Numerical modelling

of thermo-poro-elastic

behaviour

of the injection wellbore and its impact on CO
2

storage in coalbeds
Shailendra Mohite
1

Imperial College London, Exhibition Road,

Abstract

The CO
2
storage associated with enhanced recovery of methane from unmineable coalbeds has been an
attractive option for its potential to provide a clean energy resource while simultaneously reducing atmospheric
emissions. However, the significant volumetric strain induced by CO
2
adsorption controls the stress and cleat
permeability due to matrix swelling, consequently reducing its storage capacity in coalbeds.
Thermal stimulation of coalbed near wellbore region by fracturing of the matrix was used to achieve better
injectivity of CO
2
. Recent laboratory studies conducted on coal subjected to cyclic temperature gradient
(freezing to 248
o
K and subsequent thawing to 313
o
K) have indicated permanent transformation in elastic
properties and swelling characteristics by way of an increase in coal matrix porosity, fracture permeability and
reduction in elastic modulus, resulting in reduced swelling of coal during CO
2
injection. The influence of this
reduced matrix swelling was studied in near wellbore region on permeability through a numerical modelling
approach using COMSOL Multiphysics ® software.
The 2-D thermo-poro-elastic coupled model enabled formation and radial propagation of a thermally
stimulated zone (TSZ) using cold N
2
at different temperatures as low as 173
o
K and pressures up to 6 MPa,
under constrained conditions of confining stress, for a vertical injection well in coalbed.
A practical application of the model through a case study on cavitated well revealed a dual advantage on
permeability improvement due to macro-fractures along with reduced swelling of matrix induced by CO
2
in
TSZ. The base case total stress reduction after thermal stimulation was 2.2 MPa equivalent to permeability
improvement by 16% near wellbore region, whereas the cavitated well case study had a stress reduction of 2.4
MPa, equivalent to permeability improvement by 10%. Sensitivity analysis on elastic properties portrayed that
low rank coals could likely have significant influence on CO
2
injectivity, after thermal stimulation.

Keywords : ECBM, injection wellbore, matrix swelling, thermal stimulation, permeability, coal rank, COMSOL


Copyright 2012, Brazilian Petroleum, Gas and Biofuels Institute - IBP
This Technical Paper was prepared for presentation at the Rio Oil & Gas Expo and Conference 2012, held between September, 17-
20, 2012, in Rio de Janeiro. This Technical Paper was selected for presentation by the Technical Committee of the event according to
the information contained in the final paper submitted by the author(s). The organizers are not supposed to translate or correct the
submitted papers. The material as it is presented, does not necessarily represent Brazilian Petroleum, Gas and Biofuels Institute’
opinion, or that of its Members or Representatives. Authors consent to the publication of this Technical Paper in the Rio Oil & Gas
Expo and Conference 2012 Proceedings.

IBP1565_12

1. Introduction

The coalbed methane (CBM) reservoirs are soft rock, naturally fractured shallow formations acting as low
pressure gas reservoirs and usually saturated with water. The stress changes due to drawdown, shrinkage of the
coal matrix during methane production and swelling due to gas adsorption which can alter the permeability of
coalbeds.
The porous structure of coal is normally described using Warren and Root concept, wherein the coal
matrix blocks are considered to be rectangular parallelepipeds or cubes, and the fractures are considered to be
parallel cleats between matrix blocks. The two orthogonal cleat sets, perpendicular to the bedding are commonly
referred as face (dominant) and butt (subordinate) cleats, as in Fig. A-1 (Cui et al. 2007).
Permeability is negligible in the matrix of coal. Gas transport is Fickian in nature and described by
molecular and surface diffusion processes simultaneously taking place in the macro and micro-pores,
respectively. Gas migration within coalbeds involves three stages : desorption from internal coal surface,
diffusion through coalbed matrix and Darcy flow in the cleat network. The permeability has been recognised as
one of the most important parameter controlling gas productivity during primary and enhanced coalbed methane
(ECBM) recovery process.
Permeability changes during primary recovery of methane due to two competitive effects : (a) effective
stress increases due to pressure decrease; (b) matrix shrinkage due to methane desorption, thus increase the
permeability. The CO
2
and N
2
induced ECBM processes engender different (secondary) recovery mechanism,
wherein with the injection of N
2
, partial pressure of methane in the coal seam reduces. Accordingly, methane
desorbs from the coal surface and is driven to the production wells due to sweeping by the injected N
2
. The
injection of CO
2
leads to preferential adsorption over methane and has an extra benefit of its storage.
However, the field evidence has demonstrated that well injectivity due to CO
2
declines from initial 5 to 3
million std. ft
3
/day at the early stage and then rebounded at Allison pilot unit of San Juan basin, but the loss in
injectivity is mainly attributed to estimated two-orders of magnitude reduction in permeability (Reeves et al.
2003, Shi and Durucan 2004), whereas in Quinsam (Canada) coal, it can cause up to 6 order of magnitude
reduction (Cui et al. 2007). Significant decline in CO
2
injectivity has also been encountered in other field trials
(Van Bergen et al. 2006 and Shi et al. 2008).
While there has been detailed measurement data on pressure conditions to determine changes in
permeability, possible hydrocarbon sweep efficiency and thereby CO
2
storage volumes, there is very limited
information on evaluation of thermal effects on wellbore region of coalbeds. The cold N
2
injected induces
temperature effects via thermal fracturing that affect the flow / injectivity, which has been the key interface for
modelling in this study. The well injectivity is proposed to be maintained and improved by adopting a CO
2
-
alternating-N
2
cyclic injection strategy. Hence, this paper on thermo-poro-elastic (TPE) numerical modelling
work is a distinctive effort from basics to test the potential and thus enhance permeability of coalbeds near
wellbore region, as illustrated in process flow of the model, in Fig. A-2.
The main objectives of this paper are : (a) assess impact of thermal stimulation of coal matrix on CO
2

injectivity near wellbore for ECBM process and its storage; (b) develop a coupled thermo-poro-elastic model to
study elastic properties and swelling characteristics of coal.

2 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds


2. Background

Coal has the capacity to adsorb more CO
2
than methane

or N
2
in an approximate ratio of 4:2:1, until the gases in
both the sorbed and gaseous states are each in equilibrium, as per a combination of laboratory studies and field
demonstrations sponsored by US Dept. of Energy (DOE) (Reeves 2001). The commercial CBM production
has been carried out by reservoir pressure depletion with a primary recovery possible around 50% while mature
plays in USA have experienced 60% to 80% recovery (Shi and Durucan 2005).
During the early 90’s, the ECBM secondary recovery process was by using CO
2
and later on along with
N
2
as efficient means for recovery without excessively lowering the reservoir pressure. Whereas, an inert gas
like N
2
and hence low affinity for coal, when injected it decreases the partial pressure of gaseous methane in the
matrix. As a result, methane gets desorbed into the cleats to achieve partial pressure equilibrium. The elevated
pressures and lowering of temperature affected by N
2
injection causes expansion of the coal fractures on the
preferential permeability trends as observed in the Tiffany unit (Reeves et al. 2002; Erickson 2002).
The gas storage by sorption in coal beds represents approx. 95-98% of the gas in the coal seams. Methane
and CO
2
adsorption is usually described by a Langmuir-type isotherm, indicating that the adsorption is
dominated by micro-pore-filling (< 2 nano m) process. N
2
injection is similar to inert gas stripping since it is
feebly adsorbing than methane, due to reduction of partial pressure of methane and promoting its desorption
without lowering the total reservoir pressure. Since there is a significant change in volume of the coal matrix
due to desorption and adsorption of gas, it leads to wide variations in the permeability and in turn controls
production of CBM. Typically, coal can adsorb about half as much N
2
at a given pressure than methane, as
depicted in the Fig. A-3 (Shi et al. 2008).
The relationship between gas concentration and pressure is a non-linear function defined by the Langmuir
equation (Eq. 1) : (Langmuir 1916)


) 031 . 0 (
p P
p V
C
B
L
L
m
µ
+
=

……...........………...............………………………………………………… (1)
where, C
m
: matrix gas concentration (scf/ft
3
), V
L
: dry and ash-free Langmuir volume constant (scf/ton),
P
L
: Langmuir pressure constant (psia), p : pressure in fracture system (psia) and

is the bulk density (kg/m
3
).

