content, this work and
literature data.................................................................................................................... 93
Figure 3.13 TOC-normalized adsorption capacity at 3 MPa as function of (a) Kaolinite content
(%) and (b) Illite content (%) .............................................................................................. 94
Figure 3.14 (a) TOC- normalized adsorption capacity at 3 MPa this work and (b) TOC-
normalized adsorption capacity at 3 MPa this work and literature data as function of Tmax
(maturity) .......................................................................................................................... 95
Figure 3.15 Langmuir adsorption capacity nL as function of BET ........................................... 96
Figure 4.1 Shale Gas and Shale Oil Basins (EIA, 2013) ........................................................ 103
Figure 4.2 X-ray diffraction pattern showing the bulk mineralogy of sample CH-1 ............. 106
Figure 4.3 Manometric-calorimetric device (Mouahid et al., 2012b) .................................. 108
Figure 4.4 Excess sorption isotherms of pure components and with fitted three parameters
function........................................................................................................................... 111
Figure 4.5 Heat of Adsorption.......................................................................................... 112
Figure 4.6 Equimolar mixture and individual adsorption ................................................... 114
Figure 4.7 Selectivity of CO2 over CH4............................................................................... 115
Figure 5.1 (a) Native silica nanoparticles (b) Functionalized silica nanoparticles.................. 124
Figure 5.2 Adsorption capacity for pure CH4 and CO2 ....................................................... 125
Figure 5.3 Excess sorption isotherms of pure components and with fitted three parameters
function........................................................................................................................... 126
Figure 5.4 Heat of adsorption.......................................................................................... 128
Figure 5.5 Equimolar mixture and individual sorption ...................................................... 129

11
 FIGURE LIST

Figure 5.6 Selectivity of CO2 over CH4. ............................................................................. 130

Figure 5.7 Adsorption capacity for pure CH4 and CO2 ........................................................ 131
Figure 5.8 Excess sorption isotherms of pure components and with fitted three parameters
function........................................................................................................................... 132
Figure 5.9 Heat of Adsorption.......................................................................................... 133
Figure 5.10 (a) Equimolar mixture and individual sorption (b) Individual components
adsorption in native and functionalized nanoparticles ........................................................ 134
Figure 5.11 Selectivity of CO2 over CH4 .............................................................................. 135
Figure 5.12 Sorption2 of CH4 and CO2 in (a) functionalized and (b) native particles.
Comparisson between experimental data and IAST method ............................................... 137

12
13
14
 TABLE LIST

Table 3.1 Colombia shale gas reservoir properties and resources (Modified from EIA, 2013) .. 77
Table 3.2 Mineralogic analysis of the samples ...................................................................... 81
Table 3.3 Rock-Eval analysis, Hydrogen Idex, Oxigen Index and Specific Surface Area (BET
method) ............................................................................................................................ 82
Table 3.4 Langmuir model fitting parameters at 50°C ........................................................... 90
Table 3.5 Langmuir model fitting parameters at 75°C ........................................................... 90
Table 4.1 Characterization of black shale (CH-1) ............................................................... 107
Table 4.2 Adsorption data for CO2 and CH4 pures ............................................................... 110
Table 4.3 Parameters for Langmuir Model with  ads fixed ................................................... 111
Table 4.4 Differential enthalpy of adsorption H ads (KJ/mol) ............................................ 112
Table 4.5 Adsorption data for mixture and individual components of the equimolar mixture
CH4 - CO2 ......................................................................................................................... 115
Table 5.1 Samples Characterization................................................................................... 123
Table 5.2 Adsorption data for pure CO2 and CH4................................................................. 126
Table 5.3 Parameters for Langmuir Model with  ads fixed ................................................... 127
Table 5.4 Differential enthalpy of adsorption H ads (KJ/mol) ............................................ 127
Table 5.5 Adsorption data for mixture and individual components of the equimolar mixture
CH4 - CO2 ........................................................................................................................ 129
Table 5.6 Adsorption data for CO2 and CH4 pures ............................................................... 131
Table 5.7 Parameters for Langmuir Model with  ads fixed ................................................... 132
Table 5.8 Differential enthalpy of adsorption H ads (KJ/mol) ............................................ 133
Table 5.9 Adsorption data for mixture and individual components of the equimolar mixture
CH4 - CO2.......................................................................................................................... 135

15
16
17
18
INTRODUCTION

19
20
 INTRODUCTION

My doctoral thesis was carry out in the Laboratory of Complex Fluids and their Reservoirs
(LFC-R) UMR5150 (CNRS / Total) of the Université de Pau et des Pays de l’ Adour (UPPA), thanks
to a funding from the Colombian National Agency Colfuturo, the Colombian AssociationofPublic
Universities (ASCUN), the Embassy of France and the Universidad Industrial de Santander. Its
subject was motivated by fundamental issues for Colombia: the search for new sources of
energy and the environmental concern of carbon dioxide emissions.

The increasing demand for energy has led in the last 20 years to the study of so -called
unconventional reserves. Issues related to the exploration and production of heavy crudes or
tar sands and more recently tight gas or shale gas, have attracted the attention of oil companies
and many research teams.

Shale gas, also known as source rock gas, is a natural gas contained in clay rocks rich in organic
matter (the organic matter content is referred as TOC - Total Organic Content). These rocks may
have a shale structure. Unlike conventional natural gas that is retained in a permeable rock for
easy operation, shale gas is trapped in the pores of a rock made impermeable by the clay
contained inside it. Extraction of shale gas, which is particularly difficult, requires the systematic
use of the combined techniques of directional drilling and expensive high -volume hydraulic
fracturing. Reservoir rocks containing shale gas may also contain shale oil (petroleum), but in
much smaller proportions.

Large-scale shale gas exploitation began in the 2000s, mainly in the United States, when the
price per barrel was permanently established above a high threshold. Environmental problems
associated with shale gas extraction, including intensive use and pollution of water supplies,
increased earthquakes and greenhouse gas emissions, in some countries, including the United
States, are very controversial.

Colombia has unconventional shale gas reserves in the Middle Magdalena Valley (MMV),avalley
between the Central Cordillera and Eastern Colombia. Preliminary explorations have shown a
potential for shale gas reserves in this sedimentary basin.

This context and this issue of Colombian shale gas are the origin of my doctoral thesis. As
previously mentioned, shale gas can be represented by a very low permeability mineral matrix
in which the kerogen is dispersed and over which methane is adsorbed. The reserves calculation
(Total Gas In Place, GIP) therefore assumes the estimation of these quantitie s of adsorbed

21
 INTRODUCTION

methane. In natural gas the main constituent is methane CH4 , which can be present in three
ways:

 As free gas, in the pores of the source rock,

 As dissolved gas, in the eventual presence of water contained in the pores, and
 As adsorbed gas, in kerogen and possibly in the mineral matter.

The amounts adsorbed represent up to 85% of the total amount of gas. In this context, in
connection with the Colombian oil company (Ecopetrol) which exploits the depositsof the MMV,
the main objective of my PhD studies was to contribute to the estimation of the reserves of
methane contained in various wells of the MMV.

Although methane adsorption in shale gas has been studied extensively in shales from Europe,
the United States and China, the data presented in this dissertation is, to our knowledge, the
first experimental results concerning to Colombian shales. This detailed study presented in
chapter 3, is the structuring axis of my PhD work. In addition to this study and knowing that the
separation, capture and storage of CO2 are issues related to shale gas and global environmental
issues, we conducted two complementary studies. The first concerns the potentiality of gas
shale for the geological storage of CO 2 with the comparative study of methane and CO 2
adsorption in a characterized kerogen. Finally, in collaboration with the University of Vigo, we
worked on the development of new adsorbents (silica nanoparticles) for the separation of
CH4/CO2.

The document is organized as follows:

 Chapter I is devoted to fundamental reminders on the adsorption phenomenon.

 Chapter II provides a detailed description of the experimental techniques used for the
measurement of adsorbed quantities and / or adsorption enthalpies in the case of pure
substances or binary mixtures.

 Chapter III presents the detailed study of the gas shale adsorption in reservoirs from the
Middle Magdalena Valley. As a matter of confidentiality, the company has not authorized
us to publish the names of the reservoirs concerned. Note that this work was the subject
of an international publication (Effect of Organic Matter and Thermal Maturity on
Methane Adsorption Capacity on the Middle Magdalena Valley Basin in Colombia,Energy

22
 INTRODUCTION

Fuels 2017, 31, 11698-11709) attached to this document. The geochemical and structural
characterizations were carried out in the Department of Geology of the Universidad
Autonoma de Madrid, in collaboration with Prof. Manuel Pozo.

 In Chapter IV, we present a study on selective CH 4/CO2 adsorption on a shale gas

characterized for the potential use of CO 2 for enhanced recovery processes and the
geological storage of CO 2 in depleted reservoirs. The study of MMV samples being
circumscribed in the case of methane, our choice fell on a previously characterized shale
sample (Pozo et al., 2017). This work has been published in an international journal
(Adsorption of pure CO 2 and CO2/CH4 mixture on a black shale sample: Manometry and
microcalorimetry measurements, Journal of Petroleum Science and Engineering, 159
(2017) 307-313).

 Finally, Chapter V is dedicated to the potential of silica nanoparticles for the separation
and storage of CO2. These particles were synthesized by Professor Veronica Salgueirino in
Physics Department at the University of Vigo.

The main results and conclusions are summarized in the last part.

23
24
25
26
INTRODUCTION GÉNÉRALE

27
28
 INTRODUCTION GÉNÉRALE

Ma thèse de doctorat a eu pour cadre le Laboratoire des Fluides Complexes et leurs Réservoirs
UMR5150 (CNRS/Total) de l’Université de Pau et des Pays de l’Adour (UPPA), grâce à un
financement de l’Agence Nationale Colombienne Colfuturo, de l’Association Colombienne des
Universités Publiques (ASCUN), de l’Embase de la France et de l’Universidad Industrial de
Santander. Son sujet a été motivé par des enjeux fondamentaux pour l a Colombie: la recherche
de nouvelles sources d’énergie et la préoccupation environnementale des émissionsde dioxyde
de carbone CO2.

La demande toujours croissante d’énergie a suscité lors des 20 dernières années l’émergence
de l’étude des réserves dites non conventionnelles. Les problématiques en lien avec
l’exploration et la production des bruts lourds ou sables bitumineux et plus récemmentdestight
gas ou shales gaz, gaz de schiste de français, ont mobilisé l’attention de compagnies pétrolières
et de nombreuses équipes de recherche.

Le gaz de schiste, également appelé gaz de roche-mère et connu sous le nom de shale gaz, est
un gaz naturel contenu dans des roches argileuses riches en matières organiques (le contenude
matière organique est désigné par le TOC: Total Organic Content). Ces roches peuvent avoirune
structure litée de schiste. Contrairement au gaz naturel conventionnel qui est retenu dans une
roche perméable permettant une exploitation facile, le gaz de schiste est piégé dans les
porosités d'une roche rendue imperméable par l'argile qu'elle contient. L'extraction du gaz de
schiste, particulièrement difficile, nécessite le recours systématique aux techniquescombinées
du forage dirigé et de la fracturation hydraulique à grands volumes particulièrement coûteuses.
Les roches-réservoir contenant du gaz de schiste peuvent également contenir de l'huile de
schiste (pétrole), mais dans des proportions beaucoup plus faibles.

L'exploitation à grande échelle du gaz de schiste a débuté durant les années 2000,
principalement aux Etats-Unis, lorsque le prix du baril était établi durablement au-dessus d'un
seuil élevé. Les problèmes environnementaux associés à l'extraction du gaz de schiste,
notamment l'utilisation intensive et la pollution des réserves d'eau, l'augmentation des
tremblements de terre ainsi que l'émission de gaz à effet de serre, entraînent danscertainspays,
y compris aux États-Unis, est très polémique.

La Colombie dispose de réserves non conventionnelles de gaz de schiste dans le Middle

Magdalena Valley (MMV): une vallée située entre les Cordillères centrales et Est de laColombie.
Des explorations préliminaires ont montré les potentialités en réserves de gaz de schiste de ce
bassin sédimentaire.

29
 INTRODUCTION GÉNÉRALE

Ce contexte et cette problématique des gaz de schiste colombiens sont à l’origine de ma thèse
de doctorat. Comme précédemment évoqué, les gaz de schiste peuvent être représenté s par
une matrice minérale très peu perméable dans laquelle sont disséminés des réacteurs
organiques, le kérogène, sur lesquels le méthane vient s’adsorber. L’estimation des réserves
(Total Gas In Place, GIP) suppose donc l’estimation de ces quantités de méthane adsorbé. Car,
le gaz, un gaz sec dont le constituant majoritaire est le méthane CH 4, peut être présent soustrois
formes:

 sous forme libre dans les pores de la roche mère,

 sous forme dissoute dans l’eau éventuelle contenue dans les pores,
 sous forme adsorbé dans le kérogène et éventuellement dans la matière minérale.

Les quantités adsorbées représentent jusqu’à 85% de la quantité de gaz totale. Dans ce cadre,
en lien avec la société pétrolière colombienne (Ecopetrol) qui exploite les gisements de la
(MMV), l’objectif principal de mon doctorat était de contribuer à l’estimation des réserves de
méthane contenues dans différents puits de la MMV.

Si l’adsorption de méthane dans les gaz de schiste a fait l’objet de nombreuses études dansdes
shales en provenance d’Europe, des Etats-Unis, de Chine, les données présentées dans ce
mémoire de doctorat sont, à notre connaissance, des premiers résultats expérimentaux
concernant des shales colombiens. Cette étude détaillée dans le chapitre 3 est l’axe structurant
de mon travail de doctorat. En complément à cette étude et sachant que la séparation, le
captage et le stockage de CO 2 sont des problématiques connexes aux gaz de schistes et des
enjeux environnementaux globaux, nous avons mené deux études complémentaires. La
première concerne la potentialité des shales gaz pour le stockage géologique duCO 2 avecl’étude
comparative de l’adsorption méthane et CO 2 dans un kérogène caractérisé. Enfin, en
collaboration avec l’Université de Vigo, nous avons travaillé sur le développement de nouveaux
adsorbants (nanoparticules de silice) pour la séparation CH4/CO2 .

Le document est organisé de la manière suivante :

 Le chapitre I est consacré à des rappels fondamentaux sur le phénomène d’adsorption.

 Le chapitre II propose une description détaillée des techniques expérimentales utilisées

pour la mesure des quantités adsorbées et/ou des enthalpies d’adsorption dans le casde
corps purs ou de mélanges binaires.

30
 INTRODUCTION GÉNÉRALE

 Le chapitre III présente l’étude détaillée de l’adsorption des shales gaz issus de réservoirs
de la Middle Magdalena Valley. Par mesure de confidentialité, la société ne nous a pas
autorisés à publier le nom des réservoirs concernés. Notons que ce travail a fait l’objet
d’une publication internationale (Effect of Organic Matter and Thermal Maturity on
Methane Adsorption Capacity on Shales from the Middle Magdalena Valley Basin in
Colombia; Energy Fuels 2017, 31, 11698-11709) jointe en annexe à ce document. Les
caractérisations géochimiques et structurales ont été menées au se in du départementde
Géologie de l’Université Autonome de Madrid en collaboration avec le Pr. Manuel Pozo.

 Dans le chapitre IV, nous présentons une étude sur l’adsorption sélective CH 4/CO2 surun
gaz de schiste caractérisé en vue de l’utilisation potentielle du CO2 pour les procédés de
récupération assistée et le stockage géologique du CO 2 dans des réservoirs déplétés.
L’étude des échantillons de la (MMV) étant circonscrite au cas du méthane, notre choix
s’est porté sur un échantillon de shale précédemment caractérisé (Pozo et al. 2017). Ce
travail a fait l’objet d’une publication internationale aussi (Adsorption of pure CO 2 and
CO2 /CH4 mixture on a black shale sample: Manometry and microcalorimetry
measurements; Journal of Petroleum Science and Engineering, 159 (2017) 307-313).

 Enfin le chapitre V est dédié à la potentialité des nanoparticules de silice pour la

séparation et le stockage du CO 2 . Ces particules ont été synthétisées par la Pr. Veronica
Salgueirino du département de Physique de l’Université de Vigo.

Les résultats marquants et principales conclusions sont résumés dans la dernière partie.

31
32
33
34
 CHAPTER 1
FUNDAMENTAL ASPECTS OF ADSORPTION

35
36
 Chapter 1. Fundamental Aspects of Adsorption

This chapter presents some fundamental aspects about adsorption, considering this
phenomenon is the main physical topic investigated in this research work.

1.1 GENERAL DEFINITIONS AND TERMINOLOGY

Adsorption occurs whenever a solid surface is in contact with a gas or liquid, it is defined as the
increase in the density of the fluid in the vicinity of an interface. Adsorption has a great
technological importance, because some adsorbents are used as desiccants, catalysts, or
catalysts supports; others are used for the separation of gases, the purification of liquids,
pollution control, etc.

The term “adsorption” is related to a process in which molecules accumulate in the interfacial
layer; desorption denotes the opposite process. The solid material on which adsorption occurs
is the “adsorbent”. The substance in the adsorbed state is defined as the “adsorbate” but is
called the “adsorptive” when considering the bulk fluid. The penetration process by the
adsorbate molecules into the bulk solid or fluid phase is called “absorption”. Additionally, the
relation, at constant temperature, between the amount adsorbed and the equilibriumpressure,
or concentration, is known as the adsorption isotherm.

Adsorption may result either from physical or chemical interactions between the adsorbate and
the adsorbent. The terms physisorption or chemisorption refer to these two phenomena.
Adsorption is accompanied by a decrease in free energy and entropy of the adsorption system
and, thereby, this process is exothermic.

Rouquerol et al., (2014), detail as follows the most important distinguishing features between
physisorption and chemisorption:

a. Physisorption is a general phenomenon with a relatively low degree of specificity.

b. Chemisorbed molecules are linked to reactive parts of the surface and the adsorptionis
necessarily confined to a monolayer. Whereas, at high relative pressures, physisorption
generally occurs as a multilayer.
c. A physisorbed molecule keeps its identity and on desorption returns to the fluid phase
in its original form (reversible process).

37
 Chapter 1. Fundamental Aspects of Adsorption

d. The energy of chemisorption is the same order of magnitude as the energy change in a
comparable chemical reaction. Physisorption is always exothermic, but the energy
involved is generally not much larger than the energy of condensation of the adsorptive.
e. An activation energy is often involved in chemisorption and at low temperature the
system may not have sufficient thermal energy to attain thermodynamic equilibrium.
Physisorption systems generally attain equilibrium fairly rapidly, but equilibrationmay
be slow if the transport process is rate-determining.

1.2 TYPES OF ADSORPTION ISOTHERMS

1.2.1 Physisorption Isotherms

The amount of gas adsorbed, n a, divided by the mass, ms of solid, is dependent on the
equilibrium pressure, p, the temperature, T, and the nature of the gas–solid system. For a given
gas adsorbed on a particular solid at a constant temperature, one may write:

𝑛𝑎
= 𝑓(𝑝) 𝑇 (1.1)
𝑚𝑠

and if the gas is below its critical temperature:

𝑛𝑎 𝑝
= 𝑓 ( 0) (1.2)
𝑚𝑠 𝑝 𝑇

Where po is the saturation pressure of the adsorptive at T.

Equations (1.1) and (1.2) represent the adsorption isotherm which is the relationship between
the amount adsorbed by unit mass of solid and the equilibrium pressure (or relative pressure),
at a known temperature. These information is usually represented by means of a graphic.

The adsorption isotherms have different characteristic shapes, which is important because
theses shapes provide useful preliminary information about the pore structure ofthe adsorbent.

According to an extended IUPAC classification, the most of the vapour isotherms maybe divided
into nine groups (Figure 1.1).

38
 Chapter 1. Fundamental Aspects of Adsorption

B B

Figure 1.1 Classification of vapour adsorption isotherms (Rouquerol et al., 2014)

Types I, II, III, IV and V were similar to those originally proposed by Brunauer, Deming, Deming
and Teller (1940).

The type I(a) and I(b) isotherms are reversible and concave to the relative pressure axis. They
rise sharply at low relative pressures and reach a plateau, which is means that the amount
adsorbed by unit mass of solid approach a limiting value, as relative pressure approach to 1.
Type I(a) corresponds to the filling of narrow micropores, and Type I(b) indicates the presence
of wider ones.

Type II is concave to the relative pressure axis, then almost linear and finally convex. Thisshape
is characteristic of a non-porous or macroporous adsorbent, which allows unrestricted
multimolecular adsorption occur at high relative pressure . When the equilibriumpressure reach

39
 Chapter 1. Fundamental Aspects of Adsorption

the saturation vapour pressure, the adsorbed layer becomes a bulk liquid or solid; the point B
usually represents the completion of the monomolecular layer and the beginning of the
formation of the multimolecular layer. The value of the ordinate at point B gives an estimation
of the amount of adsorbate required to cover the unit mass of solid surface with a complete
monolayer. In Type II(a) there is a complete reversibility of the desorption-adsorptionisotherm
whereas in Type II(b) there is a narrow hysteresis loop as the result of inter-particle capillary
condensation.

Type III isotherm is convex to the relative pressure axis over the entire range. This shape
indicates that the interactions between adsorbent and adsorbate are weak on non-porous or
macroporous adsorbent.

The isotherms Type IV (a and b) are typical for mesoporous adsorbents; the initial part of them
are similar to Type II isotherm, they are stabilized at high relative pressures with a characteristic
saturation plateau, although it could be short or reduced to an inflection point. Type IV(a) is
more common than Type (b), and has a hysteresis loop which is related with the filling and
emptying of the mesopores by capillary condensation; the isotherms Type IV(b) are completely
reversible.

The type V isotherm indicates weak interactions adsorbent-adsorbate on microporous or

mesoporous solid. It is initially convex to the relative pressure axis and it is stabilized at high
relative pressures as type IV isotherms. This isotherm usually has a hysteresis loop which is
associated with pore filling and emptying.

The isotherm type VI is associated with layer by layer adsorption, for this reason it has “steps”;
the sharpness of these steps depends on the system and the temperature. This type ofisotherm
is not common.

1.2.2 Chemisorption Isotherms

Generally in chemisorption the isotherms are comparable to Type I(a) of physisorption
isotherms. The plateau is due to the completion of a chemically bound monolayer. Usuallythese
isotherms are known as Langmuir isotherms.

40
 Chapter 1. Fundamental Aspects of Adsorption

1.3 GENERAL ASPECTS ON THE THERMODYNAMIC OF ADSORPTION AT THE

GAS-SOLID INTERFACE

The aim of this section is simply to introduce the most appropriate quantities involved on the
interpretation of adsorption isotherm, which are obtained by the experimental m ethods
described in the next section 1.5.

Following the work of Rouquerol et al. (2014), we adopt in this presentation the following
conventions and simplifying assumptions:

(a) The case of a single gas adsorbed on a solid adsorbent

(b) The adsorbent is supposed inert
(c) By the use of “Gibbs dividing surface” (GDS), we define the associated surface excess
properties.

1.3.1 Single gas adsorption up to 1 bar

Suppose an adsorbent-adsorbate system, we do not have any information about the structure
of the adsorbed layer. Also suppose that local concentration c=dn/dV, decrease with z (z is the
distance from the adsorbent surface); at z=t (t, is the thickness of the adsorbed layer) the value
of gas concentration, cg, is constant. (Figure 1.2).

Figure 1.2(a) represents the hypothetical variation of local concentration. It is possible to

identify three different zones.

The zone I is occupied only by the absorbent, it is assumed no penetration of gas into the solid
(i.e. no absorption), therefore the adsorptive concentration in the solid is zero (cs =0).

The zone II represents the adsorbed layer. It is located between z=0 and z=t. The local
concentration, c, is higher than gas concentration cg in zone III and it depends on z.

In zone III, located at zt, the adsorbable gas has a uniform concentration, cg, because it is far
away from the solid surface. In this zone the concentration only depends on the equilibrium
pressure and temperature.

41
 Chapter 1. Fundamental Aspects of Adsorption

Figure 1.2 The layer model and the Gibbs representation of the surface excess amount.
(Rouquerol et al., 2014)

According to the previous figure, the volume of the adsorbed layer, V a, can be expressed by
means of the following expression:

𝑉a = 𝐴 𝑡 (1.3)

Where A is the interfacial area and t, the thickness as is mentioned earlier.

The amount adsorbed, na, of substance in the adsorbed layer, which corresponds to area (d+e)
in figure 1.2(a), is:

𝑉a 𝑡
𝑛a = ∫0 𝑐 d𝑉 = 𝐴 ∫0 𝑐 d𝑧 (1.4)

The total amount of the adsorbable substance in the total system, n, is the sum of the amount
adsorbed, na, and the amount that remains in the gas phase, cgV g, thus na is:

𝑛a = 𝑛 − 𝑐 g 𝑉 g (1.5)

where V g is the volume occupied by the gas at the concentration cg.

42
 Chapter 1. Fundamental Aspects of Adsorption

In order to have the exact value on na, it is necessary to know the values of V g or the variation of
c with z, which is very difficult. For this reason Gibbs proposed an alternative approach;itmakes
use of the concept of “surface excess” to quantify the amount adsorbed.

Figure 1.2(b) shows a reference system, which is divided in two zones by an imaginary surface
called GDS located close to the adsorbent surface. This system has the same total volume asthe
real one, so that:

𝑉 = 𝑉 s,o + 𝑉 g,o = 𝑉 s + 𝑉 a + 𝑉 g (1.6)

Where V s,o and V g,o are the volumes of zone I and zone II respectively, (Figure 1.2b).

In this system, n, is the surface excess amount (area d in figure 1.2(b)), which can be expressed
as:

𝑛𝜎 = 𝑛 − 𝑐 g 𝑉 g,o (1.7)

Where n is the total amount of the adsorptive (areas d+e+f), and cgV g,o is the amount which
would be present in the volume V g,o if the final equilibrium concentration, cg , would be constant
up to the GDS (areas e+f).

For convenience, GDS is located on the surface which is accessible to the adsorptive used, so
that V g,oV a+V g, thus:

𝑛𝜎 = 𝑛 − 𝑐 g (𝑉 a + 𝑉 g ) (1.8)

to obtain na , we can combine equations (1.5) and (1.8)

𝑛a = 𝑛𝜎 + 𝑐 g 𝑉 a (1.9)

At pressures up to 1 bar, the quantity cgV a is negligible in comparison to n , so nan.

Nevertheless, at higher pressures these quantities are different.

The “surface excess concentration”,, is defined as:

𝑛a
Γ= (1.10)
𝐴

Where A is the surface area, which is associated with the mass of adsorbent, ms . Therefore, the
specific area, a, is equal to A/ms .

Normally, the quantity measured and registered is the specific surface excess amount n/ms ,
where

43
 Chapter 1. Fundamental Aspects of Adsorption

𝑛σ
= Γ𝑎 (1.11)
𝑚s

n/ms , depends on the equilibrium pressure, p, and the adsorbent temperature, T. As is usual
to maintain constant the value of T, the following relation, is the adsorption isotherm:

𝑛σ
= 𝑓 (𝑝)𝑇 (1.12)
𝑚s

1.3.2 Adsorption for pressures above 1 bar

Adsorption process is more efficient at temperatures much lower than ambient but is more
expensive. In order to compensate the lack of efficiency for the higher temperature s (TTamb),
is normal to use higher pressures (5-50 bar, and even more).

Under these conditions, the concentration of the adsorptive in the phase gas is no negligible as
compared to its concentration in the adsorbed phase.

In figure 1.3, there are three concentrations profiles at different pressures ( p3p2p1). Athigher
pressures, the total amount adsorbed (area d+e) increase, but the surface excess amount (area
d) first increase and then decrease. The main importance of this surface excess amount is that
it avoids any experimental uncertainties and assumptions (e.g. concerning the volume V gofthe
gas phase).

In order to understand and to interpret the experimental data it is necessary to evaluate the
actual amount adsorbed and the space occupied by the adsorbed phase. To obtain the amount
adsorbed is required to know the volume containing the adsorbed layer, V a; sometimes V a is
equal to the micro-pore volume. This value can be evaluated from a standard adsorption
isotherm of N 2 at 77K. With V a and V s (obtained by experimental data or by theoretical
evaluation) it is possible to have V g from equation (1.6).

Then, with equation (1.5) the amount adsorbed is determined.

44
 Chapter 1. Fundamental Aspects of Adsorption

Figure 1.3 Effect of a maximum surface excess (Rouquerol et al., 2014)

1.4 HEATS OF ADSORPTION

The experimental determination of the energetics of adsorption, usually proceedsbymeasuring

a differential heat of adsorption, which may be the isosteric heat qst, the adiabatic or isothermal
calorimetric heats, or the integral heat of adsorption.

1.4.1 The Integral and Differential Heats of Adsorption

Consider an isolated system, consisting of a container and a thermostat of infinity capacity, in
which na moles of adsorbate at p1 is transferred isothermally from the gas phase to the surface
of an adsorbent that was previously at vacuum. The final equilibrium pressure is p 2. (Figure 1.4).

The integral heat of adsorption, qint, corresponds to the heat adsorbed by the bath. As is an
isolated system, the initial total energy is equal to the final.

𝐸1 + 𝑛𝐸 g = 𝐸2 + (𝑛 − 𝑛a )𝐸 g + 𝑛a 𝐸 a (1.13)

Where:

45
 Chapter 1. Fundamental Aspects of Adsorption

E1 and E2: Energy or bath at p1 and p2, respectively

Eg: Energy of gas
Ea: Energy of adsorbate
n: moles of an ideal gas at p1
na: moles adsorbed

Figure 1.4 Isothermal adsorption in an isolated system (Ross et al., 1964)

Applying the first law of thermodynamics:

𝑞 int = (𝐸2 − 𝐸1 ) + 𝑤 (1.14)

As no work is done, (w=0), thus:

𝑞 int = 𝐸2 − 𝐸1 = 𝑛a (𝐸 g − 𝐸 a ) (1.15)

The differential heat of adsorption, qdiff, is defined as:

𝜕𝑞 int
𝑞 diff = ( ) (1.16)
𝜕𝑛a 𝑇

46
 Chapter 1. Fundamental Aspects of Adsorption

1.4.2 The Isothermal Heat of Adsorption

In this case, the system is a thermostat bath of infinity capacity and a container, it has a
frictionless piston in one side. There is an ideal gas in equilibrium with an adsorbent inside the
container, then the volume of container is infinitesimally decreased by the piston(workisdone).
In this case:

𝑑𝐸total = −𝑑𝑤 (1.17)

The isothermal heat of adsorption, qth, is defined as:

𝜕𝐸1
𝑞 th = ( ) (1.18)
𝜕𝑛a 𝑇

With E1: Initial total energy of the bath.