3. Pore Pressure, Permeability and Stress

During the pressure-depletion primary recovery, in situ stresses increase due to pore pressure decrease, where
cleats get compressed resulting in the decrease of cleat permeability and porosity. On the other hand, with
desorption of methane the coal matrix would shrink causing cleat apertures to increase thus permeability and
cleat porosity would increase. However, the shrinkage of coal matrix would also cause in situ stresses to
decrease.
Therefore, the increase or decrease of permeability and porosity during production depends on the pore
pressure and induced shrinkage / swelling due to sorption. A part of above shrinkage strains convert to the
relaxation or decrease of horizontal in-situ stresses that may cancel out the increase due to pore pressure decline.
During secondary ECBM recovery when CO
2
injection begins, the gas flows into the cleats first, and then
diffuses into the matrix due to the concentration gradient. The gas flowing through the cleats is considered to be
gas seepage controlled by the permeability of fracture in the coal seam but the gas diffusion in the matrix, which
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 3


is a thermodynamic process, is controlled by both the permeability and the adsorption (Harpalani and Chen
1995).
Effective stress is the difference between the horizontal in-situ stress and pore pressure (o - p) or is the
total stress acting normal to the cleats minus the fluid pressure (Mazumder et al. 2006). Coalbed permeability
varies exponentially with changes in effective stress acting across the cleats through the cleat-volume
compressibility. The relation between the change of effective stresses and permeability is monotonic and was
presented by McKee et al. (1988) and Seidle et al. (1992), wherein absolute permeability varies exponentially
with changes in horizontal stress as in Eq. 2 (Durucan and Edwards 1986, McKee et al. 1988; Palmer and
Mansoori 1998) :


) ( C 3
0
o f
e k k
o o÷ ÷
= ............…………….....................…………………………………………………… (2)

where; k
0
: initial permeability, C
f
: cleat volume compressibility with respect to effective horizontal stress
normal to the cleats and change of effective stress ) (
0
o o ÷ , in McKee et al’s (1988) model is the change of
mean effective stresses while in Seidle’s et al’s (1992) model is the change of effective horizontal stress.
The shrinkage/swelling is proportional to the volume of gas desorbed/ adsorbed, rather than change in
sorption pressure (Harpalani and Chen 1995; Seidle and Huitt 1995). The changes in horizontal stress in a
uniaxial reservoir due to sorption resulting from pore pressure are represented by (Shi and Durucan 2005) :

¿
=
÷
÷
+ ÷
÷
÷ = ÷
n
1 j
0 j j sj 0 0
) V V (
) 1 ( 3
) p p (
) 1 (
o
v
E
v
v
o o .......................................................…………… (3)
where; p and p
0
: change in fluid pressure from initial, E : Young’s modulus, v : Poisson ratio, o
sj
:
shrinkage / swelling coefficient, V
j
and V
jo
are specific adsorbed gas volumes for component ‘j’ at current and
initial conditions. The two terms on the right side of above equation are referred to as ‘cleat compression’ and
‘matrix shrinkage / swelling terms’, respectively, which determine magnitude of ) (
0
o o ÷ .

4. Problem definition and influence of thermally induced stress

Thermal stimulation by injecting cold fluids in typical rocks has shown that cooled regions contract which
substantially reduce the hoop (circumferential) stress component around wellbore causing hydraulic fractures at
considerably lower pressures (Perkins and Gonzales 1985). The thermal hoop stress (o
uu
) due to cooling are
negative and relaxed from its original value of horizontal stress (o
h
) whereas by contrast and equilibrium, the
warmer regions have higher hoop stress In between, a transition zone prevails with very steep stress gradients
(Charles et al. 1996).
There have been some references in literature for relationship between thermal contraction and matrix
shrinkage associated with gas desorption in coalbeds, besides stress-strain relationships for a thermo-elastic
porous media. Hence, it was believed that thermal effects or freezing may help in improving coalbed
permeability. These references are briefly as under :
1. The elastic properties of porous rock depend on the porosity and crack density (Walsh, 1965a and b;
Reiss 1980). Matrix fracturing can induce micro-cracks thereby reducing elastic modulus of the rock.
2. There are some field observations for use of cryogenic N
2
as a hydraulic fracturing fluid for
underperforming CBM wells. The low thermal conductivity of coal combined with significant
shrinkage due to cryogenic N
2
caused a high stress resulting fracture of the coal matrix. It was
concluded that liquid N
2
can be injected into CBM reservoir for sustaining thermal effects on shrinkage
(McDaniel et al. 1997)
3. Replacing the thermal expansion terms by an analogous shrinkage terms yields the corresponding
relationships for an isothermal gas-desorbing porous media (Shi and Durucan 2004; Palmer 2009). In a
non-isothermal body, by reducing its temperature the fabric fractures, leading to an increase in its
porosity. An enhanced permeability zone is expected near wellbore region with cold gas injection
influenced hydraulic fracturing. This is analogous to matrix shrinkage where cleat porosity increases
on gas desorption on drawdown. This differential contraction with cold temperature can be coupled
with pore fluid pressure and velocities associated with fluid flow for overall solution at the stress, using
a complete set of Navier-type equations, heat diffusion equation and pore pressure diffusion equation in
a 2-D form. Navier’s equation is a system of expressions that relate stresses (o), strains (c) and
displacements (u) in solids. The pressure, temperature and the stress field due to non isothermal
injection of fluid in a coalbed reservoir can be described by a coupled thermo-poro-elasitc model, as in
Eq. 4. The swelling effect can be conveniently expressed similar to the temperature load term, as per
following Navier’s type equation, (Abousleiman and Ekbote 2005) :

4 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds


i , i , B ij , j jj , i
T p u
2 1
G
Gu | o
v
v
+ =
÷
+ ............…….................................……………………...……(4)

where;
) 1 ( 2
E
G
u ÷
=
: Lamé elastic (shear modulus) constant, u : solid displacement, v : Poisson
ratio, o
B
: Biot’s effective stress coefficient, p : pressure, as a source term, | : thermal coefficient, T :
temperature, i and j are components. The left hand side indicates elastic deformation influenced by
equivalent fluid pressure and sorption induced body forces. The right hand side represents influence of
pore pressure and thermal effects, respectively. The above theory on shrinkage of the matrix by thermal
stimulation associated with injection of cold fluids has been useful in formulating the governing
equation of the numerical model, based on the modified Eq. 3 incorporating a third term on the right
side, pertaining to the temperature load term as under (source : Imperial College London, PhD transfer
report of Syed, A. 2010) :

) T T (
) 1 ( 3
E
) V V (
) 1 ( 3
) p p (
) 1 (
0 T
n
1 j
0 j j sj 0 0
÷
÷
+ ÷
÷
+ ÷
÷
÷ = ÷
¿
=
o
v
o
v
E
v
v
o o ..................……(5)
where; o
T
: thermal expansion coefficient, T and T
0
are current and initial temperature, respectively.

5. 2-D Model assumptions, parameters and physics sequences

The model with major assumptions as in Table-1, contains a set of application modes that define the
mathematical modelling equations and can be fitted to the conceptual model of physics, with boundary and
initial conditions. The application modes of the FEM software tool (COMSOL Multiphysics ® 2008, 2010) can
be used for linking the variables and coupling the physics represented anywhere in the model, as in Table 2.
The sign convention in this study used has been positive for compressive stress is positive and negative for
tensile stress.

Table 1 : Main assumptions for thermo-poro-elastic model


Table 2 : Governing equations for thermo-poro-elastic model

COMSOL®
application modes
Governing laws and equations of phenomena
Fluid Flow
- Darcy’s equation
s f
Q ) gz p (
k
. =
(
¸
(

¸

+ V ÷ V µ
q

…....................................................................……(6)
Heat Transfer
- Fourier’s equation
| | T . C T kp .
t
T
C
pL p
V ÷ = V ÷ V +
c
c
u

……………………………………………..…(7)
Solid Mechanics
- Biot’s poro-elastic theory and Exponential law, as per Eq. 2

H g ) u .(
) 2 1 )( 1 ( 2
E
u
) 1 ( 2
f B
2
V = V V
÷ +
+ V
÷
µ o
v v v
E
…........................................................…(8)

(
¸
(

¸

+
÷ ÷ ÷ =
p P
p V
K ) T E 3 ( p ) 8 . Eq ( LHS
L
L
s T
o A o

…...................................................…(9)

Brief details on governing equation for total stress (Eq. 9) and model validation is provided in Appendix-
B and Appendix-C, respectively.

Main assumptions Parameters not being considered
- vertical well, isotropic conditions and dry ideal gas - effects of gravity, capillary pressure and skin
- confined and homogeneous reservoir, viscous gas flow
obeys Darcy’s law.
- negligible expansion of injected fluids or heat losses
inside wellbore
- µ, k, |, Cf, µ : are all constants - gas leakage or decay rates
- sorption is manifested in geomechanical stresses - coal grain deformations
- linear elasticity, Biot’s constant; o
B
= 1
- dual porosity
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 5


Solid deformation equation :
p=p
0
, T=T
0 ,
o=o
0
Injection well
R
w
=0.0762 m
(6” diameter)
p=p
w
, T=T
inj
v=0
u=0
p=p
0
, T=T
0 ,
o=o
0
6. Model base case for CO
2
induced swelling
6.1 Model definition

Most of the coalbeds are a few meters thick and are oriented horizontally and therefore the macro-porous
fracture networks which are invariably perpendicular to coalbeds, are near vertical (refer Fig. 1). The
deformation induced by changes in horizontal stresses are expected to be predominant, where pressure,
temperature and stress change transiently which are coupled using stress equations, as per Table-2.
The model considered a single-component gas flow phenomena designed as a set of coupled PDE
application mode under the Earth Science module are coupled with heat transfer. The relevant boundary
conditions have been imposed for coupled application modes as per requirement of the model features, besides
material properties, process conditions and constants.
The 2-D model consisted isothermal fluid flow that was weakly coupled with geomechanical stress
considering simplifying assumptions on single component and total plane stress under linear elastic conditions,
was developed using COMSOL Multiphysics ®. In order to store CO
2
it is required to be injected at higher
pressure than very low initial reservoir pressure considered with an assumption that the methane is depleted.
Therefore, it is expected that effective stress would increase due to sorption induced swelling. Consequently,
the current work investigates the changes in stress due to sorption induced swelling during injection of CO
2
in
the coalbed reservoir near wellbore due to greater pore pressure compared to zones further away, at different
injection pressures.
6.2 Model description and input parameters

A square field of 1000 m sides and an injection well of radius, R
w
= 0.0762 m (6” diameter) located at the centre
(0,0) was used as shown in Fig. 4.
The initial and boundary conditions for pressure controlled conditions are : u=v=0 at oO, p=p
0
, T=T
0
, and
o=o
0
at far field stresses 15 MPa, k=k
0
. The simulations have been carried out for predicting the coalbed
stress behaviour to CO
2
injection at 3 MPa, 6 MPa and 10 MPa pressure, with initial reservoir pressure set at
1000 Pa and initial temperature at 313
o
K (same as reservoir), assuming it as a field depleted of methane after
primary recovery. CO
2
injection was carried out for a period of 1 day to examine the impact of pressure
distribution and sorption on body load as a compressive stresses. A follower injection pressure at wellbore was
maintained at the wellbore boundary as a body load, in Solid Mechanics mode.








Table 3 : Main input parameters for
CO2 base case model





The main parameters are given in Table 3 and other input data are provided in Appendix-D.
6.3 Simulation Results and Discussion

Using the solutions at the end of stipulated time scale between Fluid Flow and Heat Transfer transient analysis
into the Solid Mechanics mode, overall stress calculations were available, based on a set of constitutive
governing PDE equations and boundary conditions.
The data was extracted along the X-axis (u=0) from wellbore (0.0762, 0) to the edge of field (500, 0)
pertaining to initial state stress (o
0
) and stress tensors in x and y directions for both; with and without sorption
Main input parameters units Value
Cleat Permeability, k m
2
5e-15
Cleat Porosity, | -- 0.1
Swelling Coeff., os Kg/m
3
0.5
Langmuir pressure, PL M Pa 0.77
Langmuir volume, VL m
3
/kg 0.022
Fig. 4 : Base case model features
6 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
k
/
k
0
Radial distance/Wellbore radius ratio, r/R
w
Permeability near wellbore region : CO
2
injection for 1 day at
different pressures
X-dir, 3 MPa Y-dir,3 MPa
X-dir, 6 MPa Y-dir, 6 MPa
X-dir, 10 MPa Y-dir, 10 MPa
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
R
a
d
i
a
l
a
n
d
T
a
n
g
e
n
t
i
a
l
S
t
r
e
s
s
e
s
,


M
P
a
Radial distance / Wellbore radius ratio, r/R
w
Influence on Total Stress due to CO
2
injection for 1 day
at different pressures
X-dir, base case Y-dir, base case
X-dir, 3 M Pa Y-dir, 3 M Pa
X-dir, 6 M Pa Y-dir, 6 M Pa
X-dir, 10 M Pa Y-dir, 10 M Pa
loads (a typical poro-elastic condition) for further analysis. The influence of various pressures on the
compressive stress distribution close to wellbore due to CO
2
injection has been compared with base case initial
state stress as depicted in Fig. 5.
















There was a substantial increase in total stress calculated as per Eq. 9, from 16 MPa to 28 MPa at a radial
distance 4 times the wellbore radius. This has a direct impact on lowering of permeability near the wellbore
region, as per Eq. 2, where cleat volume compressibility (C
f
) at 2.9E-8 Pa
-1
was used for sub-bituminous coal of
Powder River basin (Ross et al. 2009).
With increase in CO
2
injection pressure from 3 MPa to 6 MPa to 10 MPa, the sorption of CO
2
results in the
swelling of coal matrix which increases effective stress on the cleats, thereby reducing permeability. The
reduction of permeability in near the injection wellbore results in the loss of injectivity of CO
2
in coalbed.
The change in permeability (k/k
0
) based on the kinetics of gas adsorption as per Eq. 2 was available and
indicates the dominance of matrix induced phenomena over mechanical influences close to the wellbore region
as depicted in Fig. 6, wherein higher injection pressures lead to lower permeability change, using the X and Y
direction stress tensors.
7. Thermal stimulation of wellbore region using cold N
2

7.1 Wellbore Stimulation options

The permeability enhancement is the objective of various stimulation techniques. The options available for
creating stimulated zones around injection well are limited as compared to production well, mainly because old
and abandoned wells are put into service for injection purposes during ECBM recovery stage. While this paper
deals with the impact due to effective stress on cleat permeability and altered elastic properties and swelling
characteristics near well bore due to thermal stimulation as compared to the other options which are mainly
mechanical operations used during primary recovery of CBM to accelerate the desorption process and enhance
permeability, like hydraulic fracturing, horizontal wells (Durucan and Shi 2009) and cavity completions. The
thermal stimulation option proposed in this paper would provide an advantage of fracturing matrix as well.
7.2 Model definition and criteria for selecting thermally stimulated zone