Additionally,

𝜕𝑤
𝑞 th = 𝑞 diff − ( ) (1.19)
𝜕𝑛a 𝑇

The work done on the system is the isothermal heat of compression of the gas in the calorimeter,
dw=pdV g, and considering the gas as ideal:

𝑑𝑝
𝑞 th = 𝑞 diff + 𝑅𝑇 + 𝑉 g (1.20)
𝑑𝑛a

1.4.3 The Adiabatic Heat of Adsorption

Suppose an ideal gas in equilibrium with an adsorbent inside a container isolated, it has a
frictionless piston in on side; the volume is changed in an infinitesimal amount by means ofthe
piston.

In this case, after a balance of energy, the adiabatic heat is:

𝜕𝑉g
𝑞 ab = 𝑞 diff − 𝑝 (1.21)
𝜕𝑛a

1.4.4 The Isosteric Heat of Adsorption

It is defined by:

𝜕ln𝑝 𝑞 st
( ) = (𝑉g − 𝑉a) (1.22)
𝜕𝑇 𝑛a 𝑇

47
 Chapter 1. Fundamental Aspects of Adsorption

Where

qst: isosteric heat of adsorption

V g: molar volume of gas phase
V a: molar volume of adsorbed phase

The usual method for its determination is to plot ln(p) for a given na as function of 1/T, from a
series of adsorption isotherms.

On the other hand, the differential energy of adsorption can be considered as the change of
internal energy of the total adsorption system, produced by the adsorption on an infinitesimal
surface excess amount at constant temperature, volume and surface. It may be obtaineddirectly
by calorimetric measurement of the heat evolved by adsorption.

The cell of the calorimeter is considered as an open system, inside the cell are the adsorbent
and the adsorptive. If the adsorptive is introduced in a reversible way and the steps are small
enough, the internal energy can be write as:

g
d𝑈 = d𝑄rev + d𝑊rev + 𝑢 𝑇 d𝑛 (1.23)

Where:

dQrev: heat reversible exchanged with the surroundings at T

dWrev: reversible work of the gas against the external pressure
ugT: molar internal energy of adsorbable gas at T
dn: amount of adsorbable gas introduced in a step.

It is possible to obtain dWrev if the volume of the whole adsorption system is divided intwoparts,
V A (external to the calorimetric cell but in contact with the thermostat) and V C (inside the
calorimeter cell).

For a reversible compression of ideal gas by reduction of V A, the work is:

d𝑊rev = 𝑅𝑇d𝑛𝜎 + (𝑉A + 𝑉C )d𝑝 (1.24)

where dn is the amount adsorbed during the compression.

48
 Chapter 1. Fundamental Aspects of Adsorption

The work received by the calorimetric cell is:

d𝑊rev (𝐶) = 𝑅𝑇d𝑛𝜎 + 𝑉𝐶 d𝑝 (1.25)

Combining equations (1.24) and (1.25):

d𝑄𝑟𝑒𝑣 d𝑝 d𝑈𝜎
( ) + 𝑉𝐶 ( ) = [( ) ̇ (1.26)
− 𝑢g − 𝑅𝑇] = ∆ads ℎ𝑇,Γ
d𝑛𝜎 𝑇,𝐴 d𝑛 𝜎
𝑇,𝐴 d𝑛𝜎 𝑇,𝐴

̇ , because we have
This equation permits to obtain the differential enthalpy adsorption, ∆ads ℎ 𝑇,Γ
the following values:

dQrev: heat measured by calorimeter

dn: amount adsorbed
dp: increase in equilibrium pressure
V C: dead volume

1.5 EXPERIMENTAL METHODOLOGY OF ADSORPTION AT THE GAS-SOLID

INTERFACE

1.51. Introduction and Basic Aspects

This section is focused in to explain briefly the experimental procedure for the determinationof
gas adsorption isotherms (the relation between the amount of adsorbed gas and the equilibrium
pressure).

Gas adsorption measurements are not difficult to achieve, because currentl y there are many
automated instruments available for that. However, it is important to define the objectives of
the measurements; which is the most suitable technique for the particular gas-solid system;
what is the temperature and pressure range; and what is the operational procedure to follow.

In general, there are three physical parameters that can be used to obtain the gas adsorption
isotherm: pressure, mass and gas flow (in this work the equilibrium pressure was used); and
there are two experimental procedures to choose: the discontinuous (point by point) or the
continuous method.

49
 Chapter 1. Fundamental Aspects of Adsorption

1.5.2 Determination of the Amount Adsorbed

In this work, as previously mentioned, the method used is based on the measurement ofthe gas
pressure in a calibrated, constant volume, at a known temperature. This method is known as
the manometric method. With this method is possible to obtain at the same time the amount
adsorbed and the equilibrium pressure (with the same pressure transducer); maybe for this
reason this technique is the most used.

1.5.2.1 Simple Gas Adsorption Manometry (up to atmospheric pressure). Figure 1.5isanexample
of a simple set-up for obtaining the amount of adsorbed gas. This set-up is made of stainless
steel (except the adsorption bulb and its stopcock) with three valves and a pressure transducer.
The “dosing volume” is within the connecting tubes between the valves and the pressure
transducer.

Figure 1.5 Basic gas adsorption manometry (Rouquerol et al., 2014)

50
 Chapter 1. Fundamental Aspects of Adsorption

This technique uses the discontinuous (point by point) procedure. The pressure and
temperature of the dosing volume (known) are measured before to enter to the adsorptionbulb.
Later, when the equilibrium is reached, the amount adsorbed is calculated from the change in
pressure using an equation of state.

Most of the commercial or home-made gas adsorption instruments are similar to this design;its
simplicity makes their construction, tightness and maintenance easier and a low cost.

1.5.2.2 Simple Gas Adsorption Manometry (above atmospheric pressure). Nowadays,itiseasyto

build an equipment similar to the one shown in the figure 1.5, to be used for higher pressures.
Most of the high pressures experiments are not carried out above 50 bar, but nowadays there
are special fittings and pressure transducers available for pressures up to 1500 bar.

The following features must be taken into account for measurements carried out above
atmospheric pressure:

 Need to make use of metallic sample cells.

 Difficulties in the sample outgassing because of the small tubing diameter whichmakes
it difficult to reach a good vacuum.
 Need for safety devices, against the risks of explosion and also of leaks.
 Major incidence of the void volume or “dead space” of the sample bulb upon the
adsorption isotherm.

1.5.2.3 Calibration of Dosing Volumes . All measurementsofgasadsorptionwiththe manometric

technique depend on the use of a calibrated volume. This calibration can be carried out in two
different ways: directly or indirectly.

In the direct calibration a part of the equipment must be isolated, removed and weighed, filled
with an outgassed liquid of known density and then weighed again. This procedure has a
limitation, it is difficult to have a balance with a sensitivity of 0.1 mg and with a loading capacity
higher than 300 g. This means that modern stainless steel vessels used in the modernequipment
are too heavy to be weighed accurately.

51
 Chapter 1. Fundamental Aspects of Adsorption

In the indirect calibration, it is necessary to use an external calibrated volume in place of the
adsorption bulb. If there is a difference in temperature, a correction for thermal expansionmust
be done.

1.5.2.4 Determination of Dead Space Volume.Almostall the adsorptiontechniquesneedtoknow

the volume of the overall dead space. It means to know the real space accessible to the gas in
adsorption bulb filled with the sample, which is needed for calculate the amount adsorbed.One
way to achieve that involves expanding a gas into the adsorption bulb already containing the
adsorbent sample. The whole adsorption bulb is maintained at the sample temperature during
the subsequent adsorption experiment, whereas all other volumes of the set-up can be
considered to be entirely at the controlled room temperature. For a more detailed explanation,
see Rouquerol et al. 2014.

1.5.3 Gas Adsorption Calorimetry

In gas adsorption measurements, sometimes it is possible to obtain at once calorimetric data.
A discontinuous calorimetric-manometric experiment, as was used in this research, can be
designed in order to obtain the value of dQrev (the heat exchanged reversibly in an adsorption
step) which is needed with the dead volume for calculating the differential enthalpy of
adsorption. (See chapter 2 for a detailed explanation).

52
53
54
 CHAPTER 2
EXPERIMENTAL TECHNIQUES

55
56
 Chapter 2. Experimental Techniques

This chapter is dedicated to provide a description about the experimental techniques and the
associated experimental procedures used in the development of the present work. Acombined
calorimetric-manometric device was used to determine simultaneously adsorption isotherm
and heat of adsorption whereas another equipment was used in the case of the study of gas
adsorption in binary mixtures. Both devices were developed in the LFC-R by Mouahid (2010) and
Pino (2014) respectively.

2.1 CALORIMETRIC-MANOMETRIC DEVICE

2.1.1 Calorimeter description

The calorimeter used is a SETARAM C80, Calvet type, which was adapted for working at high
pressures and isothermal conditions by Bessieres (2000). It has a controller CS32 and a
microcomputer in order to acquire the data by means of LabVIEW program.

This calorimeter (figure 2.1) is a vertical cylinder above a rectangular base; it contains two holes
located within the calorimetric block, which are symmetric from the cylinder center. One hole
can host the measuring cell (1) whereas in the other is located the reference cell (2). Note that
the calorimetric block is in aluminum with a very high heat capacity in agreement withthe Calvet
Principle.

Each hole is surrounded by several thermocouples constituting the flowmeters (3) which
assumes the thermal links between calorimetric block and the cells. The heating of the
calorimetric block (4) is provided by an external heating element (7), a sheath (8) surrounding
the block allows the air circulation for cooling. The temperature is measured by a platinum
resistance probe (5) located between the cells; the regulation is carried out by a temperature
signal controlled with another platinum resistance (6) placed outside the block.

An insulation (10) between the sheath and the external metal wall, allows the thermal insulation
of the calorimeter. At the bottom of the insulation wall there are orifices (11) for air circulation.
On the top of calorimetric block, there is an internal chamber (12) which is closed by a cover (9).
This free volume is occupied by two barrels (13) which constitute an extension of the useful area
occupied by the cells.

57
Chapter 2. Experimental Techniques

Figure 2.1 Calorimeter C80 (Pino, 2014)

58
 Chapter 2. Experimental Techniques

Above the cover (9), there is a removable insulation plug (14) which insulates the calorimeter
when the cells are in place. Two metallic sleeves (15) allow the connection with the holes and
therefore the introduction of the cells.

Calorimetric signal

Following the Calvet’s principle, the calorimetric cells are placed in the cavity of the calorimetric
block whose temperature is imposed by the regulator. The two cells are surrounded by
calorimetric flowmeters to measure the thermal flow exchanged between the cell and the
calorimetric block and also to connect thermally the cells to the block which is heat conductor.

The flowmeters consist of nine slabs of eighteen thermocouples (Figure 2.2), it means 162
thermocouples which are also heat conductors.

Figure 2.2 Thermocouples distribution (Pino, 2014)

Measurement principle

̇ , a quasi-
As a cell is source of thermal power or a constant and continuous heat flux (𝑞)
stationary state is established in which this flux will be transmitted to the block through the
conductance of thermocouples. Each thermocouple transmits an elementary power, 𝑞̇𝑖. Since
the thermocouples are connected in series, it is possible to send an electromotive force (E) which
is directly proportional to the thermal power. Due to the use of the geometry proposedbyCalvet
(Calvet, 1959), all the heat exchanged between the cells (measurement and reference) andthe
calorimetric block is counted and therefore the electric signal is directly proportional to the
thermal signal. Then:

59
 Chapter 2. Experimental Techniques

𝑞̇ = 𝑘𝐸 (2.1)

Where k is the calorimeter calibration constant in (mW.µV -1).

Function of calorimetric block

The metallic calorimetric block has a large mass and a very high heat capacity relative to the
thermal effect studied, it is a source with which the system exchanges heat while remaining
under isothermal conditions. The thermal phenomenon that occurs in the measuring cell is
immediately dissipated almost entirely within the block without affecting the thermoregulation.

Determination of Calorimeter calibration constant

The electrical calibration was performed as described by Bessieres (2000) in a temperature

range between 303.15 (K) and 573.15 (K). This value is checked regularly. The principle of this
calibration is to send a known power into the active area of the calorimeter, the relationship
between the voltage of signal output by the calorimeter and the applied power is usedtoobtain
the calibration constant, k (μV.mW-1 ) for a given temperature. The variation of k on the basis of
the temperature (in ° C) is presented by the following fourth grade polynomial:

𝑘 = 32.1940 − 4.603 × 10−3 𝑇 − 1.3165 × 10−4 𝑇 2 + 2.735 × 10−8 𝑇 3 + 5.466 ×

10−10 𝑇 4 (2.2)

2.1.2 Combined manometric-calorimetric device using the C80 Calorimeter

For the simultaneous determination of adsorption enthalpies and adsorbed quantities, a
manometric measurement system was developed and coupled to the C80 calorimeter by
Mouahid (2011). The design of the system was done taking into account the internal dimensions
in the upper part of the calorimeter as well as the operation conditions of pressure and
temperature. A schematic view of this device is provided in Figure 2.3. The basic elements of
this system are a high pressure gauge (MKS Baratron type 121A, 1 to 25000 (Torr) with an
accuracy better than 0.01% of full scale) and a dosing volume. The last one is connected to the
measuring volume (which is in fact the calorimeter measuring cell) in which is placed the

60
 Chapter 2. Experimental Techniques

adsorbent. The elements can be isolated by means of ball valves thus limiting the appearance of
dead volume. All of the valves are controlled (through rods) from the upper part of the
calorimeter.

Figure 2.3 Calorimetric-manometric equipment (Pino, 2014)

61
 Chapter 2. Experimental Techniques

In order to obtain isothermal conditions, the manometric system was inserted intoan aluminum
block at the upper part of the calorimeter, just above the active zone, instead of the calorimeter
insulating plug normally used. Around this aluminum block was rolled a heater resistance
controlled by an external PID control. The same regulation was imposed into both the
calorimetric and manometric systems thus leading isothermal conditions in the entire system.

The pressure sensor signal has been synchronized with the acquisition and control signals from
the calorimeter allowing an automated monitoring of pressure, temperature and calorimetric
data, through a software written in Labview. A view of acquisition data window is shown in
Figure 2.4

Figure 2.4 Acquisition and control window from Labview program

The main characteristics of the pressure transmitter P, its operating range and its measurement
accuracy (0.01 % as above mentioned) were determined considering that the accuracy on
pressure is the main source of uncertainty in the manometric method. In this context,the choice

62
 Chapter 2. Experimental Techniques

was a MKS Baratron type 121A, which operates on pressures ranging from 0 to 3.5 (MPa) until
423.15 (K) through a signal conditioner, external to the gauge, provided by MKS Baratron. The
signal conditioner provides good stability of the output voltage signal in a range between 0-10
(V). This sensor has been calibrated at the factory at 323.15 (K), however, the voltage signal
depends on temperature variations; therefore voltage-pressure calibration curves were
generated for several temperatures between 303.15 (K) and 383.15 (K) using an already
calibrated pressure sensor, available in the laboratory. The pressure signal is also readable from
the Labview software as it can be seen in Figure 2.4.

2.1.3 Measuring principle of the adsorption isotherm

The experimental procedure is based on a mass balance conducted by the accurate
determination of the pressure, temperature and volume. This approach is based on an
expansion under isothermal conditions, of an amount of gas contained in the dosing volume Vd
to the measurement volume Vm containing the adsorbent. Such procedure has been largely
described in literature (see for example, Rouquerol et al., 2014). Therefore, we only recall a
summary of the main steps of the methodology used for obtaining the adsorption isotherms:

 The adsorbent sample is placed in the cell and its mass is carefully measured (the mass is
measured before outgassing) and chosen in order to get a large enough available adsorption
area; about 30 m2 is the minimum area required.

 The adsorbent is first dried under vacuum ((vacuum quality: ultimate pressure ≤1 × 10–2(Pa))
during 12 hours. The procedure depends on the nature of the adsorbent.

 The accessible volume in presence of the porous within the measuring vessel is then
determined through successive expansions of helium (He). The value of the accessible
volume is independent of the pressure in the study interval [0-3 (MPa)]. Helium was chosen
for this determination because it is not adsorbed by the adsorbent.

 When the value of the accessible volume is known, a quantity of gas is loaded to the dosing
volume at pressure and temperature known. Later, the volume of gas (V d) is then sent tothe

63
 Chapter 2. Experimental Techniques

measuring cell (V m). When the adsorption equilibrium is reached, the amount adsorbed can
be calculated by a material balance before and after the adsorption. The number of moles
adsorbed, 𝑛1𝑎𝑑𝑠 , during the first step can be calculated as follows:

𝑉𝑑 𝑉 +𝑉
𝑛1𝑎𝑑𝑠 (𝑇,𝑝1 ) = − 𝑑( 𝑚) (2.3)
𝑣0( 𝑇,𝑝0) 𝑣1 𝑇,𝑝1

where T, is the temperature for the adsorption isotherm, p0 is the pressure in V d before
adsorption, p1 is the equilibrium pressure after adsorption, v0 is the gas molar volume atT,p0
and v1 is the gas molar volume at T,p1. The values of the gas molar volumes at the considered
pressure and temperature conditions are determined through the use of a specificequations
of the studied gas recommended by the NIST.

 This quantity adsorbed which corresponds to an equilibrium between the adsorbed phase
and the gas phase is the first point of the adsorption isotherm.

 The isotherm is then described through an accumulative process (step by step method), the
expression to obtain the amount adsorbed for the others steps is:

1 1 𝑉𝑚
 𝑛𝑖𝑎𝑑𝑠 (𝑇,𝑝𝑖 ) = 𝑉𝑑 (∑𝑖𝑘=1 − ∑𝑖𝑘=1 )− (2.4)
𝑣2𝑘−2 𝑣2𝑘−1 𝑣𝑖+1

The overall uncertainty of the amount adsorbed (due to helium calibration, dead volume, and
pressure measurement) is determined to be lower than 2% over the entire pressure, a nd
temperature range.

2.1.4 Determination of Adsorption Heat

The main interest to combine a manometric system with a calorimeter is the simultaneous
determination of the heat of adsorption when describing the adsorption isotherm. When a
quantity of gas is introduced in the measuring cell, there is a thermal effect (exothermic) until
the adsorption equilibrium is reached. The calorimetric signal E is directly proportional to the
total heat Q, which is the result of two contributions, the latent heat of adsorption Qads and the
heat liberated by the compressed (or expanded) gas Qcomp:

64
 Chapter 2. Experimental Techniques

𝑄 = 𝑄ads + 𝑄comp (2.5)

As a consequence, the estimation of the heat of adsorption Qads requires the calculation of the
heat of compression and the total heat.

The total heat Q, can be estimated through the integration of the calorimetric signal E:

𝑄 = 𝑘 ∫ 𝐸 (𝑡) d𝑡 (2.6)

The determination of the heat of compression Qcomp is done by the following Maxwell
relationship:

𝜕𝑆
( ) = −∝ 𝑝 · 𝑉 (2.7)
𝜕𝑝 𝑇

where 𝛼𝑝 represents the isobaric expansion coefficient defined by 1/V (𝜕V/𝜕T) p.

As the calorimeter is an open system, the quantity of heat dissipated, 𝛿𝑄comp , by the
pressure drop, d𝑝 , under isothermal conditions, is:

𝛿𝑄comp = −𝛼𝑝 𝑉𝐸 𝑇d𝑝 (2.8)

This thermodynamic relationship can be used for both calorimetric cell and gas, thus the
calorimetric signal is a sum of two opposite effects: one resulting from the gas and the other
from the calorimetric cell wall:

𝛿𝑄comp = 𝛼SS 𝑉E 𝑇d𝑝 − 𝛼𝑝 𝑉E 𝑇d𝑝 (2.9)

where 𝛼SS is the isobaric coefficient of the material of which the adsorption cell is made, in this
case stainless steel, in K-1 (5,1*10-5 K-1), T the temperature in K, and 𝑉E is the volume taken into
account by the thermopiles deduced by helium measurement in m 3.

As small pressures steps p are applied in this work, the heat dissipated by the compressed
gas can be expressed as:

𝑄comp = −𝑉E (𝛼𝑃 − 𝛼SS )𝑇∆𝑝 (2.10)

The volume taken into account by the thermopiles V E can be obtained by helium expansions
from the dosing volume to the adsorption cell filled with the adsorbent. As Helium isconsidered

65
 Chapter 2. Experimental Techniques

as not adsorbed, the calorimetric signal recorded is only due to the heat dissipated by th e
compressed gas into the adsorption cell. By the knowledge of the helium isobaric coefficient
𝛼𝑝He and by calculating the integral of the calorimetric signal, it is possible to calculate V E:

𝑘 ∫ 𝐸( 𝑡) d𝑡
𝑉E = (2.11)
−(𝛼𝑝He−𝛼SS)𝑇∆𝑝

To obtain the differential enthalpy of adsorption in J. mol -1, it is necessary to combine Eqs.(2.5)
to (2.11). So, the final expression is:

𝑄ads 𝑘 ∫ 𝐸( 𝑡) d𝑡+𝑉E ( 𝛼𝑃 −𝛼SS ) 𝑇∆𝑝

∆𝐻 = = (2.12)
∆𝑛ads ∆𝑛

From the combined errors on the amount adsorbed and the reproducibility on heat
determination, the uncertainty on adsorption enthalpies is estimated to be less than 5%. Note
that the reliability of this instrument was extensively checked in previous studies (Mouahid,
2012a, 2012b; Pino, 2014) by comparisons with literature data.

2.2 SPECIFIC DEVICE FOR THE STUDY OF GAS MIXTURE ADSORPTION

A manometric system was developed for the study of gas co-adsorption in mixtures (Pino,2014).
Basically, it consists of a classical manometric system coupled with an analytical method,namely
a Gas Chromatograph, in order to obtain adsorption isotherms for gas mixtures up to 4 MPa and
in a temperature range between 303,15K and 353,15K. We provide the general description as
well as the principle measurement focusing on the main difficulties associated to the study of
gas mixture.

2.2.1 General description

A schematic view of the device is provided in Figure 2.5. The main elements of this set up are
the dosing volume (16.57 cm3 ), the measuring vessel (18.29 cm 3 ) in which the adsorbent is
placed, a pressure gauge (CPH / 6400 / WIKA, 1-50 bar) which has a precision better than 0.025%
of full scale; a recirculation pump (GK-M 24/02, max. flow: 2.81 l/min and max system pressure:

66
 Chapter 2. Experimental Techniques

15 MPa) for ensuring homogenization and for reducing the time for reaching the adsorption
equilibrium pressure; and a chromatograph (Agilent Technologies 7890A) for obtaining the gas
composition after adsorption, its precision is better than 1% for molar fraction readings.

The various parts are isolated with spherical valves, thus limiting the “dead space” volume.The
whole apparatus is regulated under isothermal conditions through the use of a heater wire
controlled by a PID regulator (Eurotherm 3208) and there were installed five thermocouples
(type K, accuracy of ±0.1 K) in different parts of the circuit, in order to check the value of the
constant temperature and ensure there is not temperature gradient during the isotherm
measured. The overall uncertainty of the amount adsorbed (due to helium calibrationprocedure
and pressure accuracy) is determined to be lower than 1% over the entire range investigated in
this study.

The main difficulty on the study of gas mixture adsorption is to assure the homogenization of
the mixture during the expansion, the diffusion time being ve ry elevated. In this context, the
recirculation pump was added in order to avoid composition gradient.

Figure 2.5 Diagram of experimental equipment for gas mixture adsorption

67
 Chapter 2. Experimental Techniques

2.2.2 Measuring principle

The protocol for obtaining adsorption isotherms is similar to that used in the calorimetricdevice.
It is a mass balance based on the accurate measurements of pressure, volume and temperature.
The adsorbent is placed in measurement cell. Then, a vacuum (vacuum quality: ultimate
pressure ≤1 × 10–2 (Pa)) is created and the volume accessible with the adsorbent inside the cell
is determined through successive helium expansions. The experimental protocol thus follows
with sample outgassing. The gas mixture which is previously analyzed using the gas
chromatograph, is then introduced from the dosing volume to the sample via a step-by-step
procedure. At each dose, the mixture is circulated around the circuit and across the sample for
at least 15 min in order to ensure that a homogeneous final mixture is obtained. Discret e
quantities (0.5 cm3 ) are then extracted from the isolated dosing volume and analyzed through
the use of the chromatograph.

Amount Adsorbed
For each dose, the total amount adsorbed is calculated by a mass balance before and after
adsorption equilibrium 2.13:

𝑔
𝑛𝑎𝑡𝑜𝑡𝑎𝑙 = 𝑛𝑑𝑡𝑜𝑡𝑎𝑙 − 𝑛𝑡𝑜𝑡𝑎𝑙 (2.13)

Where

𝑛𝑎𝑡𝑜𝑡𝑎𝑙 : Total adsorbed moles.

𝑛𝑑𝑡𝑜𝑡𝑎𝑙 : Total moles introduced in measurement cell

𝑔
𝑛𝑡𝑜𝑡𝑎𝑙 : Total moles in gas phase after adsorption

Besides, each component of the mixture contributes to the mass balance, as follows:

𝑛𝑎𝑡𝑜𝑡𝑎𝑙 = 𝑛1𝑎 + 𝑛𝑎2 (2.14)

𝑔 𝑔 𝑔
𝑛𝑡𝑜𝑡𝑎𝑙 = 𝑛1 + 𝑛2 (2.15)

68
 Chapter 2. Experimental Techniques

Individual Components Adsorbed

With the information obtained from the chromatograph when adsorption is reached, it is
possible to calculate the molar fraction, y, of each component in gas phase:
𝑔
𝑛1
𝑦1 = 𝑔 𝑔 (2.16)
𝑛1 +𝑛2

𝑦1 + 𝑦2 = 1 (2.17)

In the same way, it is possible to obtain the molar fraction, x, in adsorbed phase:

𝑛𝑎1
𝑥1 = (2.18)
𝑛1 +𝑛𝑎
𝑎
2

𝑥1 + 𝑥2 = 1 (2.19)

Thus, when the first dose of gas is sent into the measurement cell, the amount of each
component adsorbed is:

𝑛1𝑎 = 𝑛𝑎𝑡𝑜𝑡𝑎𝑙 .𝑥1 (2.20)

As it was mentioned earlier, after each step the dosing volume is evacuated and replaced with
the original mixture. So, the individual adsorbed gas, at each step, is:

𝑔 ,𝑖−1 𝑔,𝑖
𝑛1𝑎,𝑖 = (0,5 𝑛𝑡𝑜𝑡𝑎𝑙
𝑑
+ 𝑦1𝑖−1 . 𝑛𝑡𝑜𝑡𝑎𝑙 ) − (𝑦1𝑖 . 𝑛𝑡𝑜𝑡𝑎𝑙 ) (2.21)

Then, the individual amount adsorbed is obtained by a cumulative process:

𝑛1𝑎 = ∑𝑖 𝑛1𝑎,𝑖 (2.22)

The mole number of the equimolar mixture was estimated using the AGA8 EOS, especially
developed to describe gas and supercritical fluid mixtures properties.

69
 Chapter 2. Experimental Techniques

2.3 OTHER DEVICES

Characterization of shale samples was carried out in Universidad Autonoma de Madrid by

Professor Manuel Pozo. This characterization include s: X-Ray Diffraction Analysis, Rock Eval
Analysis, and Specific Surface Area (BET Analysis).

A description of the equipment used in this characterization is done in Chapter III.

70
71
72
 CHAPTER 3
STUDY OF METHANE ADSORPTION
CAPACITY ON COLOMBIAN SHALES

73
74
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

This chapter presents the results obtained from characterization and measurement of methane
adsorption capacity on Colombian Shales from the Middle Magdalena Valley Basin.

3.1 INTRODUCTION

In 2016, according to BP Statistical Review of World Energy, Colombia’s natural gas production
was almost equal to its consumption and the R/P ratio (Reserves to Production) was close to 12,
which it means that the country would have self-sufficiency until 2028. At the present time,
almost all of the country gas natural production comes from conventional reserves which could
be empty by the year 2028, meanwhile, unconventional technically recoverable gas reservesare
estimated to be twelve times greater than conventional ones. Therefore, natural gasproduction
from shale gas reservoirs could be an increasingly important source of natural gas supply.

Colombia has prospective shale gas potential within marine -deposited Cretaceous shale
formation in three main basins: Middle Magdalena Valley, Llanos and Maracaibo/Catatumbo
(Fig. 3.1). The organic-rich cretaceous shales (La Luna, Capacho and Gachetá) sourced much of
the conventional gas and oil produced in Colombia, and are similar in age to Eagle Ford shale
play in the USA. Some companies such as Ecopetrol (Colombian Petroleum Company),
ConocoPhillips, ExxonMobil, Shell and others have done shale exploration in Colombia.

In table 3.1 are shown the main properties and resources of Shale gas in Colombia. It is clear
that the country has a big potential in this kind of resources.

The Middle Magdalena Valley Basin (MMVB) is the most explored conventional oil and gas
producing basin, with over 40 discovered oil fields that produce mainly from Tertiary sandstone
reservoirs; in relation with unconventional reservoirs most of the exploration has been done in
this basin.

MMVB has 34,000 km2, it is stretched along the middle reaches of the Magdalena river and is
bounded to the north and south by the Espíritu Santo fault system (E.S.F.S) and the Girardot
foldbelt (GFB), respectively; to the northeast the basin is limited by the Bucaramanga-Santa
Marta fault system (B.S.M.F) and to the southeast by the Bituima and La Salina fault systems
(B.S.F.S); the western limit is marked by the westernmost onlap of the Neogene basin fill into

75
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

the Serranía de San Lucas (SL) and the Central Cordillera basement (CC) as is showed in Figure
3.2 (Barrero et al., 2007).

Figure 3.1 Prospective Shale Basins in Colombia (EIA, 2013)

In MMVB has been drilled seven exploratory wells. For this work, it was possible to obtain five
core samples of three of them; the black circle in Fig. 3.2 shows the location of these wellsclose
to Barrancabermeja. Due to confidentiality reasons the samples were named as S1A, S1B, S2A,
S2B and S3.

76
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

Table 3.1 Colombia shale gas reservoir properties and resources (Modified from EIA, 2013)

Figure 3.2 Location of Middle Magdalena Valley Basin (in yellow) and sample site (black circle)
(Modified from Barrero et al., 2007)

77
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

It is of general acceptance that natural gas can be stored into the shales in three different ways:
as free gas, adsorbed gas and dissolved gas (Curtis, 2002; Gasparik et al., 2012, 2014; Guo, 2013;
Ross and Bustin, 2008, 2009) being the adsorbed gas the main contribution (up to 85%) (Curtis,
2002; Liu et al., 2016; Montgomery et al., 2005; Wu et al., 2015; Yang et al., 2015). Therefore,
the quantity of adsorbed gas represents one of the most important parameters into gas shale
reserves and production estimations (Pan and Connell, 2015). In this sense, many laboratory
experiments have been carried out on methane adsorption in shale gas from different
worldwide basins with the objective to provide a better understanding of this phenomenon(Cui
et al., 2009; Gasparik et al., 2015; Hu, 2014; Ji et al., 2014; Lu et al., 1995; Ross and Bustin, 2007,
2008; Tian et al., 2016; Wang et al., 2013; Weniger et al., 2010; Zhang et al, 2012). In addition
to adsorption measurements, some works had obtained at the same time the free -gas amount
stored, having into account the confining-stress effect (Kang et al., 2014; Santos and Akkutlu,
2013; Sigal et al., 2013).