A thermo-poro-elastic model was developed for injection of cold N
2
to examine the thermally induced effects on
stress distribution based on temperature profile and thereby identify the radial extent of propagation of thermally
stimulated zone (TSZ) and propose a strategy for injection cycles. A pressure controlled model was chosen of
6 MPa for faster (than mass) and deeper penetration of cold N
2
into the coalbed and intended widening of the
cleats, assuming initial reservoir pressure at 1000 Pa. Injection of cold N
2
is a preferred choice over cold CO
2
to
avoid its high matrix swelling characteristics which would otherwise thwart the advantage envisaged from
thermal stimulation of matrix.
As the low temperature gas is injected into a warm (40
o
C or 313
o
K) coalbed, the area close to the
wellbore will experience negative thermal gradient for a specified period of time and hence is more susceptible
for thermal fracturing due to tensile load. Tensile stress failure can occur if one of the effective stresses becomes
negative. A difference between stress tensors (Ao) in x and y direction for N
2
injection vis-a-vis the initial state
stress distribution can be used to identify the failure zone based on tensile stress developed. However, the data
extracted from model was mainly compressive stress which was dominant over the thermally influenced tensile
stress, hence it was not used as criteria for identifying the extent of TSZ. Therefore, the ‘X’ co-ordinate of
radial distance where there is a rise from N
2
injection temperature (248
o
K) to sub-zero (273
o
K), has been used
as a criteria for selecting the extent of TSZ. During laboratory experiments on elastic properties of high volatile
Fig. 5 : Total stress induced by CO2 injection pressure
and sorption loads
Fig. 6 : Influence on permeability of CO2 injection at
various pressures
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 7


240.0
250.0
260.0
270.0
280.0
290.0
300.0
310.0
320.0
1.0 5.0 9.0 13.0 17.0 21.0 25.0 29.0 33.0
T
e
m
p
e
r
a
t
u
r
e
,
d
e
g
K
Radial distance / Wellbore radius, r/R
w
N
2
injection @ 248 K for 1 day to create thermally stimulated zones
in 3 cycles
1st cycle 2nd cycle 3rd cycle
240.0
250.0
260.0
270.0
280.0
290.0
300.0
310.0
320.0
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0
T
e
m
p
e
r
a
t
u
r
e
,
d
e
g
K
Radial distance / Wellbore radius, r/R
w
,
N
2
injection @ 248 K : Temperature profile over various periods
1 hr 3 hrs 6 hrs
12 hrs 1 day 3 days
bituminous coal, it was observed that elastic modulus reduced for coal subjected to sub-zero temperatures,
attributable to thermal fracturing of coal matrix.
Although the temperature profile propagates with time but due to low thermal conductivity of coal, the
duration and influence would be much lesser on coalbed subjected to low temperature further along the radial
distance and hence may not result in intended matrix fracturing. Therefore, coal was cooled periodically for 1
day for maximum matrix to fracture in a region very close to TSZ, but due to its elastic properties, there may not
be any change in the stress level further away from the wellbore. The modified properties of the matrix were
permanent in nature where re-injection strategy in additional cycles of 1 day each was adopted whenever the
sub-zero temperature bounces back to above 273
o
K.
7.3 Model description and input parameters

A square field of 1000 m sides with injection well of radius of R
w
= 0.0762 m (6” diameter) located at the
centre was constructed, with all other parameters similar to CO
2
injection case. The geomechanical stress was
calculated by a weakly coupled static Solid Mechanics mode based on inputs of pressure (p) from Darcy’s Fluid
Flow combined with thermal (dominant) and sorption loads. The initial and boundary conditions for pressure
controlled conditions are : u=v=0 at oO, p=p
0
, T=T
0
, o=o
0
at 15 MPa, k=k
0
.
The TSZ was identified sequentially and added suitably in the geometry of the subsequent injection cycles.
The elastic and transport properties of the TSZ were modified and its radius was progressively increased based
on propagation of sub-zero temperature for each cycle. The entire model had 11,154 elements in total consisting
fine mesh of 0.0762 m at the injection wellbore boundary and domain close to the wellbore (TSZ) and the
number of degrees of freedom was 2837. The main parameters used were as in Table 4.

Table 4 : Main input parameters for N2 injection model, including modified properties after thermal stimulation









7.4 Simulation Results and Discussion

The N
2
injection process for 1 day each for 3 cycles was chosen to mimic the weathering process as observed in
laboratory experiments and that can be implemented in a field scenario. Though the coalbed can be cooled for
three consecutive days so that temperature gradient can be propagated comparatively for a longer radial
distance, whereas if it is cyclic cooling, the stress hysteresis can accelerate weathering of coal matrix. The
comparative plots of temperature profile for various time periods for N
2
injection temperature at 248
o
K case,
indicates that for 1 day x 3 cycles cooling the sub-zero temperature can be achieved for ~ 14 times the wellbore
radius (refer Fig. 7) while by injection for 3 days it is much beyond at ~ 23 times r/R
w
ratio (refer Fig. 8).
















A comparison of the temperature profiles for N
2
injection at 248
o
K for 3 cycles of 1 day each, depicts the
propagation of TSZ in the radial direction in steps of 0.69 m, 0.92 m and 1.08 m radius, respectively, (refer
Table 5). At lower injection temperatures like 223
o
K and 173
o
K, there is an increase in density and reduction
in viscosity of injected fluid besides higher thermal shrinkage, hence improvements in elastic and flow
Main input parameters Units Value
Cleat Permeability, k m
2
5e-15
Cleat Porosity, | -- 0.1
Coeff. of thermal exp., oT 1/
o
K 2.0e-5
Langmuir pressure, PL MPa 6.8
Langmuir volume, VL m
3
/kg 0.009
Elastic and Flow properties Units Initial TSZ
Young’s modulus, E GPa 2.14 1.80
Poisson ratio, v -- 0.30 0.27
Swelling coefficient, oS Kg/m
3
0.50 0.40
Permeability, k mD 5 50
Fig. 7 : Temperature profiles for N2 injection pressure at
6 MPa and 248
o
K , in 1 day x 3 cycles
Fig. 8 : Hourly temperature profiles for N2 injection at 6
MPa and 248
o
K
8 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

150.0
170.0
190.0
210.0
230.0
250.0
270.0
290.0
310.0
1.0 6.0 11.0 16.0 21.0 26.0 31.0 36.0 41.0 46.0
T
e
m
p
e
r
a
t
u
r
e
,
d
e
g
K
Radial distance / Wellbore radius, r/R
w
N
2
injection @ 173 K for 1 day to create thermally stimulated zones
(3 & 6 MPa)
1st cycle, 3 MPa
2nd cycle, 3 MPa
3rd cycle, 3 MPa
1st cycle, 6 MPa
2nd cycle, 6 MPa
3rd cycle, 6 MPa
12.0
12.5
13.0
13.5
14.0
14.5
15.0
15.5
16.0
16.5
17.0
1.0 5.0 9.0 13.0 17.0
R
a
d
i
a
l
a
n
d
T
a
n
g
e
n
t
i
a
l
S
t
r
e
s
s
e
s
,

M
P
a
Ratio of Radial distance / Wellbore radius, r/R
w
Influence on total Stress due to N
2
injection for 1 day at 248
o
K,
(6 MPa)
X dir, base case
Y-dir, base case
X-dir, poro-elastic case
Y-dir, poro-elastic case
X-dir, 1st cycle
Y-dir, 1st cycle
X-dir, 2nd cycle
Y-dir, 2nd cycle
X-dir, 3rd cycle
Y-dir, 3rd cycle
160
180
200
220
240
260
280
300
320
1.0 5.0 9.0 13.0 17.0 21.0 25.0 29.0 33.0 37.0 41.0 45.0 49.0
T
e
m
p
e
r
a
t
u
r
e
,

d
e
g


K
Radial distance / Wellbore radius, r/R
w
N
2
injection at various temperatures for 1 day to create thermally
stimulated zones (at 6 MPa)
1st cycle 248 K
2nd cycle 248 K
3rd cycle 248 K
1st cycle, 223 K
2nd cycle, 223 K
3rd cycle, 223 K
1st cycle, 173 K
2nd cycle, 173 K
3rd cycle, 173 K
properties are expected due to thermally induced stimulation or matrix fracturing due to tensile stress. The
effect of lowering pressure from 6 MPa to 3 MPa on propagation of temperature was also analysed for a trade-
off with corresponding TSZs created, as illustrated in Fig. 9 and Fig. 10.
The model seamlessly calculates the changes stress basis elastic and flow properties from one zone to
another consisting different input domain properties. The numerical simulation has inherent continuity for the
shape functions of a composite cylinder like radial stress, whereas the tangential stress shows discontinuity; that
is free to deform in Y-direction (Indenbaum 1974). The discontinuity at r/R
w
= 7 and 10 where changes in Y-
direction stress were highest, have been chosen for comparing stress before and after thermal stimulation as in
Fig. 11-a and Fig. 11-b.

Table 5 : Penetration of TSZ over 3 cycles for various temperatures of cold N2 injection
N2 injection Temperature (T) 1
st
cycle 2
nd
cycle 3
rd
cycle
248
o
K r : radial distance, m 0.69 0.92 1.08
r/Rw 9 12 14
223
o
K r : radial distance, m 1.27 1.71 1.91
r/Rw 17 22 25
173
o
K r : radial distance, m 2.32 2.51 2.89
r/Rw 30 33 38

































The highlights of cold N
2
injection are as under :-
1. The benefit has been for reducing the Y-direction component of total stress from 15.7 to 14.4=1.3 MPa
(8%) under the influence of sub-zero temperature that would aid in creation of TSZ (refer Fig. 11-b).
2. At lower N
2
injection temperature of 173
o
K, there has been appreciable gain in propagating the TSZ to a
longer radial distance (approx. 3 times) than at higher temperature of 248
o
K.
3. Lowering N
2
injection pressure from 6 to 3 MPa propagated TSZ over a shorter radial distance by 45%.
8. CO
2
injection after thermal stimulation
8.1 Model description and input parameters
In order to ascertain the impact of TSZ on CO
2
sorption induced stress, the same base case model as described
in section 6.2 was used, maintaining similar initial and boundary conditions. A series of pressure controlled
simulations were run for 3, 6 and 10 MPa with CO
2
temperature of 313
o
K. The modified elastic and transport
properties (refer Table 4) of TSZ were incorporated as a new domain of 1.08 m radius (r/R
w
~ 14), which was
the maximum achieved for the 3
rd
cycle during N
2
injection at 248
o
K (refer Table 5).
Fig. 9 : N2 injection cycles at 6 MPa from 248
o
K to
223
o
K to 173
o
K
Fig. 10 : TSZ propagated at N2 injection pressure from
3 MPa to 6 MPa and corresponding injection over 3 cycles
Fig. 11-a : Schematic of hoop stress around
an injection wellbore (Charles et al. 1996)
Fig. 11-b : Reduction in total stress components, from 1
st
injection
cycle of N2
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 9


0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
1.0 5.0 9.0 13.0 17.0
k
/
k
0
Radial distance/Wellbore radius ratio, r/R
w
Permeability in near wellbore region, before and after thermal stimulaiton
CO
2
injection for 1 day for 6 MPa pressure (C
f
: 2.9 e-8 Pa
-1
)
X-dir, Before stimulation, 6 MPa X-dir, After stimulation, 6 MPa
Y-dir, Before stimulation, 6 MPa Y-dir, After stimulation, 6 MPa
22.0
23.0
24.0
25.0
26.0
27.0
28.0
29.0
30.0
31.0
32.0
1.0 5.0 9.0 13.0 17.0 21.0 25.0
R
a
d
i
a
l
a
n
d
T
a
n
g
e
n
t
i
a
l
S
t
r
e
s
s
e
s
,

M
P
a
Radial distance / Wellbore radius ratio, r/R
w
Influence on total Stress due to CO
2
injection for 1 day @ 6 MPa
X-dir, no stimulation Y-dir, no stimulation
X-dir, after stimulation Y-dir, after stimulation
8.2 Simulation results and discussion

The stress tensors for different injection pressure at r/R
w
= 4, indicated similar reductions, as in Table 6. The
enlarged plot near wellbore region in Fig. 12 (injection pressure of 6 MPa), indicated reduction in effective
stress from 27.2 MPa to 25.0 MPa (2.2 MPa), at r/R
w
~ 14, where the influence of TSZ ends. A comparison of
the permeability change (k/k
0
) before and after thermal stimulation at typical 6 MPa injection pressure, indicated
about 16% increase in permeability ratio from 0.48 to 0.56, as in Fig. 13. The Y-direction component of related
stress change was used for analysis of radial permeability change, as it defines flow in the direction of wellbore.

Table 6 : Reduction in stress after thermal stimulation for various pressures of CO2 injection









9. Case study of thermal stimulation with a cavitated well
9.1 Model description and input parameters
The model has been extended for practical application of thermal stimulation along with a cavitated well. It was
expected to provide a dual advantage on enhanced permeability from its shear-failed disturbed zone of large
radius with elastic properties and improved cleat permeability (assumed instead of plastic properties) due to
cleat fractures along with micro-structural porosity improvement after thermal stimulation.
Typical wellbore dimensions, elastic and transport properties of a cavitated well at San Juan basin have
been assumed for this case study. As compared to base case, the wellbore radius was increased to 0.7 m and the
model incorporated two zones with modified properties : TSZ of smaller radius based on N
2
injection cycles and
outer disturbed zone due to a cavitated well (R
c
=20 m). The injection pressure of N
2
and CO
2
were maintained
at 6 MPa each. The propagation of sub-zero temperature was gauged for injection of N
2
at 248
o
K, 223
o
K and
173
o
K. However, in absence of laboratory experiments for cold N
2
influence on coals from disturbed zones, a
range of modified properties was assumed as in Table 7. The model features are illustrated in Fig. 14.

Table 7 : Input for cavitated well case











Fig. 14 : Model features for case study with cavitated well

CO2 Before stimulation Before stimulation Reduction in stress tensor
Injection Stress tensor (MPa) Stress tensor (MPa) Stress tensor (MPa)
Pressure, (p), MPa X-direction Y-direction X-direction Y-direction X-direction Y-direction
3 M Pa 25.65 28.54 23.66 26.31 1.98 2.23
6 M Pa 25.82 28.36 23.84 26.13 1.98 2.23
10 M Pa 26.06 28.13 24.08 25.90 1.98 2.23
Fig. 12 : Influence on total stress due to CO2 injection
after thermal stimulation
Fig. 13 : Cumulative Permeability changes in near
wellbore region
Main input
parameters
Units Cavitated
well
Thermal
stimulation
Well radius, Rw
m 0.70 0.70
Young’s modulus,
E
M Pa 1500 1350-1500
Poisson ratio, v
-- 0.21 0.21-0.30
Permeability, k
mD 50 50-80
Swelling
coefficient, os
Kg/m
3
0.50 0.35-0.50
Coeff. of thermal
expansion, oT
1/
o
K 2e-5 2e-5
10 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