Adsorption is a complex process which depends on rock matrix and fluid properties, as well as
reservoir conditions (e.g. temperature and pressure). The main parameters affectingadsorption
capacity are: Total organic carbon (TOC), mineralogy, water content, temperature and pressure.
Most of the studies affirm that TOC is the main factor that controls adsorption uptake in shales
(Cui et al., 2009; Gasparik et al., 2015; Ji et al., 2014; Lu et al., 1995; Ross and Bustin, 2009;
Weniger et al., 2010; Zhang et al, 2012). It has been also reported that the type of kerogen as
well as its maturity can influence the adsorption ability in such a way that methane adsorption
capacity of kerogen decrease in the sense of type III > type II > type I (Ross and Bustin, 2008;
Zhang et al., 2012) while a higher maturity means a higher adsorption capacity (Gasparik et al.,
2015; Jarvie et al., 2007; Ross and Bustin, 2009). Others studies reported that sorption capacity
will first increase then decrease with maturity (Curtis et al., 2012; Xiong, et al., 2017).

The roles of shale mineral composition and pore structure have been largely studied (Ross and
Bustin, 2009; Xiong et al, 2017). In addition to organic matter, clay mineral may provide a
contribution upon adsorption capacity (Chalmer and Bustin, 2007, 2008; Gasparik, et al., 2012;
Jarvie et al., 2007; Ji et al., 2012; Lu et al., 1995; Ross and Bustin, 2008, 2009). Montmorillonite
and illite/smectite present a higher adsorption capacity than kaolinite, chlorite and illite (Ji et
al., 2012). At a nanometric scale, it is difficult to correlate the adsorption gas capacity directly
with the total organic content due to the pore size distribution and heterogeneity of surface
(Ross and Bustin, 2009); therefore some authors suggest that the adsorbed gas volume

78
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

evaluation should also be related to the surface area (Kim et al, 2017; Wang et al., 2016; Xia, et
al., 2017). The adsorption in clay-rich shales is due to their high internal area. So, the specific
surface area plays a significant role in gas adsorption (Mendhe et al., 2017) due to the micro-
porosity associated to organic matter. Zhang et al. (2016) reports that shales with highercontent
of clay minerals and similar TOC content have larger specific surface area. This is due do the
porosity hosted in the clay minerals.

Pressure increases the adsorption capacity to some extent, when it rises isothermally (Chen et
al., 2017). By the contrary, water content and temperature have a negative influence; water
may occupy the adsorption sites hence reducing the amount of adsorbed gas (Gasparik et al.,
2015; Gensterblum et al., 2014) a reduction in gas adsorption up to 40% has been found when
comparing moisture samples to dry samples (Ross and Bustin, 2008, 2009). Temperature is also
one of the factors influencing the state of shale gas. Gas adsorption being an exothermic
process, the adsorption capacity of shale decreases with increasing temperature (Guo, 2013;Lu
et al, 1995). The combined effect of pressure and temperature can be used during the
production stage since it represents gas desorption behavior (Wu et al., 2015). Although the
above mentioned parameters are the most studied, some works have been done focused to
dynamically changing pore volume adjustments due to adsorption layer taking up space and
overburden effects on core shale samples (Kang et al., 2014; Santos and Akkutlu, 2013; Sigal et
al., 2013;).

This review highlights that gas storage in shale is a complex multi -parameter process. An
understanding and a quantification of each parameter requires a very huge set of well -defined
experimental data. Despite the growing interest, research published on shale is mostly limited
to US and Canadian shales, China and more recently European black shales. Less studies have
been reported for South American Shales.

The objective of this chapter is to obtain methane adsorption data for selected Colombianshales
from the Middle Magdalena Valley, using a home-made manometric set-up, methane
adsorption isotherms were measured at 50°C and 75°C for pressures up to 3.5 MPa.

The methodology applied was:

 Geochemical and textural characterization of the samples

 CH4 adsorption capacity over ranges of pressure and temperature.
 Representation of the adsorption data by a three parameters Langmuir model.

79
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

 Variation of the CH4 uptake as a function of specific surface area (BET), Organic Matter
richness (TOC), clay content and thermal maturity.

3.2 SAMPLE CHARACTERIZATION

3.2.1. X-Ray diffraction analysis

Mineralogical analysis of samples was carried out by means of X-ray diffraction (XRD) using a
SIEMENS D-5000 equipment with a scanning speed of 1º2θ/min and Cu-kα radiation (40 kV, 20
mA). X-ray diffraction (XRD) is the most widely used technique for identification of minerals.
When an incident beam of X-ray interacts with crystalline matter (regular structure) the
diffraction (constructive interference) can occur for certain directions giving a set of reflections
characteristic of the analyzed substance (fingerprint). The diffraction reflections are related to
spacing of atomic planes in a sample (i.e. d-spacing) and wavelength of X-rays (λ).

XRD studies were achieved both on randomly oriented samples (bulk sample) and clay fraction
samples (<2 μm). Powdered whole-rock samples (milled and dry sieved at <63 μm for
homogenization) were scanned from 2º to 65º 2θ. The method of the mineral intensity factors
(MIF) was applied to XRD peak intensity ratios normalized to 100% with calibration constants
for the quantitative estimation of the mineral contents (Chung, 1974). The clay fraction (<2 µm)
was separated by centrifugation and samples were prepared from suspensions orientedon glass
slides. Identification of the clay fraction minerals was performed on oriented air-driedsamples,
solvated with ethylene glycol, and after heating at 550 ºC.

The XRD mineralogical analysis revealed that the five bulk-rock samples are made-up mostlyof
phyllosilicates (28-60%), quartz (11-54%) and calcite (3-50%). Other minerals include pyrite (4-
10%), apatite (<2%) and traces of gypsum (see Table 3.2). The sample S1A shows the highest
content in phyllosilicates (>50%) and pyrite. The remaining samples have a phyllosilicate content
up-to 30%, whereas quartz can reach 54% in S1B and calcite 50% in S2B. The value of d(060)
reflection is in all cases 1.49-1.50 Å indicating dioctahedral phyllosilicates. The oriented
aggregates of the clay fraction (<2μm) shows that samples are composed mostly of two clay
minerals: kaolinite and illite. In bulk sample kaolinite content is ranging between 15% (S2B) and
45% (S1A). Illite is subordinated (7-15% in bulk sample) and include always traces of illite-

80
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

smectite mixed layers. Two representative XRD patterns of bulk samples are shown in Figure
3.3.

Table 3.2 Mineralogical analysis of the samples

Sample Illite Kaolinite Quartz Calcite Pyrite Gypsum Apatite

(%w/w) (%w/w) (%w/w) (%w/w) (%w/w) (%w/w) (%w/w)

S1A 15 45 27 3 10 Id 0
S1B 11 19 54 10 6 Id 0
S2A 9 21 31 33 4 Id 2
S2B 13 15 11 50 9 Id 2
S3a 7 22 32 28 8 Id 2

Figure 3.3 XRD diffractograms of bulk samples .

3.2.2. Rock Eval Analysis

The Rock-Eval analysis of the selected whole rock samples was performed by a Rock-Eval VI
pyrolyser. The interesting parameters measured include TOC, Tmax, S1 (free hydrocarbons),S2
(oil potential), S3 (CO2 organic source and carbonate), as well as the related indices: oxygen
index (OI) and hydrogen index (HI).

Rock-Eval analysis can help to know the oil-gas potential of a rock, but also the type of organic
matter and the degree of maturation. The most interesting parameters measured are shown in
Table 3.3.

81
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

Table 3.3 Rock-Eval analysis, hydrogen index, oxygen index and Specific Surface Area (BET
method)

Sample T max TOC S1 S2 S3 HI OI SSA

S1A 459 3.8 1.93 3.75 0.14 99 3.69 6.9

S1B 463 4.7 2.00 3.61 0.23 78 4.94 10.28
S2A 478 3.1 0.35 0.70 0.18 22 5.76 13.05
S2B 487 8.8 0.39 1.90 0.42 22 4.79 26.29
S3 471 5.7 2.47 4.96 0.25 87 4.37 10.59

Tmax Thermal maturation parameter (°C) TOC Total organic carbon (wt %)
S1 Free hydrocarbons (mg HC/g rock) S2 Oil potential (mg HC/g rock)
S3 CO2 organic source (mg CO2/g rock) HI Hydrogen index
OI Oxygen index SSA Specific Surface area (m2/g rock)

Total organic carbon (TOC) is ranging between 3.12% (S2A) and 8.77% (S2B). TOC is the total
amount of organic carbon present in the sample. In shales, a TOC of 2-5% is considered good
and higher than 5% very good (Pozo et al., 2017). All analyzed samples have TOC higher than 3%
and in the case of S3 and S2B higher than 5%.

Besides TOC is important to consider the level of thermal maturation, which can be given bythe
Tmax value. This parameter is the temperature at which the maximum amount of hydrocarbons
degraded from kerogen were generated. The Tmax values range between 459ºC and 487ºC
(Table 3.3). In general, according to Peters (1986), Tmax values lower than 435ºC are considered
immature organic matter but between 435-455ºC indicate “oil window” conditions (mature
organic matter). Higher values of Tmax between 455-470ºC are considered transitional and
higher than 470ºC represents the wet gas zone (over-mature organic matter). Indeed, when
more mature is the rock, higher is the temperature (Tmax) required to release hydrocarbons
from kerogen. The sample S1A shows the lower Tmax value (459ºC), S1B, S2A and S3
intermediate values (463-478ºC) and S2B the highest (487ºC). Therefore the maturation order
is S2B > S2A > S3 > S1B > S1A. According to Tmax-HI plot (Fig. 3.4) all the samples are within the
post-mature stage but samples S1A and S1B are within the condensate-wet gas zone, whereas
samples S2A and S2B are within the dry gas window conditions, and sample S3 is betweenthem.
The samples S3, S2A and especially S2B (Tmax > 470ºC) are indicative of over-mature organic
matter.

82
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

Figure 3.4 Tmax-HI Plot.

Regarding S1 (free hydrocarbons) and S2 (oil potential) the concentrations are low and range
between 0.35 mg HC/g rock (Sample S2A) and 2.47 mg HC/g rock (Sample S3) for S1, and
between 0.7 mg HC/g rock (Sample S2A) and 4.96 mg HC/g rock (Sample S3) for S2. The relative
amounts of S1 and S2 depend on the type of organic matter but also on the duration and
temperature suffered by the rock. The parameter S3 indicates the CO 2 evolved from thermal

83
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

cracking during pyrolysis, reaching the highest value in Sample S2B (0.42 mg CO 2/g) and the
lowest in Sample S1A (0.14 mg CO 2/g).

The oxygen index (OI) is derived from the ratio (S3/TOC) x100 ranging from 3.69 (sample S1A)
to 5.76 (sample S2A). The hydrogen index (HI) is derived from the ratio (S2/TOC) x100 reaching
22 in samples S2A and S2B, between 78-87 in samples S1B and S3 and the highest value in S1A
(99). The type of kerogen present in a rock determines its quality. Type I kerogen is the highest
quality; type III is the lowest (McCarthy et al., 2011). The values of Tmax, HI and OI in the studied
samples let to include them as kerogen of type II-III (S1A, S1B, S3) to III (S2A, S2B), according to
Peters (1986), Gorin and Feist-Burkhardt (1990), and Xu and co-workers (2015) classifications
(Fig. 3.4). The maturation degree can affect the determination of kerogene type. Indeed, the
Tmax-HI and HI/OI plots are especially useful to determine kerogen type of immature rocks.
However, as a source rock is under maturation, the amount of hydrogen and oxygen relative to
carbon decreases and then the ratios tend to converge toward the origin of the plot. Therefore,
in post-mature rocks HI and OI are not actually indicative of the original kerogen quality.

3.2.3 Specific surface area (SSA)

The specific surface area (SSA) of minerals is one of the most important properties controlling
surface phenomena. The most widely used technique for determining SSA is based on gas
adsorption, notably of nitrogen gas. Adsorption isotherms, describing the amount of gas
adsorbed as a function of relative pressure (P/P 0) can exhibit different features depending on
the size of particles, the presence of organized pores, and the energetic propertiesofthe mineral
surface. Different methods of data analysis are used to derive quantitative information from
experimental adsorption curves of which the Brunauer Emmett and Teller (BET) analysis is the
most common. BET specific surface area (SSA) of powdered samples was analyzed by a
Micromeritics ASAP 2010 equipment and determined from a low-pressure nitrogen adsorption
isotherm at 77 K (-196ºC).

The BET specific surface area is ranging between 6.90 and 26.29 m 2/g, the highest value in
sample S2B (Table 3.3). Several authors (Cao et al., 2015; Gaspartik et al., 2014; Kim et al., 2017;
Li et al., 2016; Zhang et al., 2012) reported a relationship between kerogen characteristi cs
(thermal maturity, composition and type) and development of nanopores enhancing the gas
adsorption capacity of shales. This fact explain the higher BET values obtained in more matured
samples S2A and S2B (13.05-26.29 m2/g) when compared with the other samples (6.9-10.59

84
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

m2/g). The development of nanopores in kerogen can lead to an increase in adsorptionsiteswith

increasing TOC for the mature to post-mature shales. Moreover, the analyzed samples show
moderate positive correlation between TOC and BET spe cific surface area values (R2 = 0.71)
because of increasing adsorption sites with maturation and TOC content. (Fig. 3.5).

30

25 R² = 0.71

20
S1A
BET (m2 /g)

15 S1B
S2A
10
S2B
S3
5

0
0 2 4 6 8 10
TOC (%)

Figure 3.5 Correlation between BET and TOC.

3.3 METHANE ADSORPTION MEASUREMENTS

3.3.1 Sample preparation

Considering that the time required to reach the equilibrium on intact core samples could be
extremely long due to the low permeability of shales (Heller et al., 2014), the sampleswere split.
But depending of the organic matter distribution, the milling process may induce variations on
the available surface. Following the recommendations of Gasparik et al. (2014, 2015) cuttings
of different particles sizes were tested. For each sample, two particles sizes fractions ranging
from 1 to 1.5 mm and from 1.5 to 5 mm were investigated. A known mass of each sample was
introduced into the adsorption cell (Fig. 3.6) and dried at 105°C during 24h to remove the
moisture. The mass was carefully measured and chosen in order to get a large enough available
adsorption area, about 30 m2 is the minimum area required.

85
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

3.3.2 Manometric adsorption setup.

The instrument used is a HP manometric device. A schematic view of this “homemade”
apparatus is provided in Figure 3.6. The main elements are the reference or dosing cell (33.57
cm3), the adsorption cell (16.56 cm3) in which the adsorbent is placed, the pressure transducer
(MKS Baratron type 121 A, with an uncertainty of 0.01 % in the full scale ranging from the
vacuum to 3.5 (MPa)). The various parts are isolated with spherical valves, thus limitin g the
“dead space” volume. The whole apparatus is regulated under isothermal conditions through
the use of a heater wire controlled by a PID regulator (Eurotherm 3208). Five thermocouples
(type K, accuracy of ±0.1 K.) were placed in different parts of the circuit in order to check the
isothermal conditions. The overall uncertainty of the amount adsorbed (due to helium
calibration procedure and pressure accuracy) is lower than 1% over the entire [P,T] range
investigated.

Figure 3.6 Schematic diagram of the HP/HT manometric set-up.

3.3.3 Calculation of excess adsorption

The adsorption isotherms were determined using an accumulative process, with a value of
increased pressure about 2-3 bar between successive gas doses. The procedure consists in
expanding a gas from the dosing volume V d (dosing cell) into the adsorption cell (V m) which
contains the sample under isothermal conditions. Application of this experimental methodology
requires the previous determination of the two volumes V d and V m. The uncertainty of these
measures is inferior to 0.5% in both cases. The mass balance involves the void volume orvolume

86
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

accessible in presence of the adsorbent, which is a key parameter for the adsorption capacity.
Void volume was determined through helium expansions at each temperature and fordifferent
pressures. The choice of helium was dicted considering it is an inert, non-sorbing gas (Zhang et
al., 2012 and Lu et al., 1995). An additional drying is performed after that for 8-10 h. The
methane molar volume considered at the experimental conditions (P, T) is determinedwiththe
Span and Wagner EOS (Wagner and Span, 1993). The return to the thermodynamic equilibrium
was controlled by the pressure value. It should be observed that it was reached in a range from
45 to 60 min.

CH4 adsorption isotherms were measured at 50°C for all the samples up to 3.5 MPa and for some
of them (S1A, S2B and S3) at 75°C (up to pressures 2 MPa). The measurements were performed
with high accuracy up to a quite restricted pressure range (up to 3.5 MPa). Then a
phenomenological model applied to these data, allows to extend the pressure range and to
assess the CH4 uptake. Considering the difficulty associated to the very low adsorpti oninshales,
a set of three measurements was performed for each isotherm. The reproducibility was always
superior to 99% (Average Absolute Deviation -ADD- inferior to 1%). The experimental data are
displayed in Figure 3.7 (50°C) and in Figure 3.8 (75°C). Note that the sample S1B has very low
adsorption capacity in comparison with the others samples or literature data (Gasparik et al.,
2012, 2014; Guo, 2013). Alteration of this sample by oxidation may be the cause of this
degradation. In this context, this sample will not be considered in the study. Figure 3.9 reports
the effect of temperature for the Sample S2B. As the saturation of the sample was observed at
75°C, the pressure range was limited to pressures around 2 (MPa).

87
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

0.16

0.14

0.12 S1A
n exc (mol/Kg)

S1B
0.1
S2A
0.08
S2B
0.06
S3

0.04

0.02

0
0 0.5 1 1.5 2 2.5 3 3.5
P (Mpa)

Figure 3.7 Methane adsorption capacity at 50°C.

0.06

0.05

S1A
0.04
n exc (mol/Kg)

S2B
0.03

0.02 S3

0.01

0
0 0.5 1 1.5 2 2.5
P (Mpa)

Figure 3.8 Methane adsorption capacity at 75°C.

88
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

0.16
n exc (mol/Kg)

0.12 50C
75C

0.08

0.04

0
0 0.5 1 1.5 2 2.5 3 3.5
P (Mpa)

Figure 3.9 CH4 adsorption capacity for sample S2B at 50°C and 75°C.

3.3.4 Parameterization of excess adsorption isotherms

As reservoir pressures are higher than the experimental ones, it is necessary to extrapolate data
to wells pressure conditions. Therefore the experimental data were parameterizedusingafitting
procedure. Several approaches have been developed to this purpose (Chareonsuppanimit et
al., 2012; Gasparik et al., 2012; Hou et al., 2014; Rexer et al., 2013). The modified Langmuir
model is used as a standard model to describe vapor isotherms on shales (Rexer et al., 2013),
and it is widely accepted in the petroleum industry (Kim et al., 2017). This model includes the
adsorbed gas density as a fitting parameter. As this value is difficult to assess, most of the
studies have shown a good match when assuming it as the methane liquid density at normal
boiling point (Dresibach et al., 1999; Gasparik et al., 2012; Gensterblum et al., 2014; Yang et al.,
2014, 2015).

The experimental data were correlated using a three parameters Langmuir model describedby
Gensterblum (2009) and applied by Gasparik (2012):

𝑝 𝜌𝑔(𝑝,𝑇) 𝜌𝑔(𝑝,𝑇)
𝑛𝑒𝑥𝑐𝑒𝑠𝑠
𝑎𝑑𝑠 = 𝑛𝐿 · (1 − ) = 𝑛𝑎𝑏𝑠𝑜𝑙𝑢𝑡𝑒
𝑎𝑑𝑠 · (1 − ) (3.1)
𝑝+𝑝𝐿 𝜌𝑎𝑑𝑠 𝜌𝑎𝑑𝑠

In which, nexcess
ads is the adsorbed amount of gas (mol/Kg) at p (MPa); p L is the Langmuir pressure,

corresponding to the pressure at which half of the adsorption sites are occupied (monolayer);
nL is the amount adsorbed (mol/Kg) when all the monolayer is filled (maximum Langmuir

89
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

capacity);  g is the gas density (kg/m3) at p and T; and  ads (kg/m3) is the adsorbed phase density,
which was assumed as a fixed value of 421kg/m 3 (Weniger et al., 2010; Yang et al., 2015; Yuan
et al., 2015).

A standard deviation was calculated according to Pozo et al., (2017):

1
∆𝑛 = · √∑𝑁
1 (𝑛 𝐸𝑋𝑃 − 𝑛 𝐹𝐼𝑇 )
2 (3.2)
𝑁

Where N is the number of experimental data points and n EXP and nFIT are experimental measured
data and fitted data, respectively.

Fitting parameters are shown in Tables 3.4 and 3.5, the values of n in all the samples indicate
that the fitting procedure was successful and that Langmuir model represents the adsorption
behavior in a good way without restrictions. The values obtained for n L are similar to those
already reported in literature for shales or black shales (Gasparik et al., 2012, 2014). By the way,
such parameter should be regarded as useful information for the future assessment and
exploitation of the shales wells. Additionally, their knowledge represent a meaningful
information to study the effect of individual contributions to the methane adsorption.

Table 3.4 Langmuir model Fitting parameters at 50 °C

SAMPLE nL (mol/Kg) pL (MPa) n
S1A 0.0848 0.711 0.00098
S2A 0.1054 1.661 0.00017
S2B 0.1926 0.680 0.00040
S3 0.0783 0.315 0.00001

Table 3.5 Langmuir model Fitting parameters at 75 °C

SAMPLE nL (mol/Kg) pL (MPa) n
S1A 0.0169 0.071 0.00002
S2B 0.0562 0.118 0.00010
S3 0.0537 0.229 0.00021

90
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

3.4 VARIATION OF THE CH 4 ADSORPTION CAPACITY

3.4.1 As function of organic matter

CH4 adsorption capacity (n exc) shows a moderate positive correlation with TOC, taking into
account that the TOC of S1A and S2A are nearly similar (3.8 and 3.1). Even if no linear relation -
ship can be fitted to the adsorption data, the TOC remains as controlli ng factor of the adsorption
uptake. A small discrepancy is observed for sample S3. This is shown in Figure 3.10 where is
plotted the excess adsorption capacity (n exc) at 3 MPa (and 50°C) versus TOC.

When the adsorption uptake is plotted at a lower pressure (See Figures 3.11a and b at 0.3 MPa
and 0.5 MPa), a linear law is observed between TOC and adsorption. This is due to the filling of
micro-pores of the organic matter that occurs at a first stage during the adsorption process.

0.16
nexc 3 MPa (mol/Kg)

0.12

S1A
0.08
S2A
S3
0.04
S2B

0
0 2 4 6 8 10
TOC (%)

Figure 3.10 CH4 adsorption capacity n exc (at 3MPa and 50°C) as a function of TOC .

91
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

0.06
(a)

nexc 0.3 MPa (mol/Kg) 0.04

S1A
S2A

0.02 S2B
S3

0
0 2 4 6 8 10
TOC (%)

0.08
(b)
nexc 0.5 MPa (mol/Kg)

0.06

S1A
0.04
S2A
S2B
0.02
S3

0
0 2 4 6 8 10
TOC (%)

Figure 3.11. CH4 adsorption capacity at 50°C as function of TOC. (a) At 0.3 MPa and
(b) At 0.5 MPa.

3.4.2 As function of clay content

In order to take into account clay effect, the methodology proposed by Gasparik et al., (2014)
was followed. TOC-normalized adsorption capacities were plotted versus the total clay content
for all the samples (Fig. 3.12a and b). Form the obtained graphic, discrepancy of the adsorption
capacities of the three samples with analogous clay contents are depicted. Meanwhile, the
sample S1A, with a higher clay content shows adsorption capacity similar to the others samples.
This lack of correlation between the clay content and the adsorption capaci ties is also observed
with the literature data (Gasparik et al., 2012) (Fig. 3.12a and b).

92
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

4
(a)
S1A
3
S2A
nL (mol/KgTOC)

S2B
2 S3
Aalburg 2
Geverik
1
Posidonia

0
0 10 20 30 40 50 60 70
total clays (%)

2.5
(b)
S1A
n exc 3MPa (mol/KgTOC)

2
S2A

1.5 S2B
S3

1 Aalburg 2
Geverik
0.5 Posidonia

0
0 10 20 30 40 50 60 70
total clays (%)

Figure 3.12 (a) TOC- normalized Langmuir adsorption capacity (n L) content and (b) TOC-
normalized adsorption capacity at 3MPa as a function of total clay content, this work and
literature data 4.

Similar results were observed for the detailed analysis of the individual clays plottedinFig. 3.13a
and b, the effect of illite and kaolinite seems to be insignificant in agreement with previous
studies developed by Xiong et al., (2017) and Zhang et al., (2012). Furthermore, while the same
studies report that an important content of smectite can affect the adsorption capacity, the
samples studies on the present work only presented trace amounts of illite -smectite.

93
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

2.5
(a)
n exc 3MPa (mol/KgTOC)

1.5
S1A
S2A
1
S2B
S3
0.5

0
0 10 20 30 40 50
Kaolinite(%)

2.5
(b)
n exc 3MPa (mol/KgTOC)

1.5
S1A
S2A
1
S2B
S3
0.5

0
0 4 8 12 16 20
Illite (%)

Figure 3.13 TOC-normalized adsorption capacity at 3MPa)as a function of (a) Kaolinite content
(%) and (b) Illite content (%).

3.4.3 As function of Thermal maturity

The effect of thermal maturity on the TOC-normalized excess adsorption capacity at 3 MPa
(nexc3 MPa) is shown in Fig. 3.14a. Limited to the samples of the present study, maturity showed
no significant effect on the total amount of methane adsorbed.

Taking into account the still relatively small data base, a general trend that correlates thermal
maturity and adsorption capacity is not observed. A wider range of samples maturitiesisneeded
in order to conclude on its effect over the adsorption capacity. Using data from Gasparik et al.,
(2012) the plot between maturity and adsorption capacity was obtained (see fig. 3.14b). A

94
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

consistency is observed between the two set of data. The TOC-normalized adsorption capacity
linearly increases with maturity within the investigated range.

In the literature, a variety of behaviors is reported between the maturity and adsorption
capacity. When the TOC-normalized adsorption capacities correlated positively with maturityin
terms of Vitrinite Reflectance (VRr), the maximal value of VRr was ~2.5%.4-5 which is in
agreement with our observation.

2.5
(a)

2
nEXC 3MPa (mol/KgTOC)

1.5 S1A

S2A
1
S2B

0.5 S3

0
455 460 465 470 475 480 485 490
Tmax (°C)

2.5 S1A S2A S2B S3 Aalburg 2 Posidonia

(b)

2
nEXC 3MPa (mol/KgTOC)

1.5

0.5

0
420 430 440 450 460 470 480 490
Tmax (°C)

Figure 3.14 (a) TOC-normalized sorption capacity at 3MPa this work and (b) TOC-normalized
adsorption capacity at 3MPa this work and literature data as function of Tmax (maturity)

95
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

3.4.4 As function of Specific Surface Area (BET area)

For these samples was obtained a linear relationship between adsorption capacity and SSAasis
shown in Fig. 3.15. A similar behavior was found by Kim et al., (2017). Sample S2B has shown
the largest adsorption capacity, this behavior is due to it has the highest values of TOC and SSA.

0.2
R² = 0.9472
0.16
nL (mol/Kg)

0.12
S1A
S2A
0.08
S2B
S3
0.04

0
0 5 10 15 20 25 30
BET (m2 /gr)

Figure 3.15 Langmuir adsorption capacity n L as function of BET

3.5 CONCLUSIONS
On the basis of samples studied, the following conclusions can be addressed:

 New and original experimental data for methane adsorption capacity were obtainedfor
Colombian Shales. To the best of our knowledge, this is the first study devotedtoMiddle
Magdalena Valley Shale Gas.

 The shales were fully characterized in terms of textural and geochemical properties.

 Experimental excess adsorption data were fitted to a Langmuir three parametermodel,

thus providing a basis for the Gas-In-Place estimations.

96
 Chapter 3. Study of Methane Adsorption Capacity on Colombian Shales

 Even if a linear regression was not found, the results support that TOC is the primary
factor controlling the methane adsorption capacity. Additionally, it was found a
moderate positive relationship between TOC and BET.

 Specific Surface Area has an important role in methane adsorption capacity, a positive
relationship was found between them.

 The thermal maturity doesn’t show significant effect on adsorption capacity. Excepted
one sample, the dataset obtained is very consistent considering the correlationbetween
methane adsorption capacity, TOC and specific area.

97
98
99
100
 CHAPTER 4
CHARACTERIZATION AND ADSORPTION
CAPACITY OF SHALE TO METHANE,
CARBON DIOXIDE AND TO AN
EQUIMOLAR MIXTURE CH4/CO2

101
102
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

This chapter presents the results obtained from characterization, measurement of adsorption
capacity and heat of adsorption for methane and carbon dioxide in a Silurian black shale sample
from Spain. This sample was previously studied only for methane adsorption capacity by Pozo
et al., (2016). Besides, the selectivity of kerogen for an equimolar mixture CH 4/CO2 is presented
too.

4.1 INTRODUCTION

The still growing demand for energy has stimulated the exploitation of unconventional
reservoirs like shale gas. Colored areas in Figure 4.1 shows the basins with shale formationswith
oil and natural gas in-place.

Figure 4.1 Shale Gas and Shale Oil Basins (EIA, 2013)

In word, there are 35,782 Tcf (1,013 E12 m3) of shale gas in place and 7,795 Tcf (221 E12 m3) of
technically recoverable gas shale reserves and 6,753 B bbl (1,074 E9 m3)of shale oil in place and
334.6 B bbls (532 E9 m3)of shale oil technically recoverable (EIA, 2013). It is clear that there is a
great potential of hydrocarbons in this kind of reservoirs.

103
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

In the particular case of the gas, natural gas produced from organic shale formations is known
as shale gas. The total amount of this natural gas is composed by three sources: the adsorbed
gas, the free gas and dissolved gas, the last one being negligible and consequently the total shale
gas content can be considered as the sum of adsorbed and free gas (Gasparik et al., 2012; Guo,
2013). The adsorbed gas ranges from 20 to 80% of total gas reserves as we ll as recovery rates
(Yu et al., 2014), for this reason adsorption is a leading criterion governing shale gas production.

In an effort to better understand the role of adsorption on production from gas shales,
numerous authors have done valuable contributions to the literature through laboratory
measurements. Much research focused on the estimation of the adsorption capacity of
methane, the most abundant component of shale gas, at defined borehole conditions (Clarkson
and Bustin, 2000; Gasparik et al., 2012, 2014a, 2014b; Guo, 2013; Mendhe et al., 2017). Most
of the studies point that methane is adsorbed onto the organic phase -known as kerogen-
whereas the adsorbed gas in the inorganic matrix is generally supposed to be negligible (Collel
et al., 2014; Ross and Bustin, 2009). In addition to natural gas interests, more recent studies
have evaluated the potential of shales for geological CO 2 storage (Charoensuppanimit et al.,
2016; Heller and Zoback, 2014). This is a remarkable observation not only for storage but also
from enhanced shale gas production point of view as it implicitly points out that strongeraffinity
of CO2 to the organic materials of shale could initiate an added mechanism of displacement of
the originally in-place CH4, when CO2 is introduced into the shale gas environment. The design
of the CO2 sequestration process requires knowledge of the adsorption behavior of CO 2
(Chareonnsuppanimit et al., 2012). These studies showed that adsorption capacity of CO 2, as
well as methane, appear to be related to the total organic carbon. Charoensuppanimit et al.
(2016) focused on Woodford shales from Payne and Hancock counties. For each shale sample,
CO2 adsorption was 2.1-8.6 times higher than methane adsorption. Adsorption capacity was
found correlated to the TOC content of these samples. Kang et al. (2011) identified that
micropores in organic matter acted as molecular sieves that allowed for only linear molecules,
such as CO2, to access their pore space. Despite these studies, gas adsorption, storage and
diffusion in organic-rich shales is a complex multiparameter process that still awaits for reliable
description in order to elucidate the effect of individual parameters. Indeed, kerogen is a very
complex structure depending on both its origin and thermal alteration. Various realisticmodels
have been proposed to represent the physical and chemical properties of kerogen through the
use of molecular simulations (Collell et al., 2014a, 2014b; Sui and Yao, 2016). In the latter, Sui
and Yao (2016) analyzed the effect of surface chemistry for CH4 /CO2 adsorption in kerogen. The

104
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

authors proposed a 3-D molecular model of kerogen and they studied the CO 2/CH4 interactions
with kerogen. Both for pure CO 2, CH4 or their equimolar mixture, the maximum adsorption of
CO2 in kerogen is larger than CH4 adsorbed in kerogen. Additionally, the simulated isostericheat
of adsorption display higher values for CO 2/kerogen than for CH4/kerogen. This is explained by
the Coulomb and van der Waals interactions between CO 2 and kerogen which play a key role in
the process of adsorption, whereas, in the CH 4 adsorption process, there was a little Coulomb
interaction between CH4 and kerogen.

Therefore, the main objective of this chapter is to understand the adsorption processintoshales.
So the following are the steps developed in order to achieve this goal:

 Characterization of the sample: mineralogical and geochemical analysis and specific

surface area.
 Measurement of adsorption capacity for CO 2, CH4 and for an equimolar mixture ofthese
components.
 Measurement of the heat of adsorption.
 Determination of selectivity of CO 2/CH4.

4.2 SAMPLE CHARACTERIZATION

In a previous work (Pozo et al., 2016) six shales samples were studied in order to obtain a
relationship among composition, properties and methane adsorption. For this research we
chose which had the highest adsorption capacity.

The shale sample named CH-1 was collected from Silurian pelitic to metapelitic materials
outcropping near Checa town (Guadalajara province) in the Iberian Range from Central Spain.
The shale is black in color, presents lamination to slaty cleavage and contains abundant
graptolite fossils dating it as Silurian (Gutierrez-Marco and Storch, 1998).

Mineralogical analysis was carried out by means of X-ray diffraction (XRD) using a SIEMENS D-
5000 equipment with a scanning speed of 1º2θ/min and Cu-kα radiation (40 kV, 20 mA). XRD
studies were achieved both on randomly oriented sample (bulk sample) and clayfractionsample
(<2 μm). Powdered (<63 μm) whole-rock sample was scanned from 2º to 65º 2θ. The method of
the mineral intensity factors (MIF) was applied to XRD peak intensity ratios normalized to 100%
with calibration constants for the quantitative estimation of the mineral content. Clay fraction

105
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

sample was prepared from suspensions oriented on glass slides. Identification of the clay
fraction minerals was performed on oriented air-dried samples, solvated with ethylene glycol,
and after heating at 550 ºC (Moore and Reynolds, 1989). The porosity and BET specific surface
area determined by a low-pressure adsorption isotherm at 77 K, were analyzed by a
Micromeritics ASAP 2010 equipment. Organic carbon, N, H and S were analyzed by means of a
LECO CHNS-932 elemental analyser. This mineralogical analysis was carried out in Geology
Department at Universidad Autonoma de Madrid

Besides, a detailed compositional analysis of this black shale including kerogen type and
maturation was performed in a previous work aimed to correlate their structural and
geochemical properties and the methane uptakes of different shale samples (Pozo et al.,2017).
The bulk-rock sample analyzed consists mostly of phyllosilicates (94%) with minor contents of
quartz (3%), potassium feldspar (3%) and rutile (traces) as is shown in Table 4.1. Within
phyllosilicates predominate illite-mica and mixed-layer illite-smectite, in traces (<5%) chlorite,
kaolinite and pyrophyllite were also identified (Fig. 4.2). The mixed-layer illite-smectite showsa
high degree of ordering (R1) with sharp 001 and 002 reflections at 24.5 Å and 12.5 Å respectively,
suggesting a rectorite-like clay mineral. A broad reflection was observed for illite -micawithtwo
main maximum at 10.11 Å and 9.98 Å indicating coexistence of NH 4+-rich illite and common illite
(Daniels and Altaner, 1990). In randomly oriented samples the value of d (060) reflection is 1.49-
1.50 Å indicating dioctahedral clay minerals.

Figure 4.2 X-ray diffraction pattern showing the bulk mineralogy of sample CH-1. (I/S) illite-
smectite mixed layer. (Ilt) illite. (Prl) pyrophyllite. (Kln) kaolinite, (Chl) chlorite. (Qz) quartz. (Fsp)
K-feldspar.

106
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

Table 4.1 Characterization of black shale (CH-1).

Clay minerals (wt %)* 94

Non-clay minerals (wt%)** 6
C (wt%) 8.60
H (wt%) 1.23
N (wt%) 0.64
S (wt%) 0.09
2
Specific surface BET (m /g) 34
t-plot external surface (m2/g) 21
t-plot micropores surface (m 2 /g) 12
Micropore average size (Å) 7.6
* Illite (K, NH4+)-rectorite > (chlorite-kaolinite-pyrophyllite)
**Quartz, K-feldspar

The oriented aggregates of the clay fraction (<2μm) corroborate the above mentioned
mineralogy highlighting the swelling under ethylene glycol solvation of the regular mixed-layer
illite-smectite toward 13 and 27 Å, a typical behavior of rectorite (Srodon, 1980).

The identified mineralogical assemblage is similar to that previously reported in shalesfromthe

same lithological formation by Bauluz and Subías, (2010). According to these authors the
occurrence of NH4-rich illite and pyrophyllite result mostly from the circulation of hydrothermal
fluids related to andesitic sills intruding the black shales.

The content of organic carbon (TOC) reaches 8.6% and the concentration of H and N is related
partially with the presence of NH4+ in illite interlayer (See Table 4.1).

The BET specific surface area is 34 m2/g. The t-plot external surface (mesoporosity) can reach
21 m2 /g whereas t-plot micropores surface is 12 m 2/g. The micropore average size is 7.6 Å. The
adsorption isotherm belongs to IUPAC type IV. The hysteresis cycle type H3 is common in sheet
minerals with bendable pores and slit shape.

4.3 ADSORPTION CAPACITY MEASUREMENTS

4.3.1 Sample preparation

Adsorption isotherms were measured on a dry shale sample, which was received in several
pieces of irregular shape; the average size of each piece was about 5 mm. Two grams of sample
were dried under vacuum into the experimental cells at 110 °C during 18 h. This procedure is

107
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

necessary to remove any moisture taken up by the sample. Later, after each adsorption
measurement, the sample was under additional vacuum at 110°C for 18 hours, in order to
withdraw any trace of fluid.

4.3.2 Calorimetric-manometric set up

CO2 adsorption uptake was measured using a manometric-calorimetric device. The differential
heat flux calorimeter used is a Setaram C80 Tian-Calvet (Fig. 4.3).

Figure 4.3 Manometric-calorimetric device (Mouahid et al., 2012b)

108
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

The inner part of the calorimeter include two calorimetric cells (reference and adsorption)
surrounded by a high thermal mass aluminium block. The sample is located in the adsorption
cell (see (3) in Fig. 4.3), which is connected to manometric apparatus inserted in the upper part
of the calorimeter. The same temperature is settled for the calorimeter and the manometric
system allowing isothermal conditions in both parts of the coupled apparatus. The main added
value of this apparatus is the simultaneous determination of the differential heat of adsorption
which is a direct measurement of the adsorbate/adsorbent interactions. Note thatthe reference
cell is used to compensate the heat flux in a blank experiment. A schematic diagram of the set-
up is provided in Fig. 4.3.

4.3.3 Co-adsorption CO2-CH4 setup

For the mixture, different pieces of the same sample we re used. The equimolar CO2-CH4mixture
adsorption was measured using a device built for this purpose as is represented in Fig. 2.5. The
principle remains identical to the pure component adsorption one (Ghoufi et al., 2009). Itallows
the measurements of isotherm mixture adsorption in a range of pressures from 0 to 3 MPa and
for temperatures between 303.15 and 423.15 (K). A circulation pump is used to homogenize the
mixture and reduces the time needed to reach equilibrium. The manometric system is coupled
with a gas chromatograph (Agilent) which allows the determination of the gas mole fraction of
each component in the mixture. The pressure transducer is also a MKS Baratron type 121A. To
guarantee the isothermal conditions is used a heater wire controlled by a PID regulator
(Eurotherm 3208); five thermocouples (type K) were installed in different parts of the circuit to
ensure that there is not a temperature gradient during the measurements. This instrumenthas
been detailed in a previous work (Pino et al., 2014).

4.3.4 Pure components (CO2 and CH4) adsorption

The sorption isotherms obtained by manometric measurements are excess adsorption
isotherms. A mass of sample (about 2 gr) is placed into the adsorption cell (see (3) in Fig 4.3)
under vacuum. After, a quantity of gas is introduced in the dosing volume (see (2) in Fig. 4.3.) of
known volume. The initial pressure P 1 and temperature T1 are measured after reaching thermal
and mechanical equilibria. These measurements provide the initial mole number of gas n 1.
Then, an expansion is performed into the whole system (dosing volume and adsorption cell).As
in the first step, the pressure P 2 and temperature T2 (T2=T1 ) are measured after equilibrium and

109
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

the mole number remaining in the gas phase n 2 is estimated. The adsorbed mole corresponds
to (n1 –n2). The isotherm is described repeating the same operations through an accumulative
process. A previous measuring cell volume determination is done using Helium expansion. The
molar volumes involved in the amount adsorbed calculations were determined thanks to the
NIST data (Span and Wagner, 2003, 1996) at the considered experimental conditions.

CO2 adsorption measurement was performed at 323K, and pressures in a range from
atmospheric pressure to 3 MPa. We also reproduced the CH 4 adsorption isotherm, already
studied in a previous work (Pozo et al., 2017) in which the effect of thermal events on various
shales samples was depicted.

Fig. 4.4 displays the pure CH4 and CO2 isotherms. These results, listed in table 4.2, shows that
the CH4 and CO2 adsorption isotherms exhibit Langmuir type I behaviour, which is typical for
microporous materials. This is consistent with the fact that adsorption occurs mainly onto
kerogen. CO2 adsorption is about three times larger than CH4. At low pressures, the slope ofCO2
isotherm is higher than CH4 . This slope directly correlated to the Henry constant law (K H) isafirst
indicator of the higher affinity of CO 2 with kerogen. With increasing pressure the slope decrease
as the shale sample approaches saturation, in the case of methane the decrease ads orptionrate
is faster than CO2 which is consistent with the higher CO 2 uptake. The values of CO2 uptake are
very similar to few values reported in the literature (Chareonsuppanimit et al., 2012; Hellerand
Zoback, 2014).

Table 4.2 Adsorption data for CO2 and CH4 pures

CH4* CH4 (this work) CO2

Pressure nexc Pressure nexc Pressure nexc
(MPa) (mol/Kg) (MPa) (mol/Kg) (MPa) (mol/Kg)
0.27 0.0372 0.31 0.0488 0.28 0.1812
0.65 0.0736 0.70 0.0805 0.66 0.2499
1.06 0.0964 1.11 0.0933 1.03 0.2898
1.49 0.1124 1.57 0.1125 1.41 0.3156
1.85 0.1185 1.96 0.1198 1.79 0.3338
2.20 0.1218 2.33 0.1239 2.19 0.3480
2.57 0.1245 2.70 0.1256 2.61 0.3593
2.93 0.1254 3.07 0,1257 2.97 0.3648
*Pozo et al., 2017

110
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

0.4

0.3
n exc (mol/Kg)

0.2

0.1

0
0 0.5 1 1.5 2 2.5 3 3.5
Pressure (MPa)
Fit CH4 from Pozo CO2 CH4 this work

Figure 4.4 Excess sorption isotherms of pure components and with fitted three parameters
function

As in chapter 3, to rationalize these results, the isotherm data were correlated by a Langmuir
model of three parameters (using equation 3.1), and a standard deviation was calculated
according to the equation 3.2.

The Langmuir sorption model, developed for low pressures, represents, however, a reasonable
approximation of the measured excess sorption isotherms and can thus be used as a fitting
function. The main concern to apply this approach is the estimation of the maxi mum Langmuir
capacity, which provide a value of the CO 2 adsorption capacity, in the case of this study 0.435
(mol/kg). In this work, for  ads we used 1027 Kg/m3 for CO2 (Gensterblum, et al., 2013). Fitting
parameters are shown in Table 4.3, and in Fig. 4.4 can be observed the isotherm obtained with
these values. Note that the values obtained agree with literature values obtained for kerogen
type III (Zhang et al., 2012, Gasparik et.al. 2014a)

Table 4.3 Parameters for Langmuir Model with  ads fixed

Component pL (MPa) nL (mol/kg)  ads (Kg/m3) n

Methane* 1.00 0.222 423 0.0276
Carbon dioxide 0.45 0.435 1027 0.0025
*Pozo et al., 2017

111
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

4.3.5 Differential heat of adsorption

The calorimetric–manometric allows a simultaneous measurement of the differential enthalpy
of adsorption when describing the isotherms. The values obtained for methane and carbon
dioxide are reported in the Table 4.4.

Table 4.4 Differential enthalpy of adsorption H ads (KJ/mol)

CH4 (Pozo et al., 2017) CH4 (this work) CO2
Pressure H ads Pressure H ads Pressure H ads
(MPa) (KJ/mol) (MPa) (KJ/mol) (MPa) (KJ/mol)
0.10 28 0.31 25 0.28 33
0.55 35 0.70 35 1.03 44
0.90 30 1.11 28 1.41 36

In Fig. 4.5, are shown the values of differential enthalpy of adsorption, ∆Hads. The average value
for methane is 30 kJ/mol and for carbon dioxide is about 40 kJ/mol. As expected, these values
are consistent with physisorption, and also highlights the better affinity of the supercritical CO2
with the shale sample. The enthalpy profiles –nearly constant- also suggest an interaction with
a homogeneous adsorbent. This behavior confirms that both CO 2 and CH4 are mostly adsorbed
by the organic phase, namely the kerogen.

50

40
H ads (KJ/mol)

30
CH4 Pozo et al.
20 CH4 this work
CO2
10

0
0 0.5 1 1.5
Pressure (Mpa)

Figure 4.5 Heat of Adsorption

112
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

4.3.6 Equimolar mixture adsorption

The procedure is analogous for pure gas. A mass of the sample ( about 2 gr) is placed into the
adsorption cell (see Fig. 2.5) of known volume, under vacuum. The accessible volume is
determined by successive Helium expansions after what shale outgassing is carried out. Gas
mixture, which is previously analysed using the chromatograph, is introduced in the dosing cell
without going through the adsorption cell (see Fig. 2.5) until equilibrium is reached (P 1 and T1
are measured and initial mole number of gas n 1 is calculated). In the next step, the gas expands
on the whole system (dossing cell and adsorption cell, of known volumes). When both thermal,
mechanical equilibria and homogenization are reached (checked by constant values of both
pressure and composition) values of P 2 and T2 (T2=T1) are measured and the mole number
remaining in gas phase n 2 is estimated. A discrete quantity of gas (0.5 cm 3) is extracted from the
isolated dosing volume and analysed with the chromatograph. This measure provides the
composition of the gas phase from which can be deduced the composition of the adsorbed
phase. The total amount adsorbed is calculated by a mass balance (n 1-n2 ). This procedure used
to calculate the amounts of each individual components adsorbed was detailed in Pino et al.
(2014). Basically, we wait for the equilibrium about 60 minutes and we perform several analyses
to check reproducibility. The same operation is repeated through an accumulati ve process to
describe the isotherm. The uncertainty on the adsorbed amount is estimated to be less than
3%.

Adsorption isotherm was measured in the same thermodynamics conditions that for the pure
compounds. All the data are shown in Fig 4.6 and summarized in Table 4.5. As expected, pure
CO2 in gas mixture is much preferentially adsorbed than CH 4. When one compare these
isotherms with those of the single gas component, it is relevant that the CO 2 adsorbed amount
is affected by the presence of CH4 in the gas phase but the CH4 amount decreases more than
CO2. All the data for the pure component in gas mixture are also summarized in Table 4.5.
Additionally, the selectivity of CO 2 over CH4 was calculated from equation (4.1) and reported in
Fig. 4.7 as a function of pressure:

y1
⁄y
Selectivity= x1 2 (4.1)
⁄x2

In this equation y is mol fraction in the bulk phase and x is mol fraction in the adsorbed phase.
(1 for CH4 and 2 for CO2 ). This selectivity exhibits a non -monotonic evolution as a function of

113
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

the pressure, with a significant increase at low pressure followed by a maximum reachingavalue
of 4.7 at 1.5 (MPa). This profile can be interpreted as follows: CO 2 reaches the saturation
capacity whereas there is still a slight increase in CH4 adsorption at intermediate and high
pressures. A low and medium pressures, it is clearly stated that the CO 2 molecules preferentially
occupy the adsorption sites. This affinity between CO 2 and kerogen is consistent both with the
obtained enthalpy values and with the slope of the isotherms at very low coverage in the Henry
zone.

Even if the experiment were not carried out in conditions compatible with in-well conditions,
these results confirms that shale may be a promising candidate for both the gas recovery and
the CO2 storage.

0.4

0.3
(mol/Kg)

0.2
n exc

0.1

0
0 0.5 1 1.5 2 2.5 3
Pressure (MPa)
ADSORCION TOTAL CO2 mezcla CH4 mezcla CH4 puro CO2 PURO

Figure 4.6 Equimolar mixture and individual sorption

114
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

Table 4.5 Adsorption data for mixture and individual component of the equimolar mixture CH4-
CO2

Pressure nexc mixture nexc CH4 nexc CO2 y2 y1 x2 x1

(MPa) (mol/Kg) (mol/Kg) (mol/Kg) CO2 CH4 CO2 CH4
0.40 0.1374 0.0630 0.0744 0.4585 0.5415 0.5415 0.4585
0.75 0.2161 0.0604 0.1557 0.4782 0.5218 0.7206 0.2794
1.08 0.2680 0.0576 0.2104 0.4885 0.5115 0.7850 0.2150
1.39 0.3077 0.0577 0.2500 0.4936 0.5064 0.8124 0.1876
1.71 0.3322 0.0609 0.2713 0.4971 0.5029 0.8167 0.1833
2.08 0.3592 0.0714 0.2878 0.4990 0.5010 0.8013 0.1987

4
Selectivity

0
0 0.5 1 1.5 2 2.5
Pressure (MPa)

Figure 4.7 Selectivity of CO2 over CH4

4.4 CONCLUSIONS
On the basis of experimental measurements, the following conclusions can be addressed:

 Adsorption behavior of CO 2 and CH4 exhibit type I Langmuir. CO 2 uptake is clearlygreater

than CH4.

 Representative fits of excess sorption isotherms were obtained using a 3-parameter

Langmuir-based excess sorption function. The sorption isotherms at different
temperatures can be adequately represented assuming constant Langmuir sorption

115
Chapter 4. Characterization and Adsorption Capacity of Shale to Methane, Carbon Dioxide and
to an Equimolar Mixture CH4/CO2

capacity (nL) and sorbed phase density (ρ ads ) and temperature-dependent Langmuir
pressure (pL).

 The differential heat of adsorption is nearly constant for CO 2 and CH4. This is
representative of an interaction with a homogeneous substrate. The highest values
obtained for CO2 support the highest affinity between CO 2 and kerogen.

 The investigation of the equimolar mixture CO 2 /CH4 supports the higher selectivity of
CO2 over CH4. This result is consistent with the values obtained for the differential heat
of adsorption.

116
117
118
 CHAPTER 5
NEW ADSORBENTS FOR CH4/CO2
SEPARATION

119
120
 Chapter 5. New Adsorbents for CH4 /CO2 separation

This chapter furthers the study devoted to the understanding of the adsorption mechanism on
shales gas samples. One of the main challenges today is to reduce emissions of so-called
greenhouse gases, which are a sub product of the use of fossil fuels and landfills and shale gas.
We try to contribute on the development of new adsorbents with the objective to upgrade CH4
natural gas from the complete removal of CO 2 molecules. In Chapter 4, we explored the way to
store CO2 into depleted shale gas through the study of CO 2 /kerogen interaction. In the present
chapter, we try to contribute on the development of new adsorbents with the objective to
upgrade CH4 natural gas from the complete removal of CO 2 molecules.

5.1 INTRODUCTION

Global warming is considered to be caused by the emissions of greenhouse gas es (GHG).Toface

this problem Carbon Capture and Storage (CCS) techniques are being proposed as feasible
technologies to effectively minimize CO 2 emissions (Metz et al., 2005; United Nations, 1998).

The commercially available techniques for CO 2 capture are based on absorption with liquid
amine compounds such as monoethanolamine (MEA) or diethanolamine (DEA). Howeverthese
absorption process present as disadvantages the liquid mixture corrosivity, the amine waste
during operation, and the high energy cost of regeneration process (DOE/OS-FE, 1999;
Tontiwachwuthikul et al., 1991). As a consequence, in the last decade has been an increasing
interest in developing efficient selective adsorbents for CO 2 capture as an alternative to liquid
absorption.

Adsorption is becoming one of the most used processes for gas separation, and for this reason
new adsorption materials are being studied and developed because of their ability for
separating, in a selective way, specific compounds such as methane and carbon dioxide from
complex mixtures (Goetz, et al., 2006 and Bao, et al., 2011); and because the energy
consumption is lower than traditional separation techniques (Santori et al. 2014 and Pino, et al.
2014). This method is based on the selective adsorption of one component of a gas mixture at
relatively high pressures and its release upon decreasing pressure. This process requiresahighly
selective adsorbent with a high CO 2 capacity but not too high affinity for the others molecules.
Otherwise, the regeneration step can become negative in economic terms. The adsorbent
performance indicator is clearly measured by three parameters: C0 2 adsorption capacity, the

121
 Chapter 5. New Adsorbents for CH4 /CO2 separation

selectivity and the heat of adsorption which is a measure of the required energy to release CO2
molecules.

Thus, zeolithes, activated carbons, mesostructured silicas and silicas nanoparticles, and metal-
organic frameworks have been largely investigated from the removal of carbon dioxide.ForCO 2
removal, mesostructured supports are functionalized with organic molecules containing basic
amino groups, which act as specific sites for CO 2 capture by chemical adsorption (Knowles, et
al., 2006; Calleja et al., 2011; Xu, et al., 2002; Sanz et al., 2010; Rosenholm et al., 2006; Yue et
al., 2008, 2006).

In this context, the present study focuses on synthesized silica nanoparticles (natives and
functionalized with amines). We performed adsorption measurements of CO2 and CH4 pure and
in an equimolar mixture at 50°C in a pressure range between 0.1 and 3 MPa. The chapter is
organized as follows:

 Characterization of the sample.

 Measurement of adsorption capacity for CO 2, CH4 and for an equimolar mixture ofthese
components.
 Measurement of the heat of adsorption by calorimetry.
 Determination of selectivity of CO 2/CH4

5.2 SAMPLE CHARACTERIZATION

The samples were provided by Prof. Veronica Salgueriño (Departamento de Fisica Aplicada-
University of Vigo (Spain).

5.2.1 Native silica samples

To synthesize the silica nanoparticles, a widely known modification of the typical Stöbermethod
was used. In order to obtain the nanoparticles with the narrowest size distribution all reactions
were tempered at 25 ºC and performed under the same stirring conditions.

122
 Chapter 5. New Adsorbents for CH4 /CO2 separation

5.2.2. Functionalized samples

Native silica nanoparticles were functionalized with –NH2 groups taking advantage of the
hydrolysis and condensation of the 3-aminopropyltriethoxysilane (APS) at the surface. This
process leads to the covalent bonding of the Si atom of the APS organosilane molecule to the
surface of the silica nanoparticles, terminated by silanol groups that can react with various
coupling agents, or analogously, with Si atoms distributed all over the silica matrix during its
condensation. The process can be proven and quantified by means of fluorescence spectra.

All the samples (native and functionalized) were checked by Transmission electron microscopy
(TEM), in order to appreciate their morphology and the average size distributions (fitted to
Gaussian). Figure 5.1 includes TEM images of the nanoparticles for the two samples, all of them
with a rather perfect spherical shape, and with average size distributions. Table 5.1 shows BET
and APS information.

Table 5.1 Samples Characterization

Sample BET (m2 /g) APS (mol/g silica)

Native 286 0
Functionalized 203 0.4 E-04

123
 Chapter 5. New Adsorbents for CH4 /CO2 separation

(a)

(b)

Figure 5.1 (a) native silica nanoparticles (b) Functionalized silica nanoparticles

124
 Chapter 5. New Adsorbents for CH4 /CO2 separation

5.3 ADSORPTION CAPACITY IN NATIVE SILICA SAMPLES

5.3.1 Pure components (CO2 and CH4)

CO2 and CH4 adsorption measurements were performed at 323K, and pressures in a range from
atmospheric pressure to 3 MPa. Fig. 5.2 displays the pure CH4 and CO2 isotherms. Two
independent sets of measurements were performed to confirm the reproducibility of our
procedures. The replicate measurements agree within expected experimental uncert ainty
claimed to be less than 3%. These results, listed in table 5.2, shows the CH4 and CO2 adsorption
isotherms exhibit Langmuir type I behavior. Further, the adsorption of CO 2 is about twice larger
than that of methane. As summarized in Table 5.2, the CO2 adsorption capacity increases up to
2.55 mol kg−1 at 3 MPa and 323.15 K, whereas CH4 adsorption capacity is nearly 0.52 mol kg−1 in
similar conditions. This total amount is still significantly smaller compared to the bestadsorbent
material reported in the literature (for example, 25 mol kg−1 for activated carbon) (Millwardand
Yaghi, 2005). Anyway, normalized to the BET, this CO2 adsorption uptake is a very promising
approach (Sanz et al., 2012).

3.0

2.5
n exc (mol/Kg)

2.0

1.5
CO2 pure
CH4 pure
1.0

0.5

0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

P (Mpa)

Figure 5.2 Adsorption capacity for pure CH4 and CO2

125
 Chapter 5. New Adsorbents for CH4 /CO2 separation

Table 5.2 Adsorption data for pure CO2 and CH4

CH4 CO2
Pressure nexc Pressure nexc
(MPa) (mol/Kg) (MPa) (mol/Kg)
0.25 0.0801 0.24 0.3714
0.61 0.2019 0.58 0.8117
0.98 0.3072 0.98 1.2267
1.37 0.3956 1.41 1.5503
1.72 0.4717 1.81 1.8566
2.08 0.5420 2.25 2.1412
2.50 0.6034 2.65 2.3765
2.96 0.6551 3.06 2.6114

As was mentioned in Chapters 3 and 4, to rationalize these results, the isotherm data were
correlated by a Langmuir model of three parameters, by means of Equation 3.1 and the standard
deviation was calculated with Equation 3.2.

Fitting parameters are shown in Table 5.3, and in Fig. 5.3 can be observed the isotherm obtained
with these values.

2.5

2
n exc (mol/Kg)

1.5 CO2
CH4
1 Fit

0.5

0
0 0.5 1 1.5 2 2.5 3 3.5
Pressure (MPa)

Figure 5.3 Excess sorption isotherms of pure components and with fitted three parameters
function

126
 Chapter 5. New Adsorbents for CH4 /CO2 separation

Table 5.3 Parameters for Langmuir Model with  ads fixed

Component pL (MPa) nL (mol/kg)  ads (Kg/m3) n

Methane 4.63 1.775 421 0.002
Carbon dioxide 4.23 6.488 1027 0.010

5.3.2 Differential heat of adsorption

The calorimetric–manometric device allows a simultaneous measurement of the differential

enthalpy of sorption when describing the isotherms. The values obtained are reported in the
Table 5.3.

Table 5.4 Differential enthalpy of adsorption H ads (KJ/mol)

CH4 CO2
Pressure H ads Pressure H ads
(MPa) (KJ/mol) (MPa) (KJ/mol)
0.25 22 0.24 24
0.61 37 0.58 19
0.98 16 0.98 22
1.37 15 1.41 24
1.72 17 1.81 21
2.08 21 2.25 24
2.50 25 2.65 21
2.96 32 3.06 21

In Fig. 5.4, are shown the values of differential enthalpy of adsorption, ∆Hads. The average value
for methane is 21 kJ/mol and for carbon dioxide is about 22 kJ/mol, these values are consistent
with physisorption. The enthalpy profiles –nearly constant- also suggest an interaction with a
homogeneous adsorbent.

127
 Chapter 5. New Adsorbents for CH4 /CO2 separation

30

20
H ads (KJ/mol)

CO2
10
CH4

0
0.0 1.0 2.0 3.0 4.0
P (MPa)

Figure 5.4 Heat of Adsorption

5.3.3 Equimolar mixture CH4/CO2 adsorption

Adsorption isotherm was measured in the same thermodynamics conditions that for the pure
compounds. All the data are shown in Fig 5.5 and summarized in Table 5.5. Data for the pure
component in gas mixture are also summarized in Table 5.5. The individual components
adsorption were estimated through the methodology presented in the chapter II. As expected,
CO2 is much preferentially adsorbed than CH4. When one compares these isotherms with those
of the single gas component, it is relevant that the CO 2 adsorbed amount is less affected by the
presence of CH4 in the gas phase, whereas the CH4 amount more significantly decreases.
Additionally, the selectivity of CO 2 over CH4 was calculated from equation (4.1) and reported in
Fig. 5.6 as a function of pressure.

The observed selectivity which is increasing with pressure with a maximumn nearly 7 is a
promising result. Anyway, this value remains significantly lower than the highestvaluesreported
(Furmaniak et al., 2013; Denayer et al., 2009; Othman, 2009).