240
250
260
270
280
290
300
310
320
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
T
e
m
p
e
r
a
t
u
r
e
,

d
e
g

K
Radial distance / Wellbore radius ratio, r/R
w
N
2
injection @ 248 K for 1 day and 2 days with cavitated well in 3
cycles (6 MPa)
1st cycle 2nd cycle (low k)
2nd cycle (24 hrs) 2nd cycle (48 hrs)
3rd cycle (low k) 3rd cycle (24 hrs)
3rd cycle (48 hrs)
0
5
10
15
20
25
30
35
40
45
50
55
1.0 5.0 9.0 13.0 17.0 21.0 25.0 29.0 33.0
R
a
d
i
a
l

a
n
d

T
a
n
g
e
n
t
i
a
l

S
t
r
e
s
s
,



M
P
a
Radial distance / Wellbore radius ratio, r/R
w
Influence on Stress due to CO
2
injection for 1 day @ 6 MPa
(with and without cavitated well)
X-dir, base case, with cav.well Y-dir, base case, with cav well
X-dir, with sorption, w/o cav. well Y-dir, with sorption, w/o cav. well
X-dir, with sorption, with cav. well Y-dir, with sorption, with cav. well
Base case
Before stimulation, w/o cavitated well
Cavitated well
9.2 Simulation results and discussion

The radius of wellbore being large at 0.7 m (27.5”diameter ), besides thermal conductivity of coalbed being low,
injection of N
2
that was maintained at maximum 6 MPa pressure did not indicate major increases in radius of
TSZ, over the 1 day injection cycles. Like in the base case, cold N
2
injection was modelled for temperatures of
248
o
K, 223
o
K and 173
o
K. The propagation of the TSZ up to sub-zero temperature was as per Table 8.

Table 8 : Penetration of TSZ over 3 cycles for various temperatures of cold N2 injection with cavitated well

Temperature 1
st
cycle 2
nd
cycle 3
rd
cycle
248
o
K r : radial dist., m 2.17 2.59 2.87
r/Rw 3.1 3.7 4.1
223
o
K r : radial dist., m 3.15 3.78 3.98
r/Rw 4.5 5.4 5.7
173
o
K r : radial dist., m 4.91 5.53 6.65
r/Rw 7.0 7.9 9.5

A sensitivity of permeability between 50 mD to 80 mD did not have significant effect on sub-zero
temperature propagation while longer duration of injection cycles of 2 days, resulted in an increase of r/R
w
by
nearly 46% (from 3.7 to 5.4 for 2
nd
cycle and from 4.1 to 5.7 for 3
rd
cycle from 248
o
K to 223
o
K), as in Fig. 15.
A maximum TSZ of 4.1 m radius was chosen for simulating CO
2
injection to evaluate the influence by
adopting 2 zones with modified properties. The elastic and transport properties considered for remaining zone
of coalbed were : E = 2.0 GPa, v = 0.30, o
S
= 0.5 kg/m
3
and cleat permeability, k = 5 mD. This case study
indicated a reduced total stress level by 5 MPa due to cavitated well, than without it, as in Fig. 16.
There were 2 discontinuities on the Y-direction stress tensor due to 2 zones incorporated with modified
properties. The modified properties as per Table 7, were subjected to sensitivity analysis. A best combination of
elastic properties was chosen, basis maximum reduction in total stress level with thermal stimulation, after TSZ
at r/R
w
=6.6, with CO
2
injection at 6 MPa for 1 day.
















The modified elastic properties of case-6 above were used for simulations at various CO
2
injection
pressures of 3 MPa, 6 MPa and 10 MPa. The best case-6 values of stress tensors for different radial distances for
CO
2
injection at 6 MPa, indicated a maximum reduction of 3.01 MPa at the wellbore, as in Table 9.

Table 9 : Reduction in stress tensors for CO2 injection after thermal stimulation


Fig. 15 : Temperature profiles for N2 injection at 248
o
K, in
1 day and 2 days x 3 cycles, for cavitated well
Fig. 16 : CO2 induced stress reduced from 30 MPa to
25 MPa, with a cavitated well, before stimulation
Ratio of radial Before stimulation Before stimulation Reduction in stress tensor
distance/wellbore Stress tensor (MPa) Stress tensor (MPa) Stress tensor (MPa)
radius, r/Rw X-direction Y-direction X-direction Y-direction X-direction Y-direction
1.0 (wellbore) 5.89 43.29 5.88 40.27 7.32 3.01
4.4 (end of TSZ) 23.78 25.71 22.32 22.79 1.46 2.92
6.6 (after TSZ influence) 24.31 25.20 22.41 22.78 1.89 2.42
28.6 (disturbed zone) 24.40 34.23 22.26 32.22 2.14 2.01
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 11


0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
1.0 2.0 3.0 4.0 5.0
R
a
d
i
a
l

a
n
d

T
a
n
g
e
n
t
i
a
l

S
t
r
e
s
s
e
s
,



M
P
a
Radial distance / Wellbore radius ratio, r/R
w
Influence on Total Stress due to CO
2
injection for 1 day @ different pressures :
before and after stimulation
X-dir, 3 M Pa, no TSZ Y-dir, 3 M Pa, no TSZ
X-dir, 6 M Pa, no TSZ Y-dir, 6 M Pa, no TSZ
X-dir, 10 M Pa, no TSZ Y-dir, 10 M Pa, no TSZ
X-dir, 3 Mpa, after stimulation Y-dir, 3 M Pa, after stimulation
X-dir, 6 M Pa, after stimulation Y-dir, 6 M Pa, after stimulation
X-dir, 10 M Pa, after stimulation Y-dir, 10 M Pa, after stimulation
Total stress reduction due to stimulation : 2.4 MPa
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
1.0 5.0 9.0 13.0 17.0 21.0 25.0 29.0 33.0
k
/
k
0
Radial distance/Wellbore radius ratio, r/R
w
Permeability change : CO
2
injection at 6 MPa for cavitated well, C
f
= 2.9 e-8 Pa
-1
X-dir, Before stimulation X-dir, After stimulation
Y-dir, Before stimulation Y-dir, After stimulation
Improvement in permeability : 10%
















The highlights of this case study are as under :
1. The influence of disturbed zone by cavitated well has been advantageous : the sorption induced stress
by CO
2
in Y-direction at wellbore reduced by 9 MPa and by 5 MPa away from it, than without it
2. With higher injection pressures of CO
2
, the Y-direction stress reduced by 2.4 MPa but there was no
particular advantage of lower stress away from the wellbore or in far field (refer Fig. 17).
3. Permeability improvement was 10% due to stimulation during CO
2
injection at 6 MPa pressure (refer
Fig. 18). Substantial additional storage capacity under the influence of 3 times bigger radius of TSZ
can be created near wellbore of 9 times larger volume as compared to a normal 6” diameter wellbore.

10. Sensitivity analysis

The influence of Langmuir parameters on total stress including elastic properties : E, v, and o
S
was analysed
depending on the coal rank to correlate the matrix deformation with permeability change, with wellbore radius
of 0.0762 m, TSZ radius of 1.1 m, but without the influence of cavitated well. Identical geometry, initial
conditions and boundary conditions for the models were maintained as in the base case.
10.1 Elastic properties

As the rank of coal increases, there is an increase in adsorption capacity, volumetric strain and a resultant
swelling coefficient. The highest permeability loss caused by coal swelling is associated with the stronger
adsorbing gas and mechanically weakest coals (Durucan et al. 2009). There have been studies on high pressure
gas adsorption effects on porous media where adsorption induced swelling models have been used to relate the
impact of micro-fracturing on swelling coefficient of coal, which is related to its adsorption capacity and elastic
properties (Pan and Connell 2007). The elastic and transport properties assumed for the analysis were as in
Table 10 (source : Imperial College London laboratory experiments for research).