128
 Chapter 5. New Adsorbents for CH4 /CO2 separation

3.0

2.5
Total Adsorption

2.0 CO2 in mixture

n exc (mol/Kg)

CH4 in mixture
1.5 CO2 pure

CH4 pure
1.0

0.5

0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

P (Mpa)
Figure 5.5 Equimolar mixture and individual sorption

Table 5.5 Adsorption data for mixture and individual components of the equimolar mixture
CH4-CO2

Pressure nexc mix. nexc CH4 nexc CO2 y2 y1 x2 x1

(MPa) (mol/Kg) (mol/Kg) (mol/Kg) CO2 CH4 CO2 CH4
0.32 0.607 0.161 0.447 0.466 0.534 0.735 0.265
0.62 1.024 0.199 0.824 0.483 0.517 0.810 0.190
0.92 1.302 0.202 1.099 0.489 0.510 0.844 0.156
1.28 1.615 0.224 1.390 0.494 0.506 0.861 0.139
1.69 1.877 0.272 1.606 0.497 0.503 0.855 0.145
2.20 2.269 0.280 1.989 0.499 0.501 0.877 0.123

129
 Chapter 5. New Adsorbents for CH4 /CO2 separation

6
Selectivity

0
0 0.5 1 1.5 2 2.5
P (Mpa)

Figure 5.6 Selectivity of CO2 over CH4

5.4 ADSORPTION CAPACITY IN FUNCTIONALIZED SILICA NANOPARTICLES

5.4.1 Pure components (CO2 and CH4)

CO2 and CH4 adsorption measurements were performed at 323K, and pressures in a range from
atmospheric pressure to 3 MPa. Fig. 5.7 displays the pure CH4 and CO2 isotherms; data are
summarized in Table 5.6. CO2 isotherm shows a similar behavior as in the case of native silica
nanoparticles both at low and high pressure. The CO2 adsorption uptake, 2.34 mol Kg-1 at 3MPa,
is lower than in the case of native nanoparticles silica. This seems to be first indicator that the
amino groups are not sufficiently contributing to the chemisorption process. And the decrease
of CO2 uptake can be explained by the diminution of the BET area due to the functionalization
process.

130
 Chapter 5. New Adsorbents for CH4 /CO2 separation

2.5

2.0
n exc (mol/Kg)

1.5

CO2 pure
1.0
CH4 pure

0.5

0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
P (Mpa)

Figure 5.7 Adsorption capacity for pure CH4 and CO2

Table 5.6 Adsorption data for CO2 and CH4 pure

CH4 CO2
Pressure nexc Pressure nexc
(MPa) (mol/Kg) (MPa) (mol/Kg)
0.29 0.1178 0.29 0.4174
0.65 0.2595 0.59 0.7501
1.02 0.3899 0.97 1.0785
1.39 0.5063 1.37 1.3789
1.77 0.6135 1.77 1.6437
2.19 0.7139 2.17 1.8970
2.60 0.7990 2.57 2.1292
3.00 0.8810 2.95 2.3361

The isotherm data were correlated by a Langmuir model of three parameters (equation 3.1) and
standard deviation was calculated too (equation 3.2). Fitting parameters are shown in Table 5.7,
and in Fig. 5.8 can be observed the isotherm obtained with these values.

131
 Chapter 5. New Adsorbents for CH4 /CO2 separation

Table 5.7 Parameters for Langmuir Model with  ads fixed

Component pL (MPa) nL (mol/kg)  ads (Kg/m3) n

Methane 6.51 2.923 421 0.001
Carbon dioxide 4.32 5.978 1027 0.010

2.5

2.0
n exc (mol/Kg)

1.5
CO2

1.0 CH4
Fit
0.5

0.0
0.0 1.0 2.0 3.0 4.0
Pressure (MPa)

Figure 5.8 Excess sorption isotherms of pure components and with fitted three parameters
function

5.4.2 Differential heat of adsorption

The calorimetric-manometric device allows a simultaneous measurement of the differential
enthalpy of sorption when describing the isotherms. The values obtained are reported in the
Table 5.8.

In Fig. 5.9, are shown the values of differential enthalpy of adsorption, ∆Hads. The average value
for CH4 is 30 KJ/mol and for carbon dioxide is about 28 KJ/mol. In both cases, these values are
higher than for native particles, but not as high as expected for chemisorption process. In the
case of CH4, the high values can be explained by the uncertainties due to the very lowadsorption
quantities. The values obtained for CO 2 clearly confirms that the degree of amino
functionalization is not enough for CO 2 capture through the chemisorption process.

132
 Chapter 5. New Adsorbents for CH4 /CO2 separation

Table 5.8 Differential enthalpy of adsorption H ads (KJ/mol)

CH4 CO2
Pressure H ads Pressure H ads
(MPa) (KJ/mol) (MPa) (KJ/mol)
1.39 25 1.37 27
1.77 27 1.77 26
2.19 38 2.17 29
2.60 25 2.57 30
3.00 35 2.95 29

50

40
H ads (KJ/mol)

30 CO2

CH4

20

10

0
0.0 1.0 2.0 3.0 4.0
P (MPa)

Figure 5.9 Heat of Adsorption

5.4.3 Equimolar mixture adsorption

Adsorption isotherm was measured in the same thermodynamics conditions that for the pure
compounds. All the data are shown in Fig 5.10 (a) and (b) and summarized in Table 5.9. As
expected, pure CO2 in gas mixture is preferentially adsorbed than CH4. Data for the pure
component in gas mixture are also summarized in Table 5.9. Additionally, the selectivity of CO2
over CH4 was calculated from equation (4.1) and reported in Fig. 5.11 as a function of pressure.
As a conclusion the selectivity is not improved with functionalize samples.

133
 Chapter 5. New Adsorbents for CH4 /CO2 separation

2.5

2
n exc (mol/Kg)

1.5 Total adsorption

CO2 mixture

1 CH4 mixture
CO2 pure
CH4 pure
0.5

0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
P (Mpa)

Figure 5.10 (a) Equimolar mixture and individual sorption

2.5

2 CH4 functionalized
n exc (mol/Kg)

CH4 natives
1.5
CO2 functionalized

1 CO2 natives

0.5

0
0 0.5 1 1.5 2 2.5 3
Pressure (MPa)

Fig. 5.10 (b) Individual components adsorption in native and functionalized nanoparticles

134
 Chapter 5. New Adsorbents for CH4 /CO2 separation

Table 5.9 Adsorption data for mixture and individual component of the equimolar mixture CH4-
CO2

Pressure nexc mix. nexc CH4 nexc CO2 y2 y1 x2 x1

(MPa) (mol/Kg) (mol/Kg) (mol/Kg) CO2 CH4 CO2 CH4
0.30 0.577 0.217 0.360 0.468 0.532 0.624 0.376
0.63 0.887 0.162 0.725 0.480 0.520 0.817 0.183
1.13 1.286 0.175 1.111 0.491 0.509 0.864 0.136
1.48 1.549 0.188 1.361 0.495 0.505 0.879 0.121
1.86 1.795 0.240 1.555 0.498 0.502 0.866 0.134
2.22 2.062 0.303 1.759 0.500 0.500 0.853 0.147

6
Selectivity

4
Natives
Functionalized

0
0 0.5 1 1.5 2 2.5
P (Mpa)

Figure 5.11 Selectivity of CO2 over CH4

5.5 IDEAL ADSORBED SOLUTION THEORY

The ideal adsorbed solution theory (IAST) developed by Myers and Prausnitz in 1965, can be
used to predict adsorption equilibrium of gas mixtures. It allows to check the consistency
between the pure compounds and the binary mixture data. This theory is based on the
assumption that the adsorbed phase is an ideal solution. So, the equilibrium between the two

135
 Chapter 5. New Adsorbents for CH4 /CO2 separation

dimensional adsorbed phases is described by means of an equation analogous to Raoult’s Law

for vapor-liquid equilibrium.

P. 𝑦𝑖 =𝑥𝑖 .𝑃𝑖0 (5.1)

Where Pi0 is the pressure of the pure component i at the same spreading pressure ofthe mixture;
P is the total adsorptive pressure; and yi and x i are the mole fractions of component i in the gas
phase and in the adsorbed phase, respectively. The superscript 0 denotes pure components.

The spreading pressure cannot be measured itself but it can be derived using a modified Gibbs-
Duhem equation under isothermal conditions:

𝑃0 𝑛𝑖 ( 𝑃𝑖)
π= ∫0 𝑖 . 𝑑𝑃𝑖 (5.2)
𝑃𝑖

where ni(Pi) is the adsorbed amount of component i at adsorptive pressure Pi. The term ni(Pi) is
fitted by the Freundlich equation on the basis of the experimental data of each individua l
component.

Using an iterative procedure, it is possible to estimate the composition of the adsorbed phase,
using the following conditions:

∑ 𝑥𝑖 = 1 (5.3)

∑ 𝑦𝑖 = 1 (5.4)

When the value of x i is known, it is possible to calculate the solid phase adsorbed amount of
each component, by means of:

𝑛𝑖 =𝑛𝑡. 𝑥 𝑖 (5.5)

In figure 5.12 are reported the amount adsorbed of CH4 and CO2 obtained experimentally and
using the IAST method.

136
 Chapter 5. New Adsorbents for CH4 /CO2 separation

2.5
(a)
2.0
n.exc(mol/Kg)

1.5

1.0

0.5

0.0
0.0 0.5 1.0 1.5 2.0 2.5
P (MPa)

2.0
(b)

1.5
n.exc(mol/Kg)

1.0

0.5

0.0
0.0 0.5 1.0 1.5 2.0 2.5
P (MPa)
CO2-IAST CO2-EXP CH4-IAST CH4-EXP

Figure 5.12 Sorption of CH4 and CO2 in (a) functionalized and (b) native particles. Comparison
between experimental data and IAST method.

137
 Chapter 5. New Adsorbents for CH4 /CO2 separation

5.6 CONCLUSIONS

On the basis of experimental measurements, the following conclusions can be addressed:

 Adsorption behavior of CO 2 and CH4 exhibit type I Langmuir. CO2 uptake is clearlygreater
than CH4.

 CO2 uptake obtained with the native sample is a very promising result for the use of
silica nanoparticles for CO 2 capture.

 The degree of functionalization is not sufficient for the chemisorpti on process.

 The differential heat of adsorption is nearly constant for CO 2 and CH4. This is
representative of an interaction with a homogeneous substrate. The valuesobtainedfor
CO2 support the fact that there is no chemisorption with the functionalized sample.

138
139
140
GENERAL CONCLUSIONS

141
142
 GENERAL CONCLUSIONS

In this thesis, we studied, from an experimental point of view, the adsorption phenomenon in
different systems of interest in the context of petroleum engineering.

In the case of Colombian shale gas reservoir, we studied five samples from three exploratory
wells. A characterization was carried out in the Universidad Autonoma de Madrid. To the best
of our knowledge our gas adsorption measurements are the first one made to gas shales from
MMV basin. Results support that TOC is the primary factor controlling the methane adsorption
capacity. Additionally, it was found a moderate positive relationship between TOC and SSAand
between methane adsorption capacity and SSA. Thermal maturity does not show a significant
effect in our samples unlike others studies reported.

Besides Colombian shales, we made further experimental measurements to a previously shale

sample studied by Pozo et al., (2017). This study made it possible to obtain original data for CO2
heat of adsorption on kerogen and for the selectivity of the equimolar mixture CO2/CH4 on
kerogen. The results shows that CO 2 is adsorbed about three times more than methane, in the
case of pure gases adsorption and about four times for the equimolar mixture. These resultsare
of great interest from an industrial point of view, because they mean that this shale could be a
candidate for CO2 injection as recovery method; and for CO 2 storage when it was depleted. From
a theoretical point of view, adsorption heat measured on kerogen represents a macroscopic
magnitude to test the different models of kerogen developed by molecular dynamicsorinother
modeling approaches.

Finally, we explored the separation of CO 2 from an equimolar CH4/CO2 mixture usingadsorbents

of silica nanoparticles. Although the results of adsorption capacity of CO 2 obtained from the
native silica are very promising, the functionalization process has to be improved. In any case,
it is a work perspective for the future.

143
144
145
146
CONCLUSIONS GÉNÉRALES

147
148
 CONCLUSIONS GÉNÉRALES

Dans cette thèse, nous avons étudié, d'un point de vue expérimental, le phénomène
d'adsorption dans différents systèmes d'intérêt dans le contexte de l'ingénierie pétrolière.

Dans le cas du réservoir de gaz de schiste colombien, nous avons étudi é cinq échantillons
provenant de trois puits exploratoires. Une caractérisation a été réalisée à l'Université
Autonome de Madrid. À notre connaissance, nos mesures d'adsorption de gaz sont les
premières effectuées sur les gaz de schistes de la MMV basin. Les résultats confirment que
le TOC est le principal facteur qui contrôle la capacité d'adsorption du méthane. De plus, il
a été trouvé une relation positive modérée entre le TOC et la surface spécifique (SSA) et
entre la capacité d'adsorption du méthane et la surface spécifique. La maturité thermique
ne montre pas d'effet significatif dans le cas de nos échantillons contrairement à d’autres
études de la littérature.

En plus des shales colombiens, nous avons effectué d'autres mesures expérimentalessurun
échantillon de schiste précédemment étudié par Pozo et al. (2017). Cette étude a permisde
produire des données originales pour la chaleur d'adsorption du CO 2 sur le kérogène et pour
la sélectivité du mélange équimolaire CO2/CH4 sur le kérogène. Les résultats montrentque
le CO2 est adsorbé environ trois fois plus que le méthane, dans le cas de l'adsorption de gaz
purs et environ quatre fois pour le mélange équimolaire. Ces résultats sont d’un grand
intérêt d’un point de vue industriel, car ils signifient que les schistes pourraient être
candidats à l'injection de CO2 comme méthode de récupération assistée et pour le stockage
de CO2, dans des réservoirs déplétés. D’un point de vue théorique, les mesures de chaleur
d’adsorption mesurées sur le kérogène représentent une grandeur macroscopique afinde
tester les différents modèles de kérogène développés dynamique moléculaire ou dans
d’autres approches de modélisation.

Nous avons enfin exploré une voie pour la séparation du CO2 d'un mélange équimolaire
CH4/CO2 à partir d’adsorbants de nanoparticules de silice. Si les résultats de capacité
d’adsorption du CO2 obtenus à partir de la silice native sont très prometteurs, le processus
de fonctionnalisation reste à améliorer. Il s’agit dans tous les cas d’une perspective de
travail pour le futur.

149
150
151
152
BIBLIOGRAPHY

153
154
 BIBLIOGRAPHY

 Bao, Z., Yu, L., Ren, Q., Lu, X., Deng. S., 2011. Adsorption of CO2 and CH4 on a magnesium-
based Metal-Organic Framework. J. Colloid Interface Sci. 353, 549-556.

 BP Statistical Review of World Energy June 2016.

 Barrero, D., Pardo, A., Vargas, C., Martínez, J., 2007. Colombian Sedimentary Basins:
Nomenclature, Boundaries and Petroleum Geology, a New Proposal. ANH, Bogotá, p.
92.

 Bauluz, B., Subías, I., 2010. Coexistence of pyrophyllite, I-S, R1 and NH4+-rich illite in
Silurian black shales (Sierra de Albarracín, NE Spain): metamorphic vs. hydrothermal
origin. Clay Miner. 45, 383-392.

 Bessières, D., Lafitte, T., Daridon, J-L, Randzio, S.L., 2005. High pressure thermal
expansion of gases: measurements and calibration. Thermochim. Acta. 428, 25-30.

 Bessieres, D., Saint-Guirons, H., Daridon, J.L., Coxam, J.Y., 2000. Apparatus for
Simultaneous Determination of the Densities and Heat Capacities of Liquids and of
Liquids with Dissolved Gas under an Extended Range of Pressure (0.1-100 MPa). Meas.
Sci. Technol. 11, N69.

 Busch, A., Alles, S., Gensterblum, Y., Prinz, D., Dewhurst, D.N., Raven, M.D., Stanjek, H.,
Krooss, B.M., 2008. Carbon dioxide storage potential of shales. Int. J. Greenhouse Gas
Control. 2, 297–308.

 Cao, T., Song, Z, Wang, S., Xia, J., 2015. A comparative study of the specific surface area
and pore structure of different shales and their kerognes. J. Sci. China Earth Sci. 58, 510-
522.

 Carbon sequestration. State of Science. Office of Science and Office of Fossil Energy. US
Department of Energy. DOE/OS-FE. Washington D.C., 1999.

 Calleja. G., Sanz, R., Arencibia, A., Sanz-Pérez, E.S., 2011. Influence of drying conditions
on amine-functionalized SBA-15 as adsorbent of CO2. Top. Catal. 54, 135-145.

 Chalmers, G.R.L. and Bustin, R. M., 2007. The organic matter distribution and methane
capacity of the Lower Cretaceous strata of Northeastern British Columbia, Canada. Int.
J. Coal Geol. 70, 223-239.

 Chalmers, G.R.L. and Bustin, R. M., 2008. Lower Cretaceous gas shales in northeastern
British Columbia, Part I: geological controls on methane sorption capacity. Int. Bull.Can.
Pet. Geol. 56, 1-21

 Charoensuppanimit, P., Mohammad, S. A., Gasem, K.A.M., 2016. Measurements and

modeling of Gas adsorption on Shales. Energy Fuels. 30, 2309-2319.

 Chareonsuppanimit, P., Mohammad, S.A., Robinson, R.L., Gasem, K.A.M., 2012. High-
pressure adsorption of gases on shales: Measurements and modeling. Int. J. Coal Geol.
95, 34–46.

155
 BIBLIOGRAPHY

 Chen, L, Jiang, Z., Liu, K., Ji, W., Wang, P., Gao, F., Hu, T., 2017. Application of Langmuir
and Dubinin-Radushkevich models to estimate methane sorption capacity on twoshale
samples from the Upper Triassic Chang 7 Member in the southeastern Ordos Basin,
China. Energ. Explor. Exploit. 35, 122-144

 Chung, F.H., 1974. Quantitative interpretation of X-ray diffraction patterns of mixtures.

II. Adiabatic principle of X-ray diffraction analysis of mixtures. J. Appl. Crystallogr.7,526-
531.

 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 Geol. 42, 241-71.

 Clarkson C.R., Jensen, J.L., Chipperfield, S., 2012. Unconventional Gas Reservoir
Evaluation: What Do We Have to Consider?. J. Nat. Gas. Sci. Eng. 8, 9–33.

 Collell, J., Galliero, G., Gouth, F., Montel, F., Pujol, M., Ungerer, P., Yiannourakou, M.,
2014. Molecular Simulation and Modelisation of Methane/ethane Mixtures Adsorption
onto a Microporous Molecular Model of Kerogen under Typical Reservoir Conditions.
Microporous Mesoporous Mat. 197, 194–203.

 Collell, J., Ungerer, P., Galliero, G., Yainnourakou, M., Montel, F., Pujol, M., 2009.
Molecular Simulation of Bulk Organic Matter in Type II Shales in the Middle of the Oil
Formation Window. Energy Fuels. 28, 457-466.

 Cui, X., Bustin, A.M.M., Bustin, R., 2009. Measurements of gas permeability and
diffusivity of tight reservoir rocks: different approaches and their applications.
Geofluids. 9, 208-223.

 Curtis, J.B., 2002. Fractured shale-gas systems. AAPG Bull, 86, 1921-1938.

 Curtis, M.E., Cardott, B.J., Sondergeld, C.H., Rai, C.S., 2012. Development of organic
porosity in the Woodford Shale with increasing thermal maturity. Int. J. Coal Geol. 103,
26-31.

 Daniels, E.J., Altaner, S.P., 1990. Clay mineral authigenesis in coal and shale from the
Anthracite region, Pennsylvania. Am. Mineral. 75, 825-39.

 Denayer, J.F.M., Couck, S., Baron, G.V., Gascon, J., Kaptejin, F., 2009. Enhancement of
CO2 and CH4 separation via amino-functionalisation of the MIL-53 metal-organic-
framework. In: Conference proceedings. AIChE annual meeting.

 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.

 Fathi, E., Akkutlu, I.Y., 2014. Multi-Component Gas Transport and Adsorption Effects
during CO2 Injection and Enhanced Shale Gas Recovery. Int. J. Coal Geol. 123, 52–61.

156
 BIBLIOGRAPHY

 Furmaniak, S., Kowalczyk, P., Terzyk, A.P., Gauden, P.A., Harris, P.J.F., 2013. Synergetic
effect of carbon nanopore size and surface oxidation on CO 2 capture from CO2/CH4
mixtures. J Colloid Interface Sci. 397, 144–53.

 Gasparik, M., Bertier, P., Gensterblum, Y., Ghanizadeh, A., Krooss, B.M., Littke, R.,2014a.
Geological Controls on the Methane Storage Capacity in Organic-Rich Shales. Int. J. Coal
Geol. 123, 34–51.

 Gasparik, M., Gensterblum, Y., Ghanizadeh, A., Weninger, P., Kross, B.M., 2015. High
pressure/High temperature Methane-Sorption Measurements on Carbonaceus Shales
by the Manometric Method: Experimental and Data-Evaluation Considerations for
Improved Accuracy. SPE. Journal. 20, 790-809.

 Gasparik, M., Ghanizadeh, A., Bertier, P., Gensterblum, Y., Bouw, S., Krooss, B.M., 2012.
High-Pressure Methane Sorption Isotherms of Black Shales from The Netherlands.
Energy Fuels. 26, 4995–5004.

 Gasparik, M., Rexer, T.F.T., Aplin, A.C., Billemont, P., De Weireld, G., Gensterblum, Y.,
Henry, M, et al., 2014b. First International Inter-Laboratory Comparison of High-
Pressure CH4 , CO2 and C2H6 Sorption Isotherms on Carbonaceous Shales. Int. J.Coal Geol.
132, 131–46.

 Gensterblum, Y., Busch, A., Krooss, B.M., 2014. Molecular concept and experimental
evidence of competitive adsorption of H2 O, CO2 and CH4 on organic material. Fuel. 115,
581-588.

 Gensterblum, Y., Merkel, A., Busch, A., Krooss, B.M., 2013. High-pressure CH4 and CO2
sorption isotherms as a function of coal maturity and the influence of moisture. Int. J.
Coal Geol. 118, 45-57.

 Gensterblum, Y., Van Hemert, P., Billemont, P., Busch, A., Charriére, D., Li, D., Wolf,K-H,
2009. A. European inter-laboratory comparison of high preessure CO 2 sorption
isotherms. I: Activated carbon. Carbon. 47, 2958–2969.

 Ghoufi, A., Gaberova, L., Rouquerol, J., Vincent, D., Llewellyn, P.L., Maurin , G. 2009.
Adsorption of CO2, CH4 and their binary mixture in Faujasite NaY: A Combination of
molecular simulations with gravimetry-manometry and microcalorimetry
measurements. 2009. Microporous Mesoporous Mat. 119, 117-128.

 Goetz, V., Pupier, O., Guillot, A., 2006. Carbon dioxide – methane mixture adsorptionon
activated carbon. Adsorption. 12, 55-63.

 Gorin, G.E., Feist-Burkhardt, S., 1990. Organic facies of Lower to Middle Jurassic
sediments in the Jura Mountains, Swizerland. Rev. Palaeobot. Palynol. 65, 349-355.

 Guo, S. 2013. Experimental Study on isothermal adsorption of methane gas on three

shale samples from upper Paleozoic strata of the Ordos Basin. J. Pet. Sci. Eng. 110, 132–
138.

157
 BIBLIOGRAPHY

 Gutiérrez-Marco, J.C., Storch, P., 1998. Graptolite biostratigraphy of the lower Silurian
(Llandovery) shelf deposits of the Western Iberian Cordillera, Spain. Geol. Mag. 135,71-
92.

 Heller, R., Vermylen, J., Zoback, M., 2014. Experimental investigation of matrix
permeabilitiy of gas shales. AAPG Bull. 98, 975-995.

 Heller, R., Zoback, M., 2014. Adsorption of Methane and Carbon Dioxide on Gas Shale
and Pure Mineral Samples. J. Unconv. Oil Gas Resour. 8, 14–24.

 Hou, Y.G., He, S., Yi, J.Z., Zhang, B., Chen, X., Wang, Y., Zhang, J., Cheng, C.Y., 2014. Effect
of pore structure on methane sorption capacity of shales. Pet Explor Dev. 41, 272-281.

 Hu, H., 2014. Methane adsorption comparison of different thermal maturity kerogens
in shale gas system. Chin. J. Geochem. 33 (4), 425-430.

 Jarvie, D.M., Hill, R.J., Ruble, T. E, Pollastro, R.M., 2007. Unconventional shale-gas
systems: The Mississippian Barnett Shale of north-central Texas as one model for
thermogenic shale-gas assessment. AAPG Bull. 91, 475-499.

 Ji, W., Song, Y., Jiang, Z., Wang, X., Bai, Y., Xing, J., 2014. Geological controls and
estimation algorithms of lacustrine shale gas adsorption capacity: a case study of the
Triassic strata in the southeastern Ordos Basin, China. J. Int. J. Coal Geol. 134-135, 61-
73.

 Ji, L.M., Zhang, T., Milliken K.L., Qu, J., Zhang, X., 2012. Experimental investigation of
main controls to methane adsorption in clay-rich rocks. Appl Geochem. 27, 2533-2545.

 Kang, S.M., Fathi, .E, Ambrose, R.J., Akkutlu, I.Y., Sigal, R.F, 2011. Carbon dioxide storage
capacity of organic-rich shales. SPE Journal. 16, 842–55.

 Kang, S.M., Shinn, Y.J., Akkutlu, I.Y., 2014. Gas Storage Capacity of Iljik and Hasandong
Shales in Gyongsang Basin, South Korea. Paper presented at the Unconventional
Resources Technology Conference, held in Denver, Colorado, USA, 25-27 August.

 Kim, J., Kim D., Lee, W, Lee, Y., Kim, H., 2017. Impact of total organic carbon and specific
surface area on the adsorption capacity in Horn River shale. J. Pet. Sci. Eng. 149, 331-
339.

 Knowles, G.P., Delaney, J.V., Chaffe, A.L., 2006. Dyethlenetriamine [propyl(silyl)]-

Functionalized (DT) Mesoporous Silicas as CO 2 Adsorbents. Ind. Eng. Chem. Res. 45,
2626-2633.

 Kyoto Protocol to the United Nations Framework Convention on Climate Change.

United Nations, 1998.

 Li, P., Jiang, Z., Zheng, M., Bi, H., Chen, L., 2016. Estimation of shale gas adsorption
capacity of the Longmaxi Formation in the Upper Yangtze Platform, China. J. Nat. Gas
Sci. Eng. 34, 1034–1043.

158
 BIBLIOGRAPHY

 Liu. G.H., Huang, Z.L., Jiang, Z., Chen, J., Chen, F., Xing, J., 2016. Gas adsorption capacity
calculation limitation due to methane adsorption in low thermal maturity shale: A case
study from the Yanchang Formation, Ordos Basin. J. Nat. Gas Sci. Eng. 30, 106-118.

 Lu, X.C., Li, F.C., Watson, A.T., 1995. Adsorption measurements in Devonian shales.Fuel.
74, 599-603.

 McCarthy, K., Rojas, K., Niemann, M., Palmowski, D., Peters, K., Stankiewicz, A., 2011.
Basic Petroleum Geochemistry for Source Rock Evaluation. Oilfield Rev. 23, 32-43.

 Mendhe, V.A.; Mishra, S.; Varma, A.K.; Kamble, A.D.; Bannerjee, M.; Sutay, T., 2017. Gas
reservoir characteristics of the Lower Gondwana Shales in Raniganj Basin of Eastern
India. J. Pet. Sci. Eng. 149, 649-664.

 Metz, B., Davidson, O., de Coninck, H., Loos, M., Meyer, L., 2005. IPCC Special Report on
Carbon Dioxide Capture and Storage. Cambridge Univ. Press, Cambridge & New York.

 Millward, A.R., Yaghi, O.M., 2005. Metal-organic frameworks with exceptionally high
capacity for storage of carbon dioxide at room temperature. J Am Chem Soc. 127,
17998–17999.

 Montgomery, S.L., Jarvie, D.M., Bowker, K.A., Pollastro, R. M., 2005. Mississippian
Barnet Shale, Fort Worth basin, north-central Texas: Gas-shale play with multi-trillion
cubic foot potential. AAPG Bull. 89,155-175.

 Moore D.M., Reynolds, R.C. Jr., 1989. X-Ray Diffraction and the Identification and
Analysis of Clay Minerals. Oxford University Press. New York.

 Mouahid A., Bessieres. D., Plantier. F., Pijaudier-Cabot, G., 2012a. Supercritical
adsorption of nitrogenon EcoSorb-activated carbon at temperatures up to 383 K and
pressures up to 2 MPa. J. Therm. Anal. Calorim. 109, 473-479.

 Mouahid, A., Bessieres, D., Plantier, F., Pijaudier-Cabot, G., 2012b. A Thermostated
Coupled Apparatus for the Simultaneous Determination of Adsorption Isotherms and
Differential Enthalpies of Adsorption at High Pressure and High Temperature. J. Therm.
Anal. Calorim. 109, 1077–1087.

 Myers, A.L., Prausnitz, J.M. 1965. Thermodynamics of mixed adsorption. AIChE J. 11,
121-127.

 Nuttall, B.C., Eble, C.F., Drahovzal, J.A., Bustin, R.M., 2005. Analysis of Devonian black
shales in Kentucky for potential carbon dioxide sequestration and enhanced natural gas
production. Report Kentucky Geological Survey/University of Kentucky. (DE-FC26-
2NT41442).

 Othman, M.R., 2009. Permeability and separability of methane and carbon dioxide
across meso-porous Mg–Al hydrotalcite and activated carbon media. Chem Eng Sci.
64(5), 925–929.

159
 BIBLIOGRAPHY

 Pan, Z. and Connell, L.D., 2015. Reservoir Simulation of free and adsorbed gas
production from shale. J. Nat. Gas Sci Eng. 22, 359-370.

 Peters, K.E., 1986. Guidelines for evaluating petroleum source rock using programmed
pyrolysis. Am. Assoc. Pet. Geol. Bull. 70, 328-329.

 Pino, D. (2014). Étude Expérimentale de l’Adsorption de gaz sur des solides micro ou
meso-poreux à haute pression. Application a des problématiques du génie pétrolier
(PhD Dissertation). Université de Pau et des Pays de l’Adour, Pau, France.

 Pino, D., Plantier, F., Bessieres, D., 2014. Experimental Determination of the Adsorption
Isotherms in Gas Mixtures under Extended Pressure and Temperature Range:
Application to the CO2 –CH4 Binary Mixture. J. Therm. Anal. Calorim. 117, 1469–1477.

 Pozo, M., Pino, D., Bessieres, D., 2017. Effect of thermal events on maturation and
methane adsorption of ammonium illite-rich Silurian black shales (Checa, Spain). Appl.
Clay Sci.136, 208-218.

 Rexer, T.F.T., Benham, M.J., Aplin, A.C., Thomas, K.M., 2013. Methane adsorption on
shale under simulated geological temperature and pressure conditions. Energy Fuels.
27, 3099-3109.

 Rosenholm, J.M., Penninkangas, A., Linden, M., 2006. Amino-functionalization of large-

pore mesoscopically ordered silica by a one-step hyperbranching polymerization of a
surface-grown polyethyleneimine. Chem. Comm. 3909-3911.

 Ross, S., Olivier, J.P., (1964). On Physical Adsorption. Interscience (New York).

 Ross, D. and Bustin, R.M., 2007. Shale gas potential of the lower Jurassic Gordondale
member, northeastern British Columbia, Canada. Bull. Can. Pet. Geol. 55, 51-75.

 Ross, D., Bustin, R.M. 2008. Characterizing the shale gas resource potential of Devonian
Mississippian strata in the Western Canada sedimentary basin: application of an
integrated formation evaluation. AAPG Bull. 92, 87-125.

 Ross, D., Bustin, R.M., 2009. The Importance of Shale Composition and Pore Structure
upon Gas Storage Potential of Shale Gas Reservoirs. Mar. Pet. Geol. 26, 916–927.

 Rouquerol, F., Rouquerol, J., Sing, K.S.W., Llewellyn, P., Maurin, G., 2014. Adsorptionby
Powders and Porous Solids. Principles, Methodology and Appli cations. Second Edition.
Elsevier Science.

 Santori, G., Luberti, M., Ahn, H., 2014. Ideal adsorbed solution theory solved with direct
search minimisation. Computers and Chemical Engineering. 71, 235-240.

 Santos, J.M. and Akkutlu, I.Y., 2013. Laboratory Measurements of Sorption Isotherm
Under Confining Stress With Pore-Volume Effects. SPE Journal. 18(5), 924-931.

160
 BIBLIOGRAPHY

 Sanzs, R., Calleja, G., Arencibia, A., Sanz-Pérez, E.S., 2010. CO2 Adsorption on Branched
Polyethyleneimine-Impregnated Mesoporous Silica SBA-15. Appl. Surf. Sci. 256, 5323-
5328.

 Sanzs, R., Calleja, G., Arencibia, A., Sanz-Pérez, E.S., 2012. Amino functionalized
mesostructured SBA-15 silica for CO2 capture: Exploring the relation between the
adsorption capacity and the distribution of amino groups by TEM. Microporous
Mesoporous Mat. 158, 309-317.

 Shi, J.Q., Durucan, S., 2008. Modeling of mixed-gas adsorption and diffusion in coalbed
reservoirs. Soc. Pet. Eng. SPE 114197.

 Sigal, R.F., Akkutlu, I.Y., Kang, S.M., Diaz-Campos, M., Ambrosse, R.J., 2013.The
Laboratory Measurement of Gas Storage Capacity of Organic Shales. Petrophysics. 54,
224-235.

 Sircar, S. 1999. Gibbsin surface excess for gas adsorption8 Gibbsian surface excess for
gas adsorption. Ind. Eng. Chem. Res. 28, 3670-3682.

 Span, R., Wagner, W., 1996. A new equations of state for carbon dioxide covering the
fluid region from the triple point temperature to 1100 K at pressures up to 800 MPa. J.
Phys. Chem. Ref Data. 1996; 25(6), 1509-1596.

 Span, R., Wagner, W., 2003. Equations of State for Technical Applications. I.
Simultaneously optimized functional Forms for nonpolar and polar fluids. Int. J.
Thermophys. 24, 1–39.

 Srodon, J., 1980. Precise identification of illite/smectite interstratifications by X -ray

powder diffraction. Clays Clay Miner. 28, 401-411.

 Sui, H., Yao J., 2016. Effect of surface chemistry for CH 4/CO2 adsorption in kerogen: A
molecular simulation study. J. Nat. Gas Sci. Eng. 31, 738-746.

 Technically Recoverable Shale Oil and Shale Gas Resources: An Assessment of 137Shale
Formations in 41 Countries outside the United States. Energy Information
Administration (EIA), 2013.

 Tian, H., Li, T., Zhang, T., Xiao, X.M., 2016. Characterization of methane adsorption on
over mature Lower Silurian – Upper Ordovician shales in Sichuan Basin, southwest
China: Experimental results and geological implications. Int. J. Coal Geol. 156, 36–49.

 Tontiwachwuthikul, P., Meisen, A., Lim, C.J.J., 1991. Solubility of Carbon -Dioxide in 2-
amino-2.methyl-1-propanol solutions. J. Chem. Eng. Data. 36 (1) 130-133.

 Wagner, W. and Span, R., 1993. Special Equations of State for Methane, Argon and
Nitrogen for the temperature range from 270 to 350K at pressures up to 30 MPa. Int. J.
Thermophys. 14(4), 699–725.

161
 BIBLIOGRAPHY

 Wang, S., Song, Z., Cao, T., Song, X., 2013. The methane sorption capacity of Paleozoic
shales from the Sichuan Basin, China. Mar. Pet. Geol. 44, 112-119.

 Wang, Y, Zhu, Y., Liu, S., Zhang, R., 2016. Pore characterization and its impact on
methane adsorption capacity for organic-rich marine shales. Fuel.181, 227-237.

 Weniger, P., Kalkreuth, W., Bush, A., Krooss, B.M., 2010. High-pressure methane and
carbon dioxide sorption on coal and shale samples from the Parana Basin, Brazil. J. Col.
Geol. 84, 190-205.

 Wu, Y., Fan, T., Jiang, S., Yang, X., Ding, H., Meng, M., Wei, D., 2015. Methane adsorption
capacities of the Lower Paleozoic Marine Shales in the Yangtze Platform, South China.
Energy Fuels. 29(7), 4160–4167.

 Xia, J., Song, Z., Wang, S., Zeng, W., 2017. Preliminary study of pore structure and
methane sorption capacity of the Lower Cambrian shales from the north Gui -Zhou
Province. J. Nat. Gas Sci. Eng. 38, 81-93.

 Xiong, F.Y., Jiang, Z.X., Li, P., Wang, X.Z., Bi, H., Li, Y.R. et al., 2017. Pore structural of
transitional shales in the Ordos Basin, NW China: effects of composition on gas storage
capacity. Fuel. 206, 504-515.

 Xu, J., Bechtel, A., Sachsenhofer, R.F., Liu, Z., Gratzer, R., Meng, Q., Song, Y., 2015. High
resolution geochemical analysis of organic matter accumulation in the Qingshankou
Formation, Upper Cretaceous, Songliao Basin (NE China). Int. J. Coal Geol. 141-142, 23-
32.

 Xu, X.C., Song, C.S., Andresen, J.M., Miller, B.G., Scaroni, A.W.T., 2002. Novel
polyethylenimine – modified mesoporous molecular sieve of MCM-41 type as high –
capacity adsorbent for CO 2 capture. Energy Fuels. 16, 1463-1469.

 Yang, F.,Ning, Z.F., Liu, H.Q., 2014. Fractal characteristics of shales from a shale gas
reservoir in the Sichuan Basin, China. Fuel. 115, 378-384.

 Yang, F., Ning, Z.F., Zhang, R., Zhao, H., Krooss B.M., 2015. Investigations on the methane
sorption capacity of marine shales from Sichuan Basin, China. Int. J. Coal Geol. 146, 104–
117.

 Yu, S., Chen, H., Yang, S., Guo, X., Zhou, Ch., Fang, B., Zhou, F., Yang, J.K., 2014. A New
Mathematical Model Considering Adsorption and Desorption Process for Productivity
Prediction of Volume Fractured Horizontal Wells in Shale Gas Reservoirs. J. Nat. GasSci.
Eng. 19, 228–36.

 Yuan, W., Pan, Z., Li, X., Yang, Y., Zhao, C., Connel, L.D., Li, S., He, J., 2015. Experimental
study and modelling of methane adsorption and diffusion in shale. Fuel. 117, 509-519.

 Yue, M.B., Chun, Y., Cao, Y., Dong, X., Zhu, J.H., 2006. CO 2 capture by as-prepared SBA-
15 with an occluded organic template. Adv. Funct Mater. 16, 1717-1722.

162
 BIBLIOGRAPHY

 Yue, M.B., Sun, L.B., Cao, Y., Wang, Y., Wang, Z.J., Zhu, J.H., 2008. Efficient CO 2 capturer
derived from as syntehesized MCM41 modified with amine.Chem. Eur. J. 14, 3442-3451.

 Zhang, T., Ellis, G.S., Rupple, S.C.; Milliken, K., Yang, R., 2012. Effect of organic matter
type and thermal maturity on methane adsorption in shale -gas systems. Org. Geochem.
47, 120-131.

 Zhang, Y., Shao, D., Yan, J., Jia, X., Li, Y., Yu, P., Zhang, T., 2016. The pore size distribution
and its relationship with shale gas capacity in organic-rich mudstone of Wufeng-
Longmaxi Formations, Sichuan Basin, China. J. Nat Gas Geosci. 1, 213-220.

163
164
165
166
ANEXES

167
 Article

Cite This: Energy Fuels 2017, 31, 11698-11709 pubs.acs.org/EF

Effect of Organic Matter and Thermal Maturity on Methane

Adsorption Capacity on Shales from the Middle Magdalena
Valley Basin in Colombia
Olga Patricia Ortiz Cancino,*,† Deneb Peredo Mancilla,‡ Manuel Pozo,§ Edgar Pérez,∥
and David Bessieres‡
†
Universidad Industrial de Santander, Bucaramanga, Santander 680002, Colombia
‡
Université de Pau et des Pays de l’Adour, CNRS, UMR 5150, Laboratoire des Fluides Complexes et leurs Réservoirs, TOTAL, 64013
Pau, France
§
Department of Geology and Geochemistry, Universidad Autónoma de Madrid (UAM), Cantoblanco, 28049 Madrid, Spain
∥
Instituto Colombiano del Petróleo (ICP), Piedecuesta, Santander 681018, Colombia

ABSTRACT: High-pressure methane adsorption isotherms were measured on five shale core samples obtained during
exploratory drilling from three boreholes located in the Colombian Middle Magdalena Valley Basin. The experiments were
carried out at 50 and 75 °C and for pressures ranging up to 3.5 MPa under dry conditions through the use of a homemade
manometric setup. The effect of the total organic carbon (TOC) content, thermal maturity, clay content, and specific surface area
(SSA) on methane adsorption capacity has been discussed. The excess adsorption data were fitted to a three-parameter (nL, pL,
and ρads) Langmuir model with the value of the adsorbed phase density, ρads, maintained fixed at 421 kg/m3, which corresponds
to liquid-phase density of methane at a normal boiling point. An excellent fit to the experimental data was achieved. The results
show that the temperature has a negative effect on the adsorption capacity, while TOC has a positive effect, even if no linear
regression was found between TOC and methane adsorption capacity. No correlation was observed between the clay content
and the TOC-normalized adsorption capacity to methane, which indicates that clay minerals do not significantly contribute to
methane adsorption in the case of our samples. In addition, there is not a general trend between TOC normalized and thermal
maturity. Among the factors investigated in the present study, TOC has the major contribution to the adsorption uptake.
A similar contribution is found for the SSA, which is consistent, considering the positive correlation between TOC and SSA. This
set of data represents meaningful information for indirect estimations of the gas in place during the future recovery strategies.
This study furthers the ongoing projects on the understanding of the adsorption effect on shale gas production and assessment.

1. INTRODUCTION Adsorption is a complex process, which depends upon the

In 2016, according to the BP Statistical Review of World rock matrix and fluid properties as well as reservoir conditions
Energy,1 the natural gas production of Colombia was almost (e.g., temperature and pressure). The main parameters affecting
equal to its consumption and the reserves/production (R/P) adsorption capacity are the total organic carbon (TOC) con-
ratio was close to 12. At the present time, almost all of the tent, mineralogy, water content, temperature, and pressure.
natural gas production of country comes from conventional Most of the studies affirm that organic matter, TOC, is the
reserves, which could be empty by the year 2028; meanwhile, main factor that controls adsorption uptake in shales.8,18−23 It
unconventional technically recoverable gas reserves are esti- has also been reported that the type of kerogen as well as its
mated to be 12 times greater than conventional reserves.2 Shale maturity can influence the adsorption ability in such a way that
gas has become an increasingly important source of natural gas the methane adsorption capacity of kerogen decreases in the
supply. sense of type III > type II > type I,7,18 while a higher maturity
It is of general acceptance that natural gas can be stored in means a higher adsorption capacity.8,20,27 Others studies reported
shales in three different ways: as free gas, adsorbed gas, and that sorption capacity will first increase and then decrease with
dissolved gas,3−8 with adsorbed gas being the main contribu- maturity.28,29
tion (up to 85%).3,9−12 Therefore, the quantity of adsorbed The roles of shale mineral composition and pore structure have
gas represents one of the most important parameters in gas been largely studied.8,28 In addition to organic matter, clay mineral
shale reserves and production estimations.13 In this sense, may provide a contribution upon adsorption capacity.4,7,8,23,27,30−32
many laboratory experiments have been carried out on methane Montmorillonite and illite/smectite present a higher adsorption
adsorption in shale gas from different worldwide basins, with capacity than kaolinite, chlorite, and illite.30 At a nanometric
the objective to provide a better understanding of this phemo- scale, it is difficult to correlate the adsorption gas capacity
menon.7,14−23 In addition to adsorption measurements, some
works had obtained at the same time the stored free-gas Received: June 30, 2017
amount, with the confining-stress effect having been taken into Revised: September 27, 2017
account.24−26 Published: September 28, 2017

© 2017 American Chemical Society 11698 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Figure 1. Location of the Middle Magdalena Valley (in yellow) and sample site (black circle) [modified with permission from ref 40. Copyright 2007
Agencia Nacional de Hidrocarburos (ANH)].

Figure 2. Schematic diagram of the HP/high-temperature (HT) manometric setup.

directly with the total organic content as a result of the pore A reduction in gas adsorption up to 40% has been found when
size distribution and heterogeneity of the surface;8 therefore, comparing moisture samples to dry samples.7,8 The tempera-
some authors suggest that the adsorbed gas volume evaluation ture is also one of the factors influencing the state of shale gas.
should also be related to the surface area.12,33−35 The adsorp- With gas adsorption being an exothermic process, the adsorp-
tion in clay-rich shales is due to their high internal area. There- tion capacity of shale decreases with an increasing tempera-
fore, the specific surface area (SSA) plays a significant role in ture.6,23 The combined effect of the pressure and temperature
gas adsorption36 as a result of the microporosity associated with can be used during the production stage because it represents
organic matter. Zhang et al.37 report that shales with a higher gas desorption behavior.11 Although the above-mentioned param-
content of clay minerals and a similar TOC content have a larger eters are the most studied, some works have been performed
SSA. This is due to the porosity hosted in the clay minerals. focused to dynamically changing pore volume adjustments as a
Pressure increases the adsorption capacity to some extent result of the adsorption layer taking up space and overburden
when it rises isothermally.38 In contrast, the water content and effects on core shale samples.24−26
temperature have a negative influence. Water may occupy the This review highlights that gas storage in shale is a complex
adsorption sites, hence reducing the amount of adsorbed gas.20,39 multi-parameter process. An understanding and quantification
11699 DOI: 10.1021/acs.energyfuels.7b01849
Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Table 1. Depth and Mineralogical Analysis of the Samples

sample depth (m) illite (%, w/w) kaolinite (%, w/w) quartz (%, w/w) calcite (%, w/w) pyrite (%, w/w) gypsum (%, w/w) apatite (%, w/w)
S1A 2835 15 45 27 3 10 id 0
S1B 2850 11 19 54 10 6 id 0
S2A 2934 9 21 31 33 4 id <2
S2B 3004 13 15 11 50 9 id <2
S3 4430 7 22 32 28 8 id <2

Figure 3. XRD diffractograms of bulk samples. Mineralogical assemblages include phyllosilicates (kaolinite and illite, in illite/smectite mixed layers),
calcite, quartz, apatite, and pyrite. S1A shows the highest percentage of phyllosilicates (60%, mostly kaolinite), whereas the quartz content can reach
54% in S1B and the calcite content can reach 50% in S2B (see Table 1).

of each parameter require a very huge set of well-defined Girardot fold belt (GFB), respectively. To the northeast, the basin is
experimental data. Despite the growing interest, research pub- limited by the Bucaramanga−Santa Marta fault system (BSMF), and
lished on shale is mostly limited to U.S. and Canadian shales, to the southeast, the basin is limited by the Bituima and La Salina fault
China shales, and more recently European black shales. Less system (BSFS). The western limit is marked by the westernmost onlap
studies have been reported for South American shales. of the Neogene basin fill into the Serraniá de San Lucas (SL) and the
Central Cordillera (CC) basement,40 as shown in Figure 1. The black
The objective of the present study is to further the set of
circle shows the location of the wells in the proximity of Barrancabermeja.
available data. Here, we report methane adsorption data for 2.2. Sample Characterization. 2.2.1. X-ray Diffraction (XRD)
selected Colombian shales from the Middle Magdalena Valley. Analysis. Mineralogical analysis of samples was carried out by means
Using a homemade manometric setup, methane adsorption of XRD using Siemens D-5000 equipment with a scanning speed of 1°
isotherms were reported at 50 and 75 °C for pressures up to 3.5 (2θ)/min and Cu Kα radiation (40 kV and 20 mA). XRD is the most
MPa. These measurements were carried out on dried samples widely used technique for identification of minerals. When an incident
from cores belonging to three different wells. beam of X-ray interacts with crystalline matter (regular structure), the
Following recommendations of previous works devoted to diffraction (constructive interference) can occur for certain directions,
shales, the methodology applied in this work is as follows: giving a set of reflections characteristic of the analyzed substance
(1) geochemical and textural characterization of the samples, (fingerprint). The diffraction reflections are related to spacing of atomic
(2) CH4 adsorption capacity over ranges of pressure and tem- planes in a sample (i.e., d spacing) and wavelength of X-rays (λ).
XRD studies were achieved on both randomly oriented samples (bulk
perature, with representation of the adsorption data by a mod- sample) and clay fraction samples (<2 μm). Powdered whole-rock
ified Langmuir approach, and (3) variation of the CH4 uptake
as a function of the Brunauer−Emmett−Teller (BET) surface
Table 2. Rock-Eval Analysis, HI, OI, and SSA (BET
area, organic matter richness, clay content, and thermal maturity.
Method)a
This comprehensive work furthers the still limited reliable
database of adsorption data on shales. To the best of our knowl- Tmax TOC
edge, this is the first study devoted to Colombian gas shales. sample (°C) (%) S1 S2 S3 HI OI SSA
S1A 459 3.8 1.93 3.75 0.14 99 3.69 6.9
2. MATERIALS AND EXPERIMENTAL METHODOLOGY S1B 463 4.7 2.00 3.61 0.23 78 4.94 10.28
S2A 478 3.1 0.35 0.70 0.18 22 5.76 13.05
2.1. Materials. Five core samples were obtained during exploratory S2B 487 8.8 0.39 1.90 0.42 22 4.79 26.29
drilling (stratigraphic wells) of three boreholes located in the Middle
S3 471 5.7 2.47 4.96 0.25 87 4.37 10.59
Magdalena Valley Basin in Colombia. As a result of confidentiality
a
reasons, detailed locations are not disclosed, and the samples are Tmax, thermal maturation parameter; TOC, total organic carbon
named S1A, S1B, S2A, S2B, and S3. (wt %); S1, free HC (mg of HC/g of rock); S2, oil potential (mg of
The Middle Magdalena Valley Basin is 34 000 km2. It is stretched HC/g of rock); S3, CO2 organic source (mg of CO2/g of rock); HI,
along the middle reaches of the Magdalena river and is bound to the hydrogen index; OI, oxygen index; and SSA, specific surface area
north and south by the Espiritu ́ Santo fault system (ESFS) and the (m2/g of rock).

11700 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

nitrogen gas. Adsorption isotherms, describing the amount of gas

adsorbed as a function of the relative pressure (p/p0), can exhibit
different features depending upon the size of particles, the presence of
organized pores, and the energetic properties of the mineral surface.
Different methods of data analysis are used to derive quantitative infor-
mation from experimental adsorption curves, of which BET42 analysis
is the most common. The BET SSA of powdered samples was analyzed
by Micromeritics ASAP 2010 equipment and determined from a low-
pressure nitrogen adsorption isotherm at 77 K (−196 °C).
2.2.3. Rock-Eval Analysis. The Rock-Eval analysis of the selected
whole rock samples was performed by a Rock-Eval VI pyrolyser.
The interesting parameters measured include TOC, Tmax, S1 [free hydro-
carbons (HC)], S2 (oil potential), S3 (CO2 organic source and carbonate),
and related indices: oxygen index (OI) and hydrogen index (HI).
2.3. Methane Adsorption Measurements. 2.3.1. Sample
Preparation. Considering that the time required to reach the equilib-
rium on intact core samples could be extremely long as a result of the
low permeability of shales,43 the samples were split. However, depend-
ent upon the organic matter distribution, the milling process may
induce variations on the available surface. Following the recom-
mendations of Gasparik et al.,20 cuttings of different particles sizes
were tested.5 For each sample, two particle size fractions ranging from
1 to 1.5 mm and from 1.5 to 5 mm were investigated. A known mass
of each sample was introduced into the adsorption cell (Figure 2) and
dried at 105 °C for 24 h to remove the moisture. The mass was care-
fully measured and chosen to obtain a large enough available adsorp-
tion area; about 30 m2 is the minimum area required.
2.3.2. Manometric Adsorption Setup. The instrument used in the
present study is a high-pressure (HP) manometric device. A schematic
view of this “homemade” apparatus is provided in Figure 2. This exper-
imental setup and the measurement principle have been previously
described.44 The main elements are the reference or dosing cell
Figure 4. Tmax−HI plot. (33.57 cm3), the adsorption cell (16.56 cm3), in which the adsorbent
is placed, and the pressure transducer (MKS Baratron type 121 A,
samples (milled and dry sieved at <63 μm for homogenization) were with an uncertainty of 0.01% in the full scale ranging from the vacuum
scanned from 2° to 65° (2θ). The method of the mineral intensity to 3.5 MPa). The various parts are isolated with spherical valves, thus
factors (MIFs) was applied to XRD peak intensity ratios normalized to limiting the “dead space” volume. The whole apparatus is regulated
100% with calibration constants for the quantitative estimation of the under isothermal conditions through the use of a heater wire con-
mineral contents.41 The clay fraction (<2 μm) was separated by cen- trolled by a proportional−integral−derivative (PID) regulator (Euro-
trifugation, and samples were prepared from suspensions oriented on therm 3208). Five thermocouples (type K, accuracy of ±0.1 K) were
glass slides. Identification of the clay fraction minerals was performed placed in different parts of the circuit to check that the isothermal
on oriented air-dried samples and solvated with ethylene glycol after conditions are applied along the circuit during the measurement.
heating at 550 °C. The overall uncertainty of the amount adsorbed (as a result of the
2.2.2. SSA. The SSA of minerals is one of their most important helium calibration procedure and pressure accuracy) is determined to
properties controlling surface phenomena. The most widely used be lower than 1% over the entire (p and T) range investigated in this
technique for determining SSA is based on gas adsorption, notably of study.

Figure 5. Correlation between BET and TOC. Note a moderate linear correlation.

11701 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Figure 6. Methane adsorption capacity at 50 °C.

Figure 7. Methane adsorption capacity at 75 °C.

Figure 8. CH4 adsorption capacity for sample S2B at 50 and 75 °C.

2.3.3. Calculation of Excess Adsorption. The adsorption iso- increased pressure about 2−3 bar between successive gas doses.
therms were determined using an accumulative process, with a value of The procedure consists of expanding a gas from the dosing volume VD

11702 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Table 3. Langmuir Model Fitting Parameters at 50 °C in which nexcess

ads is the adsorbed amount of gas (mol/kg) at p (MPa),
pL is the Langmuir pressure, corresponding to the pressure at which
sample nL (mol/kg) pL (MPa) Δn half of the adsorption sites are occupied (monolayer), nL is the amount
S1A 0.0848 0.711 0.00098 adsorbed (mol/kg) when all of the monolayer is filled (maximum
S2A 0.1054 1.661 0.00017
Langmuir capacity), ρg is the gas density (kg/m3) at p and T, and
ρads (kg/m3) is the adsorbed phase density, which was assumed as a
S2B 0.1926 0.680 0.00040 fixed value of 421 kg/m3 for the adsorbed phase density.12,19,52
S3 0.0783 0.315 0.00001 Standard deviation was calculated according to53
N
Table 4. Langmuir Model Fitting Parameters at 75 °C Δn =
1
∑ (nexp − nfit)2
N
sample nL (mol/kg) pL (MPa) Δn 1

S1A 0.0169 0.071 0.00002 where N is the number of experimental data points and nexp and nfit are
S2B 0.0562 0.118 0.00010 experimental measured data and fitted data, respectively. The results
S3 0.0537 0.229 0.00021 are shown in the next section.

3. RESULTS
(dosing cell) into the adsorption cell (VM), which contains the sample
under isothermal conditions. Application of this experimental meth- 3.1. Mineralogy. The XRD mineralogical analysis revealed
odology requires the previous determination of the two volumes VD that the five bulk-rock samples are mostly made up of phyllo-
and VM. The uncertainty of these measures is inferior to 0.5% in both silicates (28−60%), quartz (11−54%), and calcite (3−50%).
cases. The mass balance involves the void volume or volume accessible Other minerals include pyrite (4−10%), apatite (<2%), and
in the presence of the adsorbent, which is a key parameter for the traces of gypsum (see Table 1). The sample S1A shows the
adsorption capacity. The void volume was determined through helium highest content in phyllosilicates (>50%) and pyrite. The
expansions at each temperature and for different pressures. The choice remaining samples have a phyllosilicate content up to 30%,
of helium was dictated considering it is an inert, non-sorbing gas.18,23
An additional drying is performed after that for 8−10 h. The methane
whereas quartz can reach 54% in S1B and calcite can reach 50%
molar volume considered at the experimental conditions (p and T) is in S2B. The value of d(060) reflection is in all cases 1.49−1.50 Å,
determined with the Span and Wagner equation of state (EOS).45 The indicating dioctahedral phyllosilicates. The oriented aggregates
return to the thermodynamic equilibrium was controlled by the pres- of the clay fraction (<2 μm) show that samples are mostly com-
sure value. It should be observed that it was reached in a range from posed of two clay minerals: kaolinite and illite. In the bulk
45 to 60 min. sample, the kaolinite content is ranging between 15% (S2B)
2.3.4. Parameterization of Excess Adsorption Isotherms. Because and 45% (S1A). Illite is subordinated (7−15% in the bulk
reservoir pressures are higher than the experimental pressures, it is sample) and always include traces of illite/smectite mixed
necessary to extrapolate data to well pressure conditions. Therefore, layers. Two representative XRD patterns of bulk samples are
the experimental data were parametrized using a fitting procedure.
Several approaches have been developed4,46−48 to this purpose.
shown in Figure 3.
The modified Langmuir model is used as a standard model to describe 3.2. Organic Matter Richness and Thermal Maturity.
vapor isotherms on shales,48 and it is widely accepted in the petroleum Rock-Eval analysis can help to know the oil−gas potential of a
industry.33 This model includes the adsorbed gas density as a fitting rock but also the type of organic matter and degree of matura-
parameter. Because this value is difficult to assess, most of the studies tion. The most interesting parameters measured are shown in
have shown a good match when assuming it as the methane liquid Table 2.
density at a normal boiling point.4,12,39,49,50 TOC is ranging between 3.12% (S2A) and 8.77% (S2B).
The experimental data were correlated using a three-parameter TOC is the amount of organic carbon present in the sample.
Langmuir model described by Gensterblum et al.51 and applied by In shales, TOC of 2−5% is considered good and higher than
Gasparik et al.4
5% very good.53 All analyzed samples have TOC higher than
p ⎛ ρ (p , T ) ⎞ ⎛ ρg (p , T ) ⎞ 3% and, in the case of S3 and S2B, higher than 5%.
excess
nads = nL ⎜1 − g ⎟ = nads
absolute⎜
1− ⎟
⎜
p + pL ⎝ ρads ⎠ ⎟ ⎜ ρads ⎟⎠
Besides, TOC is important to consider the level of thermal
⎝ maturation, which can be given by the Tmax value. This parameter

Figure 9. CH4 adsorption capacity nexcess (at 3 MPa and 50 °C) as a function of TOC.