Table 10 : Elastic and transport properties used for sensitivity analysis

Properties High rank coal, fixed carbon : 90.4 (daf %) Low rank coal, fixed carbon : 56.4 (daf %)
No TSZ With TSZ No TSZ With TSZ
E, GPa 2.14 1.80 3.20 2.95
v 0.30 0.27 0.30 0.25
oS, kg/m
3
0.50 0.40 0.50 0.425
k, mD 5 50 1 10
VL (CO2), m
3
/kg 0.022 0.022 0.015 0.015
PL (CO2), MPa 0.77 0.77 3.5 3.5
Cf, Pa
-1
2.9e-8 2.9e-8 1.0e-8 1.0e-8


Fig. 17 : CO2 induced total stress at TSZ (r/Rw=4.4) for
various injection pressures
Fig. 18 : Combined influence of TSZ and cavitated well
on Permeability change, CO2 injection at 6 MPa
12 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.0 5.0 9.0 13.0 17.0 21.0 25.0
k
/
k
0
Radial distance/Wellbore radius ratio, r/R
w
Permeability change near wellbore region : CO
2
injection for at 3 MPa,
for high and low rank coals
X-dir, High rank, before stimulation X-dir, High rank, after stimulation
Y-dir, High rank, before stimulation Y-dir, High rank, after stimulation
X-dir, Low rank, before stimulation X-dir, Low rank, after stimulation
Y-dir, Low rank, before stimulation Y-dir, Low rank, after stimulation
High rank, C
f
= 2.9e-8
Low rank, C
f
= 1.0e-8
0.00
0.20
0.40
0.60
0.80
1.00
1.20
5.0E-09 1.0E-08 1.5E-08 2.0E-08 2.5E-08 3.0E-08
k
/
k
0
Compressibility factor, C
f
Permeability change at wellbore in Y-direction : CO
2
injection at
3 MPa, for high and low rank coals
High rank, before stimulation High rank, after stimulation
Low rank, before stimulation Low rank, after stimulation
1.E-05
1.E-04
1.E-03
1.E-02
1.E-01
1.E+00
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
k
/
k
0
Reservoir Pressure, MPa
Sensitivity of C
f
and elastic properties on Permeability change
High rank, C
f
= 2.9e-7
High rank, C
f
= 2.9e-8
Low rank, C
f
= 1.0e-7
Low rank, C
f
= 1.0e-8
After stimulation
0
5
10
15
20
25
30
35
40
45
1.0 5.0 9.0 13.0 17.0 21.0 25.0
R
a
d
i
a
l
a
n
d
T
a
n
g
e
n
t
i
a
l
S
t
r
e
s
s
e
s
,

M
P
a
Radial distance / Wellbore radius ratio, r/R
w
Influence on total Stress due to CO
2
injection for 1 day @ 3 MPa,
with high and low rank coals
X dir, base case Y-dir, base case
X-dir, High rank, before stimulation Y-dir, High rank, before stimulation
X-dir, high rank, after stimulation Y-dir, High rank, after stimulation
X-dir, Low rank, before stimulation Y-dir, Low rank, before stimulation
X-dir, Low rank, after stimulation Y-dir, Low rank, after stimulation
Low rank coal induced stress
lower by 8 MPa
High rank coal
Low rank coal
Base case














The plots on stress and permeability change on Fig. 19 and Fig. 20, respectively indicate the following :
1. Substantial decrease (32%) in total stress level from 25 MPa for high rank to 17 MPa for low rank coal
2. Permeability change are same at 16 % for low rank coal and high rank coal, for the equivalent
reduction in stress tensor in Y-direction
3. Low rank coals can have bigger CO
2
injection rates and thus higher ECBM recovery, due to enhanced
role of the effective stress
4. High rank coals have stronger strain capacity of CO
2
and resultant permeability reduction can
negatively influence CO
2
injectivity and replacement of CBM.
10.2 Cleat compressibility

The permeability change (k/k
0
) depends on the response of coalbed to stresses and compressibility data (C
f
)
obtained from history match data. The base case assumption of C
f
has been as per Table 10 and a range of
resultant k/k
0
have been plotted on Fig. 21 and Fig. 22 for comparing high rank coal with a low rank coal at 3
MPa pressure. The higher pore pressures induce opening of cleats thereby improving the permeability change.
The low rank coals did not induce permeability changes over a wide range of C
f
values.










11. Discussion

- The numerical model that was based on theoretical principles and laboratory experimental data has
demonstrated that cold N
2
injection can effectively lower volumetric strains and permanently transform
the elastic properties and swelling characteristic of coal near wellbore region. There was a strong
influence of N
2
injection temperatures and pressures but marginal influence of CO
2
injection pressures.
This is mainly due to combined total stress induced by wellbore pressure, high far field confining stress
and swelling stress which are dominant as compared to the low initial reservoir pressure, causing a
sharp decline in pressure near wellbore to rapidly attain equilibrium (refer Fig. 12 and Fig. 13).
- The injection pressure when increased by over 3 times from 3 MPa to 10 MPa showed a proportional
decrease in permeability change (65% compared to 19%) using the Y-direction component of effective
stress, due to increased sorption and induced swelling at higher pressures, as in Fig. 23.
Fig. 19 : Sensitivity of rank of coals on CO2 induced
total stress at TSZ (r/Rw=14.4) for 3 MPa
Fig. 20 : Sensitivity of rank of coals on Permeability
change
Fig. 21 : Sensitivity of permeability change to changes
in pore pressure v/s Cf, for high and low rank coals
Fig. 22 : Sensitivity of permeability change to rank of
coal v/s Cf
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 13


0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0
k
/
k
0
Reservoir Pressure, MPa
Permeability dependence on Pressure : CO
2
injection for 1 day
(Y-direction stress component, C
f
= 2.9e-8 Pa-
1
)
Before stimulation 3 MPa
After stimulation 3 MPa
Before stimulation, 6 MPa
After stimulation, 6 MPa
Before stimulation, 10 MPA
After stimulation, 10 MPa
3 MPa 6 MPa 10 MPa
- The radial extent of TSZ can be substantially increased in combination with a cavitated well that would
ease the severe permeability reduction around wellbore. However, it can be anticipated that the
injection volume of N
2
required would be much higher with prolonged injection cycles or higher
pressures due to large wellbore diameter. Depending upon objectives of the field (CO
2
storage or
methane recovery), operational constraints (bottom hole pressure increase or using cavitated well for
production) and economics, it would determine the utility of this option with high permeability
formations. The geological and geophysical constraints would also be important factors to decide
appropriate completion and stimulation technique.
- The relative influence of swelling coefficient (o
s
) on CO
2
sorption load after thermal stimulation
revealed over 1 MPa total stress impact. However, an integrated reservoir modelling approach would
quantify the injectivity in terms of bottom hole pressure calculations and CO
2
storage capacity.
- Cleat volume compressibility (C
f
) was very sensitive for permeability change to the extent that one-
order of magnitude increase in C
f
reduced the k/k
0
ratio by four-orders of magnitude, as in Fig. 21.
- The model confirmed that coal rank has indeed a converse influence on the CO
2
injection performance
making low rank coals more suitable for ECBM (Chen et al. 2009) and thermal stimulation, but would
be strongly influenced by its heterogeneity.