11703 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Figure 10. CH4 adsorption capacity at 50 °C as a function of TOC at (a) 0.3 MPa and (b) 0.5 MPa.

is the temperature at which the maximum amount of HC deg- rock (sample S2A) and 2.47 mg of HC/g of rock (sample S3)
raded from kerogen was generated. The Tmax values range for parameter S1 and between 0.7 mg of HC/g of rock (sample
between 459 and 487 °C (Table 2). In general, according to S2A) and 4.96 mg of HC/g of rock (sample S3) for parameter
Peters,55 Tmax values lower than 435 °C are considered immature S2. The relative amounts of parameters S1 and S2 depend upon
organic matter but Tmax values between 435 and 455 °C indi- the type of organic matter but also the duration and tempera-
cate “oil window” conditions (mature organic matter). Higher ture suffered by the rock. Parameter S3 indicates CO2 evolved
values of Tmax between 455 and 470 °C are considered from thermal cracking during pyrolysis, reaching the highest
transitional, and higher values than 470 °C represent the wet value in sample S2B (0.42 mg of CO2/g) and the lowest value
gas zone (overmature organic matter). Indeed, when more in sample S1A (0.14 mg of CO2/g).
mature is the rock, the higher is the temperature (Tmax) required The OI is derived from the ratio (S3/TOC) × 100, ranging
to release HC from kerogen. Sample S1A shows a lower Tmax from 3.69 (sample S1A) to 5.76 (sample S2A). The HI is
value (459 °C); samples S1B, S2A, and S3 show intermediate derived from the ratio (S2/TOC) × 100, reaching 22 in sam-
values (463−478 °C); and sample S2B shows the highest value ples S2A and S2B, between 78 and 87 in samples S1B and S3,
(487 °C). Therefore, the maturation order is S2B > S2A > S3 > and the highest value in sample S1A (99). The type of kerogen
S1B > S1A. According to the Tmax−HI plot, all of the samples present in a rock determines its quality. Type I kerogen is the
are within the post-mature stage but samples S1A and S1B are highest quality, and type III is the lowest.54 The values of Tmax,
within the condensate−wet gas zone, whereas samples S2A and HI, and OI in the studied samples let us to include them as
S2B are within the dry gas window conditions and sample S3 is kerogen of types II−III (samples S1A, S1B, and S3) to type III
between them (Figure 4). The samples S3, S2A, and especially (samples S2A and S2B), according to classifications by Peters,55
S2B (Tmax > 470 °C) are indicative of overmature organic Gorin and Feist-Burkhardt,56 and Xu and co-workers57 (Figure 4).
matter. The maturation degree can affect the determination of the
With regard to S1 (free HC) and S2 (oil potential), the con- kerogene type. Indeed, the Tmax−HI and HI−OI plots are
centrations are low and range between 0.35 mg of HC/g of especially useful to determine kerogen type of immature rocks.
11704 DOI: 10.1021/acs.energyfuels.7b01849
Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Figure 11. (a) TOC-normalized Langmuir adsorption capacity (nL) content and (b) TOC-normalized adsorption capacity at 3 MPa as a function of
the total clay content, in our work and literature data.4

However, when a source rock is under maturation, the amount samples up to 3.5 MPa and for some of them (samples S1A, S2B,
of hydrogen and oxygen relative to carbon decreases and then and S3) at 75 °C (up to pressures of 2 MPa). To obtain reliable
the ratios tend to converge toward the origin of the plot. There- adsorption data at a high pressure is quite complicated.59 As a
fore, in post-mature rocks, HI and OI are not actually indicative result of the limitations of our own techniques, very reliable
of the original kerogen quality. data are accessible up to moderate pressures. In the present
3.3. SSA. The BET SSA is ranging between 6.90 and work, measurements were performed with high accuracy up to
26.29 m2/g, with the highest value in sample S2B (Table 2). a quite restricted pressure range (up to 3.5 MPa). Then, a
Several authors reported a relationship between kerogen char- phenomenological model applied to these data allows us to
acteristics (thermal maturity, composition, and type) and develop- extend the pressure range and to assess the CH4 uptake. Con-
ment of nanopores enhancing the gas adsorption capacity of sidering the difficulty associated with the very low adsorption in
shales.18,33,58−60 This fact explains the higher BET values obtained shales, a set of three measurements was performed for each
in more mature samples S2A and S2B (13.05−26.29 m2/g) isotherm. The reproducibility was always superior to 99% [aver-
when compared to the other samples (6.9−10.59 m2/g). The age absolute deviation (AAD) inferior to 1%]. The experimental
development of nanopores in kerogen can lead to an increase in data are displayed in Figure 6 (50 °C) and Figure 7 (75 °C).
adsorption sites with increasing TOC for the mature to post- Note that the sample S1B has very low adsorption capacity in
mature shales. Moreover, the analyzed samples show moderate comparison to the other samples or literature data.4−6 Alteration
positive correlation between TOC and BET SSA values (R2 = of this sample by oxidation may be the cause of this degradation.
0.71) because of increasing adsorption sites with maturation In this context, this sample will not be considered in the study.
and TOC content (Figure 5). Figure 8 reports the effect of the temperature for sample S2B.
3.4. Methane Adsorption Isotherms (Dry Samples). Because the saturation of the sample was observed at 75 °C, the
CH4 adsorption isotherms were measured at 50 °C for all of the pressure range was limited to pressures around 2 MPa.
11705 DOI: 10.1021/acs.energyfuels.7b01849
Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Figure 12. TOC-normalized adsorption capacity at 3 MPa as a function of the (a) kaolinite content (%) and (b) illite content (%).

Fitting parameters are shown in Tables 3 and 4, with the a linear law is observed between TOC and adsorption. This is
values of Δn in all of the samples indicating that the fitting due to the filling of micropores of the organic matter that
procedure was successful and that the Langmuir model repre- occurs at a first stage during the adsorption process.
sents the adsorption behavior in a good way without restric- Once that the influence of TOC over the adsorption capacity
tions. The values obtained for nL are similar to those already was determined, the effect of the clay content was studied. One
reported in the literature for shales or black shales.4,59 In addi- way to do this is by comparing the results with shale samples
tion, such a parameter should be regarded as useful information for to isolated kerogen, not accessible in our case. Therefore, we
future assessment and exploitation of the shale wells.61 Additionally, follow the methodology proposed by Gasparik et al.4 TOC-
their knowledge represents meaningful information to study the normalized adsorption capacities were plotted versus the total
effect of individual contributions to the methane adsorption. clay content for all of the samples (panels a and b of Figure 11).
Form the obtained figure, discrepancy of the adsorption capacities
4. DISCUSSION of the three samples with analogous clay contents are depicted.
4.1. Effect of Organic Matter and Clay Contents on Meanwhile, sample S1A with a higher clay content shows adsorp-
CH4 Uptake. CH4 adsorption capacity (nexcess) shows a mod- tion capacity similar to the other samples. This lack of correlation
erate positive correlation with TOC, taking into account that between the clay content and the adsorption capacities is also
TOC contents of samples S1A and S2A are nearly similar (3.8 observed with the literature data4 (panels a and b of Figure 11).
and 3.1). Even if no linear relationship can be fitted to the Similar results were observed for the detailed analysis of the
adsorption data, TOC remains as controlling factor of the individual clays plotted in panels a and b of Figure 12, the effect
adsorption uptake. A small discrepancy is observed for sample of illite and kaolinite seems to be insignificant in agreement
S3. This is shown in Figure 9, where we plot the excess adsorp- with previous studies.18,28 Furthermore, while the same studies
tion capacity (nexcess) at 3 MPa (and 50 °C) versus TOC. report that an important content of smectite can affect the
When the adsorption uptake is plotted at a lower pressure (see adsorption capacity, the studied samples on the present work
panels a and b of Figure 10 at 0.3 and 0.5 MPa, respectively), only presented trace amounts of illite/smectite.
11706 DOI: 10.1021/acs.energyfuels.7b01849
Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

Figure 13. (a) TOC-normalized sorption capacity at 3 MPa in our work and (b) TOC-normalized adsorption capacity at 3 MPa in our work and
literature data4 as a function of Tmax (maturity).

4.2. Effect of Thermal Maturity. The effect of thermal 5. CONCLUSION

maturity on the TOC-normalized excess adsorption capacity at New and reliable experimental data for methane adsorption
3 MPa (n3excess
MPa
) is shown in Figure 13a. With limitation to the capacities were obtained for fully characterized Colombian
samples of the present study, maturity showed no significant shales. To the best of our knowledge, this is the first study
effect on the total amount of methane adsorbed. devoted to Middle Magdalena Valley shale gas. For this work, a
Taking into account the still relatively small database, a Langmuir three-parameter model proved adequate to represent
general trend that correlates thermal maturity and adsorption the experimental excess adsorption, allowing for the gas-in-
capacity is not observed. A wider range of sample maturities is place estimations to be obtained. The study samples showed a
needed to conclude its effect over the adsorption capacity. moderate positive relationship between TOC and SSA.
Using data from the literature,4 the plot between maturity and The results confirmed that, even if no linear correlation was
adsorption capacity was obtained (see Figure 13b). A consis- found, TOC remains a key factor defining the adsorption capacity
tency is observed between the two sets of data. The TOC- of shales. Further work is needed to determine the importance
and effect of other parameters, such as the maturity of the samples.

■
normalized adsorption capacity linearly increases with maturity
within the investigated range.
In the literature, a variety of behaviors is reported between the AUTHOR INFORMATION
maturity and adsorption capacity. When the TOC-normalized Corresponding Author
adsorption capacities correlated positively with maturity in terms *E-mail: oportizc@uis.edu.co.
of vitrinite reflectance (VRr), the maximal value of VRr was ORCID
∼2.5%,4,5 which is in agreement with our observation. Olga Patricia Ortiz Cancino: 0000-0002-7949-348X
11707 DOI: 10.1021/acs.energyfuels.7b01849
Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

David Bessieres: 0000-0002-6262-4893 Ordovician shales in Sichuan Basin, southwest China: Experimental
results and geological implications. Int. J. Coal Geol. 2016, 156, 36−49.
Notes
(17) Wang, S.; Song, Z.; Cao, T.; Song, X. The methane sorption
The authors declare no competing financial interest.

■
capacity of Paleozoic shales from the Sichuan Basin, China. Mar. Pet.
Geol. 2013, 44, 112−119.
ACKNOWLEDGMENTS (18) Zhang, T.; Ellis, G. S.; Ruppel, S. C.; Milliken, K.; Yang, R.
The scientific work is included within the research activities Effect of organic matter type and thermal maturity on methane
of the Research Group C-144 (Geomaterials and Geological adsorption in shale-gas systems. Org. Geochem. 2012, 47, 120−131.
Processes, UAM). The authors are so grateful to the Colombian (19) Weniger, P.; Kalkreuth, W.; Busch, A.; Krooss, B. M. High-
Petroleum Institute (ICP) for supplying the samples and letting pressure methane and carbon dioxide sorption on coal and shale
us publishing these results. Olga Patricia Ortiz Cancino is grateful samples from the Parana Basin, Brazil. Int. J. Coal Geol. 2010, 84, 190−
205.
to the Universidad Industrial de Santander for the opportunity to
(20) Gasparik, M.; Gensterblum, Y.; Ghanizadeh, A.; Weniger, P.;
do her Ph.D. studies at UPPA and to Colfuturo, the Embassy of Krooss, B. M. High pressure/High temperature Methane-Sorption
France, the Colombian Ministry of National Education, and the Measurements on Carbonaceus Shales by the Manometric Method:
Association of Colombian Universities (ASCUN) for financial Experimental and Data-Evaluation Considerations for Improved
support.

■
Accuracy. SPE J. 2015, 20, 790−809.
(21) Ji, W.; Song, Y.; Jiang, Z.; Wang, X.; Bai, Y.; Xing, J. Geological
REFERENCES controls and estimation algorithms of lacustrine shale gas adsorption
(1) BP. BP Statistical Review of World Energy; BP: London, U.K., June capacity: A case study of the Triassic strata in the southeastern Ordos
2016. Basin, China. Int. J. Coal Geol. 2014, 134−135, 61−73.
(2) U.S. Energy Information Administration (EIA). Technically (22) Cui, X.; Bustin, A. M. M.; Bustin, R. Measurements of gas
Recoverable Shale Oil and Shale Gas Resources: An Assessment of 137 permeability and diffusivity of tight reservoir rocks: Different
Shale Formations in 41 Countries Outside the United States; EIA: approaches and their applications. Geofluids 2009, 9, 208−223.
Washington, D.C., 2013. (23) Lu, X. C.; Li, F. C.; Watson, A. T. Adsorption measurements in
(3) Curtis, J. B. Fractured shale-gas systems. AAPG Bull. 2002, 86, Devonian shales. Fuel 1995, 74, 599−603.
1921−1938. (24) Santos, J. M.; Akkutlu, I. Y. Laboratory Measurements of
(4) Gasparik, M.; Ghanizadeh, A.; Bertier, P.; Gensterblum, Y.; Sorption Isotherm Under Confining Stress With Pore-Volume Effects.
Bouw, S.; Krooss, B. M. High-pressure methane sorption isotherms of SPE Journal 2013, 18 (5), 924−931.
black shales from the Netherlands. Energy Fuels 2012, 26, 4995−5004. (25) Sigal, R. F.; Akkutlu, I. Y.; Kang, S. M.; Diaz-Campos, M.;
(5) Gasparik, M.; Bertier, P.; Gensterblum, Y.; Ghanizadeh, A.; Ambrosse, R. J. The Laboratory Measurement of Gas Storage Capacity
Krooss, B. M.; Littke, R. Geological controls on the methane storage of Organic Shales. Petrophysics 2013, 54, 224−235.
capacity in organic-rich shales. Int. J. Coal Geol. 2014, 123, 34−51. (26) Kang, S. M.; Shinn, Y. J.; Akkutlu, I. Y. Gas Storage Capacity of
(6) Guo, S. Experimental study on isothermal adsorption of methane Iljik and Hasandong Shales in Gyongsang Basin, South Korea.
gas on three shale samples from Upper Paleozoic strata of the Ordos Proceedings of the Unconventional Resources Technology Conference;
Basin. J. Pet. Sci. Eng. 2013, 110, 132−138. Denver, CO, Aug 25−27, 2014; DOI: 10.15530/URTEC-2014-
(7) Ross, D. J. K.; Bustin, R. M. Characterizing the shale gas resource 1922165.
potential of Devonian Mississippian strata in the Western Canada (27) Jarvie, D. M.; Hill, R. J.; Ruble, T. E.; Pollastro, R. M.
sedimentary basin: Application of an integrated formation evaluation. Unconventional shale-gas systems: The Mississippian Barnett Shale of
AAPG Bull. 2008, 92, 87−125. north-central Texas as one model for thermogenic shale-gas
(8) Ross, D. J. K.; Bustin, R. M. The importance of shale composition assessment. AAPG Bull. 2007, 91, 475−499.
and pore structure upon gas storage potential of shale gas reservoirs. (28) Xiong, F.; Jiang, Z.; Li, P.; Wang, X.; Bi, H.; Li, Y.; Wang, Z.;
Mar. Pet. Geol. 2009, 26, 916−927. Amooie, M. A.; Soltanian, M. R.; Moortgat, J. Pore structural of
(9) Liu, G. H.; Huang, Z. L.; Jiang, Z.; Chen, J.; Chen, F.; Xing, J. Gas transitional shales in the Ordos Basin, NW China: Effects of
adsorption capacity calculation limitation due to methane adsorption
composition on gas storage capacity. Fuel 2017, 206, 504−515.
in low thermal maturity shale: A case study from the Yanchang
(29) Curtis, M. E.; Cardott, B. J.; Sondergeld, C. H.; Rai, C. S.
Formation, Ordos Basin. J. Nat. Gas Sci. Eng. 2016, 30, 106−118.
Development of organic porosity in the Woodford Shale with
(10) Montgomery, S. L.; Jarvie, D. M.; Bowker, K. A.; Pollastro, R.
increasing thermal maturity. Int. J. Coal Geol. 2012, 103, 26−31.
M. Mississippian Barnet Shale, Fort Worth basin, north-central Texas:
(30) Ji, L. M.; Zhang, T.; Milliken, K. L.; Qu, J.; Zhang, X.
Gas-shale play with multi-trillion cubic foot potential. AAPG Bull.
2005, 89, 155−175. Experimental investigation of main controls to methane adsorption in
(11) Wu, Y.; Fan, T.; Jiang, S.; Yang, X.; Ding, H.; Meng, M.; Wei, D. clay-rich rocks. Appl. Geochem. 2012, 27, 2533−2545.
Methane adsorption capacities of the Lower Paleozoic Marine Shales (31) Chalmers, G. R. L.; Bustin, R. M. The organic matter
in the Yangtze Platform, South China. Energy Fuels 2015, 29 (7), distribution and methane capacity of the Lower Cretaceous strata of
4160−4167. Northeastern British Columbia, Canada. Int. J. Coal Geol. 2007, 70,
(12) Yang, F.; Ning, Z. F.; Zhang, R.; Zhao, H.; Krooss, B. M. 223−239.
Investigations on the methane sorption capacity of marine shales from (32) Chalmers, G. R. L.; Bustin, R. M. Lower Cretaceous gas shales
Sichuan Basin, China. Int. J. Coal Geol. 2015, 146, 104−117. in northeastern British Columbia, Part I: Geological controls on
(13) Pan, Z.; Connell, L. D. Reservoir Simulation of free and methane sorption capacity. Bull. Can. Pet. Geol. 2008, 56, 1−21.
adsorbed gas production from shale. J. Nat. Gas Sci. Eng. 2015, 22, (33) Kim, J.; Kim, D.; Lee, W.; Lee, Y.; Kim, H. Impact of total
359−370. organic carbon and specific surface area on the adsorption capacity in
(14) Hu, H. Methane adsorption comparison of different thermal Horn River shale. J. Pet. Sci. Eng. 2017, 149, 331−339.
maturity kerogens in shale gas system. Chin. J. Geochem. 2014, 33 (4), (34) Wang, Y.; Zhu, Y.; Liu, S.; Zhang, R. Pore characterization and
425−430. its impact on methane adsorption capacity for organic-rich marine
(15) Ross, D. J. K.; Bustin, R. M. Shale gas potential of the lower shales. Fuel 2016, 181, 227−237.
Jurassic Gordondale member, northeastern British Columbia, Canada. (35) Xia, J.; Song, Z.; Wang, S.; Zeng, W. Preliminary study of pore
Bull. Can. Pet. Geol. 2007, 55, 51−75. structure and methane sorption capacity of the Lower Cambrian shales
(16) Tian, H.; Li, T.; Zhang, T.; Xiao, X. M. Characterization of from the north Gui-Zhou Province. J. Nat. Gas Sci. Eng. 2017, 38, 81−
methane adsorption on over mature Lower Silurian − Upper 93.

11708 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
Energy & Fuels Article

(36) Mendhe, V. A.; Mishra, S.; Varma, A. K.; Kamble, A. D.; matter accumulation in the Qingshankou Formation, Upper Creta-
Bannerjee, M.; Sutay, T. Gas reservoir characteristics of the Lower ceous, Songliao Basin (NE China). Int. J. Coal Geol. 2015, 141−142,
Gondwana Shales in Raniganj Basin of Eastern India. J. Pet. Sci. Eng. 23−32.
2017, 149, 649−664. (58) Cao, T.; Song, Z.; Wang, S.; Xia, J. A comparative study of the
(37) Zhang, Y.; Shao, D.; Yan, J.; Jia, X.; Li, Y.; Yu, P.; Zhang, T. The specific surface area and pore structure of different shales and their
pore size distribution and its relationship with shale gas capacity in kerognes. Sci. China: Earth Sci. 2015, 58, 510−522.
organic-rich mudstone of Wufeng-Longmaxi Formations, Sichuan (59) Gasparik, M.; Rexer, T. F. T.; Aplin, A. C.; Billemont, P.; De
Basin, China. J. Nat. Gas Geosci. 2016, 1, 213−220. Weireld, G.; Gensterblum, Y.; Henry, M.; Krooss, B. M.; Liu, S.; Ma,
(38) Chen, L.; Jiang, Z.; Liu, K.; Ji, W.; Wang, P.; Gao, F.; Hu, T. X.; Sakurovs, R.; Song, Z.; Staib, G.; Thomas, K. M.; Wang, S.; Zhang,
Application of Langmuir and Dubinin-Radushkevich models to T. First international inter-laboratory comparison of high-pressure
estimate methane sorption capacity on two shale samples from the CH4, CO2 and C2H6 sorption isotherms on carbonaceus shales. Int. J.
Upper Triassic Chang 7 Member in the southeastern Ordos Basin, Coal Geol. 2014, 132, 131−146.
China. Energy Explor. Exploit. 2017, 35, 122−144. (60) Li, P.; Jiang, Z.; Zheng, M.; Bi, H.; Chen, L. Estimation of shale
(39) Gensterblum, Y.; Busch, A.; Krooss, B. M. Molecular concept gas adsorption capacity of the Longmaxi Formation in the Upper
and experimental evidence of competitive adsorption of H2O, CO2 Yangtze Platform, China. J. Nat. Gas Sci. Eng. 2016, 34, 1034−1043.
and CH4 on organic material. Fuel 2014, 115, 581−588. (61) Murillo, C.; Gomez, O.; Ortiz Cancino, O.; Muñoz, S.
(40) Barrero, D.; Pardo, A.; Vargas, C.; Martínez, J. Colombian Aplicación de modelos para la generación de la isoterma de adsorción
Sedimentary Basins: Nomenclature, Boundaries and Petroleum Geology, a de metano en una muestra de shale y su impacto en el cálculo de
New Proposal; Agencia Nacional de Hidrocarburos (ANH): Bogotá, reservas. Revista Fuentes 2015, 13 (2), 131−140.
Colombia, 2007; pp 92.
(41) Chung, F. H. Quantitative interpretation of X-ray diffraction
patterns of mixtures. II. Adiabatic principle of X-ray diffraction analysis
of mixtures. J. Appl. Crystallogr. 1974, 7, 526−531.
(42) Brunauer, S.; Emmett, P. H.; Teller, E. Adsorption of gases in
multimolecular layers. J. Am. Chem. Soc. 1938, 60, 309−319.
(43) Heller, R.; Vermylen, J.; Zoback, M. Experimental investigation
of matrix permeabilitiy of gas shales. AAPG Bull. 2014, 98, 975−995.
(44) Pino, D.; Bessieres, D. CH4/CO2 Mixture Adsorption on a
Characterized Activated Carbon. J. Chem. Eng. Data 2017, 62, 1475−
1480.
(45) Wagner, W.; Span, R. Special Equations of State for Methane,
Argon and Nitrogen for the temperature range from 270 to 350K at
pressures up to 30 MPa. Int. J. Thermophys. 1993, 14 (4), 699−725.
(46) Hou, Y. G.; He, S.; Yi, J. Z.; Zhang, B.; Chen, X.; Wang, Y.;
Zhang, J.; Cheng, C. Y. Effect of pore structure on methane sorption
capacity of shales. Pet Explor Dev. 2014, 41, 272−281.
(47) Chareonsuppanimit, P.; Mohammad, S. A.; Robinson, R. L., Jr.;
Gasem, K. A. M. High- pressure adsorption of gases on shales:
Measurements and modeling. Int. J. Coal Geol. 2012, 95, 34−46.
(48) Rexer, T. F. T.; Benham, M. J.; Aplin, A. C.; Thomas, K. M.
Methane adsorption on shale under simulated geological temperature
and pressure conditions. Energy Fuels 2013, 27, 3099−3109.
(49) Dreisbach, F.; Staudt, R.; Keller, J. U. High Pressure Adsorption
Data of Methane, Nitrogen, Carbon Dioxide and their Binary and
Ternary Mixtures on Activated Carbon. Adsorption 1999, 5, 215−227.
(50) Yang, F.; Ning, Z. F.; Liu, H. Q. Fractal characteristics of shales
from a shale gas reservoir in the Sichuan Basin, China. Fuel 2014, 115,
378−384.
(51) Gensterblum, Y.; VanHemert, P.; Billemont, G.; Busch, A.;
Charriere, D.; Li, D.; Krooss, B. M.; deWeireld, G. D.; Prinz, K.; Wolf,
H. A. A. A European inter-laboratory comparison of high pressure
CO2 sorption isotherms. I: Activated Carbon. Carbon 2009, 47, 2958−
2969.
(52) Yuan, W.; Pan, Z.; Li, X.; Yang, Y.; Zhao, C.; Connell, L. D.; Li,
S.; He, J. Experimental study and modelling of methane adsorption
and diffusion in shale. Fuel 2014, 117, 509−519.
(53) Pozo, M.; Pino, D.; Bessieres, D. Effect of thermal events on
maturation and methane adsorption of ammonium Illite-rich Silurian
black shales (Checa, Spain). Appl. Clay Sci. 2017, 136, 208−218.
(54) McCarthy, K.; Rojas, K.; Niemann, M.; Palmowski, D.; Peters,
K.; Stankiewicz, A. Basic Petroleum Geochemistry for Source Rock
Evaluation. Oilfield Rev. 2011, 23, 32−43.
(55) Peters, K. E. Guidelines for evaluating petroleum source rock
using programmed pyrolysis. AAPG Bull. 1986, 70, 328−329.
(56) Gorin, G. E.; Feist-Burkhardt, S. Organic facies of Lower to
Middle Jurassic sediments in the Jura Mountains, Swizerland. Rev.
Palaeobot. Palynol. 1990, 65, 349−355.
(57) Xu, J.; Bechtel, A.; Sachsenhofer, R. F.; Liu, Z.; Gratzer, R.;
Meng, Q.; Song, Y. High resolution geochemical analysis of organic

11709 DOI: 10.1021/acs.energyfuels.7b01849

Energy Fuels 2017, 31, 11698−11709
 Journal of Petroleum Science and Engineering 159 (2017) 307–313

Contents lists available at ScienceDirect

Journal of Petroleum Science and Engineering

journal homepage: www.elsevier.com/locate/petrol

Adsorption of pure CO2 and a CO2/CH4 mixture on a black shale sample:

Manometry and microcalorimetry measurements
Olga Patricia Ortiz Cancino a, *, David Pino Perez b, Manuel Pozo c, David Bessieres b
a
Universidad Industrial de Santander- GMPH, Bucaramanga, Colombia
b
Universite de Pau et des Pays de l’Adour, UMR 5150, LFC-R- Laboratoire des Fluides Complexes et leurs Reservoirs, 64013 Pau Cedex, France
c
Department of Geology and Geochemistry, Universidad Autonoma de Madrid, Cantoblanco 28049 Madrid, Spain

A R T I C L E I N F O A B S T R A C T

Keywords: The study of CO2/CH4 adsorption onto kerogen is relevant for shale gas production. Despite much expanded
Black shale literature, reliable adsorption models still await for a complete description due to the complexity of kerogen. The
Kerogen objective of this study is to provide an original set of experimental data and to use the selectivity as an indicator to
Adsorption test the affinity of the sample for a component over another. The adsorption of pure CO2 and the equimolar
CH4 mixture CH4/CO2 were explored on a Silurian black shale sample. This outcropping sample was collected in a
CO2
formation, which is not considered as target for shale gas exploration. However its geochemical profile as well as
Selectivity
its thermal maturity suggest this sample is an ideal candidate to study kerogen/carbon dioxide and methane
interactions. Both isotherm and enthalpy of adsorption of carbon dioxide were measured up to 3.2 (MPa) by the
use of a combined manometric-calorimetric device. The carbon dioxide isotherm was fitted with a modified
Langmuir model allowing the determination of the adsorption uptake. The heat of adsorption is an indicator of the
affinity of the carbon dioxide with kerogen. Additionally the equimolar mixture methane/carbon dioxide
isotherm was performed up to 2 (MPa) by use of a device specially developed and built for gas mixture co-
adsorption. The adsorption of each component within the mixture was provided. The estimated selectivity
CO2/CH4 highlights a significant affinity of CO2 with the kerogen.

1. Introduction that methane is adsorbed onto the organic phase -known as

kerogen-whereas the adsorbed gas in the inorganic matrix is generally
The still growing demand for energy has stimulated the exploitation supposed to be negligible (Collell et al., 2014; Ross and Bustin, 2009). In
of unconventional gas reservoirs like shale gas. The natural gas produced addition to natural gas interests, more recent studies have evaluated the
from organic shale formations is known as shale gas. The total amount of potential of shales for geological CO2 storage (Charoensuppanimit et al.,
natural gas is composed by three sources, i.e., the adsorbed gas, the free 2016; Heller and Zoback, 2014). This is a remarkable observation not
gas and the gas dissolved in pore fluids, the last one being negligible and only for storage but also from enhanced shale gas production point of
consequently the total shale gas content can be considered as the sum of view as it implicitly points out that stronger affinity of CO2 to the organic
adsorbed and free gas (Guo, 2013; Gasparik et al., 2012). The adsorbed materials of shale could initiate an added mechanism of displacement of
gas ranges from 20 to 80% of total gas reserves as well as recovery rates the originally in-place CH4, when CO2 is introduced into the shale gas
(Yu et al., 2014), for this reason adsorption is a leading criterion gov- environment. The design of the CO2 sequestration process requires
erning shale gas production. In an effort to better understand the role of knowledge of the adsorption behavior of CO2 (Chareonsuppanimit et al.,
adsorption on production from gas shales, numerous authors have done 2012). These studies showed that adsorption capacity of CO2, as well as
valuable contributions to the literature through laboratory measure- methane, appear to be related to the total organic carbon. Char-
ments. Much research focused on the estimation of the sorption capacity oensuppanimit et al. (2016) focused on Woodford shales from Payne and
of methane, the most abundant component of shale gas, at defined Hancock counties. For each shale sample, CO2 adsorption was 2.1–8.6
borehole conditions (Clarkson and Bustin, 2000; Gasparik et al., 2014a, times higher than methane adsorption. Adsorption capacity was found
2014b, 2012; Guo, 2013; Mendhe et al., 2017). Most of the studies point correlated to the TOC content of these samples. Kang et al. (2011)

* Corresponding author.
E-mail address: oportizc@uis.edu.co (O.P. Ortiz Cancino).

https://doi.org/10.1016/j.petrol.2017.09.038
Received 20 March 2017; Received in revised form 14 August 2017; Accepted 18 September 2017
Available online 21 September 2017
0920-4105/© 2017 Elsevier B.V. All rights reserved.
O.P. Ortiz Cancino et al. Journal of Petroleum Science and Engineering 159 (2017) 307–313

identified that micropores in organic matter acted as molecular sieves each adsorption measurement, the sample was under additional vacuum
that allowed for only linear molecules, such as CO2, to access their pore at 110
C for 18 h, in order to withdraw any trace of fluid.
space. Despite these studies, gas adsorption, storage and diffusion in
organic-rich shales is a complex multiparameter process that still awaits 2.2. Calorimetric-manometric sorption set up
for reliable description in order to elucidate the effect of individual pa-
rameters. Indeed, kerogen is a very complex structure depending on both CO2 adsorption uptake was measured using a manometric-
its origin and thermal alteration. Various realistic models have been calorimetric device. The differential heat flux calorimeter used is a
proposed to represent the physical and chemical properties of kerogen Setaram C80 Tian-Calver. The inner part of the calorimeter include two
through the use of molecular simulations (Collell et al., 2014, 2009; Sui calorimetric cells (reference and adsorption) surrounded by a high
and Yao, 2016). In the latter, Sui and Yao (2016) analysed the effect of thermal mass aluminium block. The sample is located in the adsorption
surface chemistry for CH4/CO2 adsorption in kerogen. The authors pro- cell (see (3) in Fig. 1), which is connected to manometric apparatus
posed a 3-D molecular model of kerogen and they studied the CO2/CH4 inserted in the upper part of the calorimeter (Fig. 1). The same temper-
interactions with kerogen. Both for pure CO2, CH4 or their equimolar ature is settled for the calorimeter and the manometric system allowing
mixture, the maximum adsorption of CO2 in kerogen is larger than CH4 isothermal conditions in both parts of the coupled apparatus. The main
adsorbed in kerogen. Additionally, the simulated isosteric heat of added value of this apparatus is the simultaneous determination of the
adsorption display higher values for CO2/kerogen than for CH4/kerogen. differential heat of adsorption which is a direct measurement of the
This is explained by the Coulomb and van der Waals interactions between adsorbate/adsorbent interactions. Note that the reference cell is used to
CO2 and kerogen which play a key role in the process of adsorption, compensate the heat flux in a blank experiment. A schematic diagram of
whereas, in the CH4 adsorption process, there was a little Coulomb the set-up is provided in Fig. 1.
interaction between CH4 and kerogen.
In this context, the purpose of the present work should be regarded as 2.3. Determination of the isotherm and heat of adsorption of CO2
a step towards the understanding of the CO2/CH4 interactions with
kerogen. Our experimental study focused on a black shale sample The sorption isotherms obtained by manometric measurements are
outcropping in the Iberian Range from Central Spain in the province of excess sorption isotherms. To get more information on this concept, also
Guadalajara. Its choice was dictate due to its basic geochemical profile called Gibbs surface excess, the reader can refer to the following work
and its high organic content which propose this sample as an ideal (Sircar, 1999). A mass of sample (2 gr) is placed into the adsorption cell
candidate to study the adsorption and interaction mechanism of both (see (3) in Fig. 1) under vacuum. After, a quantity of gas is introduced in
CH4 and CO2 with the kerogen. A complete characterization of this the dosing volume (see (2) in Fig. 1.) of known volume. The initial
kerogen was performed on a previous work (Pozo et al., 2017). In the pressure P1 and temperature T1 are measured after reaching thermal and
present study, sorption isotherms and heat of sorption for carbon dioxide mechanical equilibria. These measurements provide the initial mole
CO2 were measured on this selected black shale whereas the CH4
adsorption was previously measured (Pozo et al., 2017). Additionally,
the selectivity of kerogen for CH4/CO2 was studied from the adsorption
measurement of the equimolar CH4-CO2 mixture at 50
C up to 2 (MPa).
To the best to our knowledge, such set of meaningful data was never
presented up to the date. The experimental selectivity CO2/CH4 in
kerogen is the main novelty of the work. Its value will provide original
information to test and discriminate the realistic models proposed for the
kerogen description.
The work is organized as follows. Section 2 is devoted to the
description of both experimental setup and measurements principles. A
fit-to-purpose calorimetric-manometric system was used to determine
simultaneously adsorption isotherms and heat of adsorption of pure
compounds. This apparatus can operate over broad ranges of pressure
(0–3.5 MPa). The study of adsorption in gas mixture is much complicated
due to the gas diffusion phenomenon. Therefore, a device was specif-
ically designed and built for gas mixture (Pino et al., 2014) which con-
sists of a dynamic manometric measure combined with an analytical
technique. As the description of these instruments together with theirs
measurement principles are extensively detailed elsewhere (Mouahid
et al., 2012a, 2012b; Pino et al., 2014) we only recall theirs most sig-
nificant features. In Section 3, we provide the experimental results for the
pure compounds and for the equimolar mixture. The selectivity of
kerogen with CO2 and CH4 is discussed on the basis of the results ob-
tained for the deduced separation factor. Conclusions of this work are
depicted in Section 4.

2. Experimental techniques

2.1. Sample preparation

Sorption isotherms were measured on a dry shale sample, which was

received in several pieces of irregular shape; the average size of each
piece was about 5 mm. Two grams of sample were dried under vacuum
into the experimental cells at 110
C during 18 h. This procedure is
necessary to remove any moisture taken up by the sample. Later, after Fig. 1. Manometric-calorimetric device (Mouahid et al., 2012b).

308
O.P. Ortiz Cancino et al. Journal of Petroleum Science and Engineering 159 (2017) 307–313

number of gas n1. Then, an expansion is performed into the whole system The heat of compression δQcomp sums two contributions of opposite
(dosing volume and adsorption cell). As in the first step, the pressure P2 effect: one resulting from the gas and the other from the vessel wall:
and temperature T2 (T2 ¼ T1) are measured after equilibrium and the
mole number remaining in the gas phase n2 is estimated. The adsorbed
mole corresponds to (n1 –n2). The isotherm is described repeating the δQcomp ¼ αSS VE Tdp  αp VE Tdp (4)
same operations through an accumulative process. A previous measuring In Eq. (4) αSS is the isobaric coefficient of stainless steel in K (the 1
cell volume determination is done using Helium expansion. The molar material of which the adsorption cell is made), T the temperature in K,
volumes involved in the amount adsorbed calculations were determined and VE is the volume taken into account by the thermopiles deduced by
thanks to the NIST data (Span and Wagner, 2003, 1996) at the considered helium calibration:
experimental conditions.
k∫ EðtÞdt
2.3.1. Adsorbed amount uncertainties VE ¼ (5)
ðαPHe  αSS ÞTΔp
After each expansion, the return to the thermodynamic equilibrium
was controlled by both the pressure and calorimetric signal reached with Combining Eq. (1) to Eq. (5) the heat of adsorptionΔH (J mol1) is
kinetics ranging from 45 to 60 min. estimated to be:
The pressure is measured by a pressure transducer MKS Baratron type
121A, with an accuracy of 0.01% of the full scale (3 MPa). For temper- Qads k∫ EðtÞdt þ VE ðαP  αSS ÞTΔp
ΔH ¼ ¼ (6)
ature, the uncertainty is about 0.1 K (temperature sensor, Pt100 is Δn Δn
located in the aluminium block of the calorimeter). The uncertainties on Expression in which Δn is the amount of gas adsorbed between each
the dosing and measuring cell volumes have been estimated to less than step Δp.
1%. The computed uncertainty on the amount adsorbed shows that the
main contribution comes from the error on the pressure measurement 2.3.3. Uncertainties
(Belmabkhout et al., 2004). This global uncertainty is estimated to The procedure used to determine VE leads to an uncertainty of ±2%
be Δn
n ¼ 1%. on Qcomp. The error associated to the total heat Q estimated by a test of
reproducibility is about 1%. Combining errors on both Qads and nads, the
2.3.2. Heat of adsorption error on the differential enthalpy of adsorption is about 5%, strongly
The heat of adsorption or differential enthalpy of adsorption ΔH (in J affected by the value of nads. The detailed procedure including both the
mol1), is also estimated at each step of the isotherm described. The experimental procedure and the uncertainties estimations can be found
calorimetric signal is representative of the total heat liberated Q, which is elsewhere (Bessieres et al., 2005; Mouahid et al., 2012a, 2012b).
the result of two contributions: the heat of adsorption Qads and the heat
liberated by the compressed (or expanded) gas Qcomp. This last contri-
bution is previously determined through Helium expansions. 2.4. Co-adsorption CO2/CH4 measurement

Q ¼ Qads þ Qcomp (1)

For the mixture, we used different pieces of the same sample. The
As a consequence, the estimation of the heat of adsorption Qads re- equimolar CO2-CH4 mixture adsorption was measured using a device
quires the calculation of the heat of compression, the total heat Q being built for this purpose as is represented in Fig. 2. The principle remains
estimated through the integration of the calorimetric signal E: identical to the pure component adsorption one (Ghoufi et al., 2009). It
allows the measurements of isotherm mixture adsorption in a range of
Q ¼ K∫ EðtÞdt (2) pressures from 0 to 3 MPa and for temperatures between 303.15 and
1 423.15 (K). A circulation pump is used to homogenize the mixture and
Expression in which K is the calorimetric detector (W⋅μV ), previ-
reduces the time needed to reach equilibrium. The manometric system is
ously measured (Bessieres et al., 2005, 2000). From other part, the heat
coupled with a gas chromatograph (Agilent) which allows the determi-
of compression Qcomp is determined through the use of the following
nation of the gas mole fraction of each component in the mixture. The
Maxwell relationship:
pressure transducer is also a MKS Baratron type 121A. To guarantee the

 isothermal conditions is used a heater wire controlled by a PID regulator
∂S
¼ ∝P ⋅V (3) (Eurotherm 3208); five thermocouples (type K) were installed in
∂p T
different parts of the circuit to ensure that there is not a temperature
gradient during the measurements. This instrument has been detailed in
where αP is the isobaric expansion coefficient defined by 1/V (∂V/∂T)P.
a previous work (Pino et al., 2014).

Fig. 2. Schematic diagram of the manometric dosing system.

309
O.P. Ortiz Cancino et al. Journal of Petroleum Science and Engineering 159 (2017) 307–313

2.4.1. Measurement principles 3. Results and discussion

The procedure is analogous for pure gas. A mass of the sample (2 gr) is
placed into the adsorption cell (see Fig. 2) of known volume, under 3.1. Black shale sample
vacuum. The accessible volume is determined by successive Helium ex-
pansions after what shale outgassing is carried out. Gas mixture, which is The shale sample named CH-1 was collected near Checa town (Gua-
previously analysed using the chromatograph, is introduced in the dosing dalajara province) in the Iberian Range from Central Spain. The shale is
cell without going through the adsorption cell (see Fig. 2) until equilib- black in color, presents lamination to slaty cleavage and contains abun-
rium is reached (P1 and T1 are measured and initial mole number of gas dant graptolite fossils dating them as Silurian (Gutierrez-Marco and
n1 is calculated). In the next step, the gas expands on the whole system Storch, 1998). A detailed compositional analysis of this black shale
(dossing cell and adsorption cell, of known volumes). When both ther- including kerogen type and maturation was performed in a previous
mal, mechanical equilibria and homogenization are reached (checked by work aimed to correlate their structural and geochemical properties and
constant values of both pressure and composition) values of P2 and T2 the methane uptakes of different shale samples (Pozo et al., 2017). This
(T2 ¼ T1) are measured and the mole number remaining in gas phase n2 is sample consists mostly of phyllosilicates (94%) with minor contents of
estimated. A discrete quantity of gas (0.5 cm3) is extracted from the quartz (3%), potassium feldspar (3%) and rutile (traces) as is shown in
isolated dosing volume and analysed with the chromatograph. This Table 1. Within phyllosilicates predominate illite-mica and mixed-layer
measure provides the composition of the gas phase from which can be illite-smectite, in traces (<5%) chlorite, kaolinite and pyrophyllite were
deduced the composition of the adsorbed phase. The total amount also identified (Fig. 3). The mixed-layer illite-smectite shows a high de-
adsorbed is calculated by a mass balance (n1-n2). This procedure used to gree of ordering (R1) with sharp 001 and 002 reflections at 24.5 Å and
calculate the amounts of each individual components adsorbed was 12.5 Å respectively, suggesting a rectorite-like clay mineral. A broad
detailed in Pino et al. (2014). Basically, we wait for the equilibrium about reflection was observed for illite-mica with two main maximum at
60 min and we perform several analyses to check reproducibility. The 10.11 Å and 9.98 Å indicating coexistence of NHþ 4 -rich illite and common
same operation is repeated through an accumulative process to describe illite (Daniels and Altaner, 1990). In randomly oriented samples the
the isotherm. The uncertainty on the adsorbed amount is estimated to be value of d(060) reflection is 1.49–1.50 Å indicating dioctahedral
less than 3%. clay minerals.
The oriented aggregates of the clay fraction (<2 μm) corroborate the
above mentioned mineralogy highlighting the swelling under ethylene
Table 1
glycol solvation of the regular mixed-layer illite-smectite toward 13 and
Characterization of black shale (CH-1). 27 Å, a typical behavior of rectorite (Srodon, 1980). The identified
mineralogical assemblage is similar to that previously reported in shales
Clay minerals (wt %)a 94
Non-clay minerals (wt%)b 6 from the same lithological formation by Bauluz and Subías (2010). Ac-
C (wt%) 8.60 cording to these authors the occurrence of NH4-rich illite and pyro-
H (wt%) 1.23 phyllite result mostly from the circulation of hydrothermal fluids related
N (wt%) 0.64
to andesitic sills intruding the black shales. The content of organic carbon
S (wt%) 0.09
Specific surface BET (m2/g) 34
(TOC) reaches 8.6% and the concentration of H and N is related partially
t-plot external surface (m2/g) 21 with the presence of NHþ 4 in illite interlayer (See Table 1).
t-plot micropores surface (m2/g) 12
Micropore average size (Å) 7.6 3.1.1. Kerogen maturity
a
Illite (K, NH4þ)-rectorite > (chlorite-kaolinite-pyrophyllite). A Rock- Eval analysis shows that the kerogen used in this study is type
b
Quartz, K-feldspar. III. It should be more a source of gas than oil. Here, the HI/OI ratio
converges towards the origin of the plot, leading to a type III

Fig. 3. X-ray diffraction pattern showing the bulk mineralogy of sample CH-1. (I/S) illite-smectite mixed layer. (Ilt) illite. (Prl) pyrophyllite. (Kln) kaolinite, (Chl) chlorite. (Qz) quartz.
(Fsp) K-feldspar.

310
O.P. Ortiz Cancino et al. Journal of Petroleum Science and Engineering 159 (2017) 307–313

interpretation. The maturation of the sample was forced by thermal 423 kg/m3 for methane (Gasparik et al., 2012; Pozo et al., 2017) and
events reaching an over-mature stage. This interpretation is also sup- 1027 kg/m3 for CO2 (Gensterblum et al., 2013). Fitting parameters are
ported by the very low hydrocarbon yields (S1, S2) despite their rela- shown in Table 3, and in Fig. 4 can be observed the isotherm obtained
tively high TOC content (4.4–7.62%). with these values. Note that the values obtained agree with literature
The adsorption isotherm belongs to IUPAC type IV, characteristic of a values obtained for kerogen type III (Zhang et al., 2012; Gasparik
wide distribution ranging from micro to mesoporous mate- et al., 2014a).
rials (2–50 nm).
CH4 and CO2 were provided by Linde Gas with a minimum purity of
3.3. Differential heat of adsorption
99.995%. The equimolar mixture CH4/CO2 was also provided by Linde
Gas. Its accuracy claimed to be better than 1% was previously checked by
The calorimetric–manometric allows a simultaneous measurement of
gas chromatograph analysis.
the differential enthalpy of sorption when describing the isotherms. The
values obtained for both methane (Pozo et al., 2017) and carbon dioxide
3.2. Pure carbon dioxide adsorption isotherm are reported in Table 4. In Fig. 5, are shown the values of differential
enthalpy of adsorption, ΔHads. The average value for methane reaches
CO2 adsorption measurement was performed at 323K, and pressures 30 kJ/mol whereas for carbon dioxide it is nearly 40 kJ/mol. In both
in a range from atmospheric pressure to 3 MPa. We also reproduced the cases, the values are nearly constant considering the uncertainties esti-
CH4 adsorption isotherm, already studied in a previous work (Pozo et al., mations. The behavior for CH4 and CO2 is consistent with physisorption
2017) in which the effect of thermal events on various shales samples was because their differential heat are lower than the minimum values re-
depicted. Due to the limitations of the own techniques, reliable data are ported for chemisorption.
accessible up to moderate pressures. In this context, an alternative is to The heat of adsorption of CO2 is larger than CH4. This results suggest a
perform measurements with high accuracy up to moderate pressures (up better affinity of the CO2 with kerogen considering that the heat or
to 3 MPa) and then fit the experimental data to a model as Langmuir, in enthalpy of adsorption is directly linked to the interactions between
order to obtain information to higher pressures; this procedure was fol- kerogen and the species. Indeed, the enthalpy profiles nearly constant
lowed on our work, the results are shown later. The present study focuses can be also regarded as the competitive effect between two opposite
on CH4/CO2 interactions with kerogen with the idea to provide a contributions. At low pressure the molecules are adsorbed to the high
meaningful set of data for the kerogen characterization.
Fig. 4 displays the pure CH4 and CO2 isotherms. These results, listed
in Table 2, shows the CH4 and CO2 adsorption isotherms exhibit Lang-
muir type I behavior, which is a typical for microporous materials. This is
consistent with the fact that adsorption occurs mainly onto kerogen. CO2
adsorption is about three times larger than CH4. At low pressures, the
slope of CO2 isotherm is higher than CH4. This slope directly correlated to
the Henry constant law (KH) is a first indicator of the higher affinity of
CO2 with kerogen. With increasing pressure the slope decrease as the
shale sample approaches saturation, in the case of methane the decrease
adsorption rate is faster than CO2 which is consistent with the higher CO2
uptake. One can note that these values of CO2 uptake are very similar to
few values reported in the literature (Chareonsuppanimit et al., 2012;
Heller and Zoback, 2014).

3.2.1. Parameterization of experimental results

To rationalize these results, the isotherm data were correlated by a
Fig. 4. Excess sorption isotherms of pure components and with fitted three parameters
Langmuir model of three parameters described by Gensterblum et al. function (filled circle: CH4 excess sorption obtained from Pozo et al., 2017, times: CH4
(2009) and applied by Gasparik et al. (2012): excess sorption obtained in this work, filled triangle: CO2 excess sorption, line: fit).



p ρg ðp; TÞ ρg ðp; TÞ
nexcess ¼ nL ⋅ 1 ¼ nabsolute ⋅ 1 (8) Table 2
ads
p þ pL ρads ads
ρads Adsorption data for CO2 and CH4 pures.

CH4a CH4 (this work) CO2

Expression, in which:
Pressure nexc(mol/Kg) Pressure nexc(mol/Kg) Pressure nexc
(MPa) (MPa) (MPa) (mol/Kg)
 nexcess
ads (mol/Kg): indicates the excess amount adsorbed at the
0.27 0.0372 0.31 0,0478 0.28 0.1812
measured pressure P (MPa).
0.65 0.0736 0.70 0,0810 0.66 0.2499
 pL (MPa): Langmuir pressure. The pressure at which half of the 1.06 0.0964 1.11 0,1003 1.03 0.2898
adsorption sites are occupied. 1.49 0.1124 1.57 0,1128 1.41 0.3156
 nL (mol/Kg): Maximum Langmuir capacity. 1.85 0.1185 1.96 0,1191 1.79 0.3338
2.20 0.1218 2.33 0,1230 2.19 0.3480
2.57 0.1245 2.70 0,1254 2.61 0.3593
A standard deviation was calculated according to equation (9).
2.93 0.1254 3.07 0,1266 2.97 0.3648
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
XN
1 a
Pozo et al., 2017.
Δn ¼ ⋅ 1
ðn  nFIT Þ2 (9)
N
Table 3
The Langmuir sorption model, developed for low pressures, repre-
Langmuir parameters with ρads fixed.
sents, however, a reasonable approximation of the measured excess
Component pL (MPa) nL (mol/kg) ρads (Kg/m3) Δn
sorption isotherms and can thus be used as a fitting function. The main
concern to apply this approach is the estimation of the maximum Lang- Methanea 1.00 0.222 423 0.0276
muir capacity, which provide a value of the CO2 adsorption capacity, in Carbon dioxide 0.45 0.435 1027 0.0025

the case of this study 0.435 (mol/kg). In this work, for ρads we used a
Pozo et al., 2017.

311
O.P. Ortiz Cancino et al. Journal of Petroleum Science and Engineering 159 (2017) 307–313

Table 4
Differential enthalpy of adsorption ΔH ads (KJ/mol).
a
CH4 CH4 (this work) CO2

Pressure (MPa) ΔH ads (KJ/mol) Pressure (MPa) ΔH ads (KJ/mol) Pressure (MPa) ΔH ads (KJ/mol)

0.10 28 0.31 25 0.28 33

0.55 35 0.70 35 1.03 44
0.90 30 1.11 28 1.41 36
a
Pozo et al., 2017.

energy positions where as the fluid/fluid interactions play an important compounds. All the data are shown in Fig. 6 and summarized in Table 5.
role when pressure is increasing. To the best of our knowledge, these The individual gas isotherms are calculated from the mole fraction of
differential heat of adsorption of CO2 onto kerogen are the first experi- each gas in the bulk gas phase. As expected, pure CO2 in gas mixture is
mental values reported in the literature. This meaningful information much preferentially adsorbed than CH4. When one compare these iso-
obtained on a well-characterized kerogen can be used to discriminate the therms with those of the single gas component, it is relevant that the CO2
different kerogen approaches proposed in the literature (Collell et al., adsorbed amount is affected by the presence of CH4 in the gas phase but
2014; Sui and Yao, 2016). Using a more realistic model of kerogen, the CH4 amount decreases more than CO2. All the data for the pure
molecular simulations should be performed to estimate the CH4 or CO2 component in gas mixture are also summarized in Table 5. Additionally,
uptakes at borehole conditions. the selectivity of CO2 over CH4 was calculated from equation (10) and
reported in Fig. 7 as a function of pressure:
3.4. Equimolar mixture adsorption y1=y
2
Selectivity ¼ (10)
x1=x
Adsorption isotherm was measured for the equimolar mixture CH4/ 2
CO2 in the same thermodynamics conditions that for the pure
In this equation y is mol fraction in the bulk phase and x is mol
fraction in the adsorbed phase, values are shown in Table 5. (1 for CH4
and 2 for CO2). This selectivity exhibits a non -monotonic evolution as a
function of the pressure, with a significant increase at low pressure fol-
lowed by a maximum reaching a value of 4.7 at 1.5 (MPa). This profile
can be interpreted as follows: CO2 reaches the saturation capacity
whereas there is still a slight increase in CH4 adsorption at intermediate
and high pressures. A low and medium pressures, it is clearly stated that
the CO2 molecules preferentially occupy the adsorption sites. This af-
finity between CO2 and kerogen is consistent both with the obtained
enthalpy values and with the slope of the isotherms at very low coverage
in the Henry zone.
Even if the experiment were not carried out in conditions compatible
with in-well conditions, these results confirms that shale may be a
promising candidate for both the gas recovery and the CO2 storage.
Fig. 5. Heat of Adsorption (open circle: CH4 (Pozo et al., 2017), filled diamond: CH4 this
work, filled triangle: CO2).

Fig. 6. Equimolar mixture sorption and individual components sorption in the mixture
(times: mixture, empty square: CO2-mixture, empty triangle: CH4 mixture). Fig. 7. Selectivity of CO2 over CH4 (filled diamond: Selectivity).

Table 5
Adsorption data for equimolar mixture and individual components in the equimolar mixture CH4-CO2.

Pressure nexc mixture nexc CH4 nexc CO2 y2 y1 x2 x1

(MPa) (mol/Kg) (mol/Kg) (mol/Kg) CO2 CH4 CO2 CH4

0.40 0.1374 0.0630 0.0744 0.4585 0.5415 0.5415 0.4585

0.75 0.2161 0.0604 0.1557 0.4782 0.5218 0.7206 0.2794
1.08 0.2680 0.0576 0.2104 0.4885 0.5115 0.7850 0.2150
1.39 0.3077 0.0577 0.2500 0.4936 0.5064 0.8124 0.1876
1.71 0.3322 0.0609 0.2713 0.4971 0.5029 0.8167 0.1833
2.08 0.3592 0.0714 0.2878 0.4990 0.5010 0.8013 0.1987

312
O.P. Ortiz Cancino et al. Journal of Petroleum Science and Engineering 159 (2017) 307–313

4. Conclusions Collell, J., Ungerer, P., Galliero, G., Yainnourakou, M., Montel, F., Pujol, M., 2009.
Molecular simulation of bulk organic matter in type II shales in the middle of the oil
formation window. Energy fuels. 28, 457–466.
In the present experimental study, pure and mixture adsorption Daniels, E.J., Altaner, S.P., 1990. Clay mineral authigenesis in coal and shale from the
properties of carbon dioxide-methane were investigated on graptolite Anthracite region, Pennsylvania. Am. Mineral. 75, 825–839.
black shale sample. This paper provides a new set of experimental data Gasparik, M., Bertier, P., Gensterblum, Y., Ghanizadeh, A., Krooss, B.M., Littke, R., 2014a.
Geological controls on the methane storage capacity in organic-rich shales. Int. J.
meaningful to understand of the adsorption mechanisms in shale Coal Geol. 123, 34–51.
and kerogen. Gasparik, M., Ghanizadeh, A., Bertier, P., Gensterblum, Y., Bouw, S., Krooss, B.M., 2012.
The main conclusions can be summarized as follows: High-pressure methane sorption isotherms of black shales from The Netherlands.
Energy fuels. 26, 4995–5004.
Gasparik, M., Rexer, T.F.T., Aplin, A.C., Billemont, P., De Weireld, G., Gensterblum, Y.,
(1) Adsorption behavior of CO2 and CH4 exhibit type I Langmuir. CO2 Henry, M., et al., 2014b. First international inter-laboratory comparison of high-
uptake is clearly greater than CH4. pressure CH4, CO2 and C2H6 sorption isotherms on carbonaceous shales. Int. J. Coal
Geol. 132, 131–146.
(2) Representative fits of excess sorption isotherms were obtained Gensterblum, Y., Merkel, A., Busch, A., Krooss, B.M., 2013. High-pressure CH4 and CO2
using a 3-parameter Langmuir-based excess sorption function. The sorption isotherms as a function of coal maturity and the influence of moisture. Int. J.
sorption isotherms at different temperatures can be adequately Coal Geol. 118, 45–57.
Gensterblum, Y., Van Hemert, P., Billemont, P., Busch, A., Charriere, D., Li, D., Wolf, K.-
represented assuming constant Langmuir sorption capacity (nL) H., 2009. A. European inter-laboratory comparison of high pressure CO2 sorption
and sorbed phase density (ρads) and temperature-dependent isotherms. I: activated carbon. Carbon 47, 2958–2969.
Langmuir pressure (pL). Ghoufi, A., Gaberova, L., Rouquerol, J., Vincent, D., Llewellyn, P.L., Maurin, G., 2009.
Adsorption of CO2, CH4 and their binary mixture in Faujasite NaY: a Combination of
(3) The differential heat of adsorption is nearly constant for CO2 and
molecular simulations with gravimetry-manometry and microcalorimetry
CH4. The highest values obtained for CO2 support the highest af- measurements. 2009. Microporous Mesoporous Mat. 119, 117–128.
finity between CO2 and kerogen. Guo, S., 2013. Experimental Study on isothermal adsorption of methane gas on three
(4) The investigation of the equimolar mixture CO2/CH4 supports the shale samples from upper Paleozoic strata of the Ordos Basin. J. Pet. Sci. Eng. 110,
132–138.
higher selectivity of CO2 over CH4. This result is consistent with Gutierrez-Marco, J.C., Storch, P., 1998. Graptolite biostratigraphy of the lower silurian
the values obtained for the differential heat of adsorption. (llandovery) shelf deposits of the western iberian cordillera. Spain. Geol. Mag. 135,
(5) The present works provide an original set of experimental CO2/ 71–92.
Heller, R., Zoback, M., 2014. Adsorption of methane and carbon dioxide on gas shale and
CH4 selectivity in kerogen. These data are meaningful to test and pure mineral samples. J. Unconv. Oil Gas. Resour. 8, 14–24.
discriminate the realistic model proposed for kerogen. Kang, S.M., Fathi, E., Ambrose, R.J., Akkutlu, I.Y., Sigal, R.F., 2011. Carbon dioxide
storage capacity of organic-rich shales. SPE J. 16, 842–855.
Mendhe, V.A., Mishra, S., Varma, A.K., Kamble, A.D., Bannerjee, M., Sutay, T., 2017. Gas
Acknowledgements reservoir characteristics of the lower gondwana shales in Raniganj basin of eastern
India. J. Pet. Sci. Eng. 149, 649–664.
The scientific work is included within the research activities of the Mouahid, A., Bessieres, D., Plantier, F., Pijaudier-Cabot, G., 2012a. Supercritical
adsorption of nitrogen on EcoSorb-activated carbon at temperatures up to 383 K and
Research group C-144 (UAM, Geomaterials and Geological Processes). pressures up to 2 MPa. J. Therm. Anal. Calorim. 109, 473–479.
Olga Patricia Ortiz Cancino is grateful to Universidad Industrial de Mouahid, A., Bessieres, D., Plantier, F., Pijaudier-Cabot, G., 2012b. A thermostated
Santander for the opportunity to do her Ph. D. studies at UPPA and to coupled apparatus for the simultaneous determination of adsorption isotherms and
differential Enthalpies of adsorption at high pressure and high temperature. J. Therm.
Colfuturo, Embassy of France, Colombian Ministry of National Education
Anal. Calorim. 109, 1077–1087.
and ASCUN for financial support. Pino, D., Plantier, F., Bessieres, D., 2014. Experimental determination of the adsorption
isotherms in gas mixtures under extended pressure and temperature range:
References application to the CO2–CH4 binary mixture. J. Therm. Anal. Calorim. 117,
1469–1477.
Pozo, M., Pino, D., Bessieres, D., 2017. Effect of thermal events on maturation and
Bauluz, B., Subías, I., 2010. Coexistence of pyrophyllite, I-S, R1 and NH4þ-rich illite in methane adsorption of ammonium illite-rich Silurian black shales (Checa, Spain).
Silurian black shales (Sierra de Albarracín, NE Spain): metamorphic vs. hydrothermal Appl. Clay Sci. 136, 208–218.
origin. Clay Min. 45, 383–392. Ross, D., Bustin, R.M., 2009. The importance of shale composition and pore structure
Belmabkhout, Y., De Weireld, G., Frere, M., 2004. High-pressure adsorption isotherms of upon gas storage potential of shale gas reservoirs. Mar. Pet. Geol. 26, 916–927.
N2, CH4, O2 and Ar on different carbonaceous adsorbents. J. Chem. Eng. Data 49, Sircar, S., 1999. Gibbsin surface excess for gas adsorption8 Gibbsian surface excess for gas
1379–1391. adsorption. Ind. Eng. Chem. Res. 28, 3670–3682.
Bessieres, D., Saint-Guirons, H., Daridon, J.L., Coxam, J.Y., 2000. Apparatus for Span, R., Wagner, W., 1996. 1996. A new equations of state for carbon dioxide covering
simultaneous determination of the densities and heat capacities of liquids and of the fluid region from the triple point temperature to 1100 K at pressures up to 800
liquids with dissolved gas under an extended range of pressure (0.1-100 MPa). Meas. MPa. J. Phys. Chem. Ref. Data 25 (6), 1509–1596.
Sci. Technol. 11. N69. Span, R., Wagner, W., 2003. Equations of State for Technical Applications. I.
Bessieres, D., Lafitte, T., Daridon, J.-L., Randzio, S.L., 2005. High pressure thermal Simultaneously optimized functional Forms for nonpolar and polar fluids. Int. J.
expansion of gases: measurements and calibration. Thermochim. Acta 428, 25–30. Thermophys. 24, 1–39.
Charoensuppanimit, P., Mohammad, S.A., Gasem, K.A.M., 2016. Measurements and Srodon, J., 1980. Precise identification of illite/smectite interstratifications by X-ray
modeling of gas adsorption on shales. Energy fuels. 30, 2309–2319. powder diffraction. Clays Clay Min. 28, 401–411.
Chareonsuppanimit, P., Mohammad, S.A., Robinson, R.L., Gasem, K.A.M., 2012. High- Sui, H., Yao, J., 2016. Effect of surface chemistry for CH4/CO2 adsorption in kerogen: a
pressure adsorption of gases on shales: measurements and modeling. Int. J. Coal Geol. molecular simulation study. J. Nat. Gas. Sci. Eng. 31, 738–746.
95, 34–46. Yu, S., Chen, H., Yang, S., Guo, X., Zhou, Ch, Fang, B., Zhou, F., Yang, J.K., 2014. A new
Clarkson, C.R., Bustin, R.M., 2000. Binary gas adsorption/desorption isotherms: effect of mathematical model considering adsorption and desorption process for productivity
moisture and coal composition upon carbon dioxide selectivity over methane. Int. J. prediction of volume fractured horizontal wells in shale gas reservoirs. J. Nat. Gas.
Coal Geol. 42, 241–271. Sci. Eng. 19, 228–236.
Collell, J., Galliero, G., Gouth, F., Montel, F., Pujol, M., Ungerer, P., Yiannourakou, M., Zhang, T., Ellis, G.S., Rupple, S.C., Milliken, K., Yang, R., 2012. Effect of organic matter
2014. Molecular simulation and modelisation of methane/ethane mixtures type and thermal maturity on methane adsorption in shale-gas systems. Org.
adsorption onto a microporous molecular model of kerogen under typical reservoir Geochem. 47, 120–131.
conditions. Microporous Mesoporous Mat. 197, 194–203.

313