Fig. 23 : Permeability change dependence on pressure for CO2 injection
12. Conclusions and future work

The model results provided insights into possible variations in stress and permeability during CO
2
injection with
the modified elastic and flow properties due to TSZ and relation of permeability change with its injection
pressure and compressibility. The base case stress reduction was 2.2 MPa equivalent to permeability
improvement by 16% near wellbore region, due to presence of TSZ. The cavitated well case study had a stress
reduction of 2.4 MPa equivalent to permeability improvement by 10%.
The gas storage volumes were estimated for base case model of 1000 m square field using Langmuir’s
isotherm (Fig. 3) at 10 MPa pressure as 40 million std. m
3
(MM SCM) for CO
2
and 20 MM SCM for methane.
As per sensitivity analysis, assuming a conservative improvement of 10% of permeability change equivalent to
higher CO
2
injectivity at same bottom hole pressure, the benefit of thermally induced stimulation is 2 MM SCM
of additional methane production equivalent to generation of 4 MM SCM extra CO
2
storage capacity.

Further studies are recommended for :
1. Laboratory testing of thermal effects on samples from cavitated well disturbed zone and low rank coals
2. Thermal stress and sorption dependent permeability change predictions integrated with reservoir
simulation model
3. Horizontal well in 3-D model
4. Field trials for additional data on periodicity of cyclic N
2
injection, saturation and heterogeneity.

14 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

Nomenclature

Field variables :
C
f
= cleat volume compressibility, Lt
2
/m, Pascal
-1

C
m
= matrix gas concentration, fraction, SCM/m
3
[scf/ft
3
]
C
p
= volumetric heat capacity of porous media, L
2
/Tt
2
, Joule/kg
o
K
C
pL
= volumetric heat capacity of flowing fluid, L
2
/Tt
2
, Joule/kg
o
K
g = gravity, acceleration of, L/t
2
, 9.81 m/sec.
2

G = Lamé elastic constant (shear modulus), m/Lt
2
, GPa
GPa = giga (1e9) Pascal, m/Lt
2

h

= heat transfer coefficient, m/Tt
3
, W/m
2

o
K
H = hydraulic head, L, m
k = cleat permeability, L
2
, mD
k
T
= thermal conductivity of fluid, mL/Tt
3
, W/m
o
K
k
p
= thermal conductivity of porous media, mL/Tt
3
, W/m
o
K
K = bulk modulus, m/Lt
2
, GPa
Kcal = kilo calories, mL
2
/t
2

mD = milli Darcy L
2
, (1e-15 m
2
)
MM = million (1e6)
MPa = mega (1e6) Pascal, m/Lt
2

m = meter, L
p = pore pressure, m/Lt
2
, Pascal [psia]
psi = pounds per square inch
P
L
= Langmuir pressure, m/Lt
2
, Pascal [psia]
Q
s
= volumetric flow rate per unit volume of reservoir, 1/t, sec.
-1

Q = heat source or sink, m/Lt
3
, W/m
3

r = radial distance, L, m
R
w
= radius of wellbore radius, L, m
R
c
= radius of cavitated well’s disturbed zone, L, m
SCM = standard cubic meter
t = time, t, seconds
T = temperature, T,
o
Kelvin (÷
o
K)
u = vector of directional velocities : ‘u’ and ‘v’ in 2-D, L/t, m/sec. (Darcy Flow)
u = vector of directional displacements : ‘u’ and ‘v’ in 2-D, L, m (Solid Mechanics)
V
L
= Langmuir volume, L
3
/m, m
3
/kg [scf/ton]

Subscripts :
0 = initial state
u = angle from X-axis
B = bulk
f = fracture or formation
i, j = direction or component ‘i’ and ‘j’
inj. = injection
L = Langmuir’s isotherm
X, Y, Z = x, y and z directions, respectively

Symbols :
o
B
= Biot’s effective stress coefficient, fraction
o
S
= swelling / shrinkage coefficient, m/L
3
, kg/m
3

o
T
= coefficient of thermal expansion, 1/T, 1/
o
K
| = thermal coefficient, m/LTt
2
, Pa/
o
K
E = Young’s modulus, m/Lt
2
, GPa
| = matrix porosity, fraction
o = stress tensor, m/Lt
2
, MPa
c = volumetric strain, fraction
µ = density, m/L
3
, kg/m
3

q = dynamic viscosity, m/Lt, Pascal sec.
v = Poisson’s ratio, fraction
A = delta or difference
V = vector operator, 1/L, m
-1

O = real domain


Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 15


Acknowledgement

The financial support provided by M/s Shell International Exploration and Production for an investigation into
the potential permeability effects of cold CO
2
injection for ECBM recovery during 2005, is gratefully
acknowledged.

References
Abousleiman, Y. and Ekbote, S. 2005. Solutions for the inclined borehole in a porothermoelastic transversely isotropic
medium. Journal of Applied Mechanics, 72, 102-113. doi : 10.11115/1.1825433
Charlez, P., Lemonnier, P., Ruffet, C., Bouteca, M.J., Tan, C. 1996. Thermally induced fracturing : analysis of a field case in
North Sea. SPE 36916, European Petroleum Conference, 22-24 Oct.1996, Milan, Italy, doi : 10.2118/36916-MS
Chen, Z., Liu, J., Elsworth, D., Connell, L., Pan, Z. 2009 Investigation of CO2 injection induced coal-gas interactions.
Proceedings of the 43rd US Rock Mechanics Symposium and 4th US-Canada Rock Mechanics Symposium 2009,
Asheville, North Carolina, USA, ARMA 09-099
COMSOL Multiphysics ®. Nov. 2008 & May 2010. Earth Science & Multiphysics Modules User’s Guide. [Version 3.5 &
4.0] Burlington, USA, COMSOL, Inc.
Cui, X., Bustin, M., and Chikatamarla, L. 2007. Adsorption-induced coal swelling and stress : Implications for methane
production and acid gas sequestration into coal seams. Journal of Geophysical Research, 112 (B 10202)
Durucan, S. and Edwards, J.S. 1986. The effects of stress and fracturing on permeability of coal. Mining Science and
Technology, 3 , 205-216.
Durucan, S., Ahsan, M., Shi, JQ. 2009. Matrix shrinkage and swelling characteristics of European coals. Energy Procedia, 1
(1), 3055-3062.
Durucan, S. and Shi, JQ. 2009. Improving the CO
2
well injectivity and enhanced CBM production performance in coal
seams, International Journal of Coal Geology, 77, 214-221.
Durucan, S., Shi, JQ., and Grattoni, C.G. 2005. Thermal stimulation in coal: an investigation into the potential permeability
effects of cold CO
2
injection for ECBM recovery. A Confidential Technical Report to Shell International E&P
Engineering ToolBox Ref. http://www.engineeringtoolbox.com, accessed 15 August 2010.
Erickson, D. 2002. Overview of ECBM commercial demonstration pilot at Tiffany unit. Internal Report, BP America Inc.
Harpalani, S. and Chen, G. 1995. Influence of gas production induced volumetric strain on permeability of coal. Geotech
Geological Engineering, 15, 303-325.
Herrin, J. and Deming, D. 1996. Thermal conductivity of US coals. Journal of Geophysical Research, 101, 381-386.
Indenbaum, V. M. 1974. Calculation of stresses in multilayer cylindrical articles made of variously combined composite
materials. Mechanics of Composite Materials, 10 (1), 48-52.
Jaeger J.C., Cook N.G.W., Zimmerman R.W. 2007. Fundamentals of Rock Mechanics 4th edition, Blackwell.
Langmuir, I. 1916. The constitution and fundamental properties of solids and liquids. Journal of the American Chemical
Society, 38, 2221-2295.
Marsden, R. J. 1997. An introduction to near-wellbore rock mechanics. Lecture Notes at Imperial College.
Mazumder, S., Karnik, A., Wolf, K. 2006. Swelling of coal in response to CO
2
sequestration for ECBM and its effect on
fracture permeability. SPE Journal, SPE 97754-PA, 11 (3) 390-397. doi : 10.2118/97754-PA
McDaniel, B.W., Grundmann, S.R., Kendrick, W.D. 1997.. Field applications of cryogenic Nitrogen as a hydraulic
fracturing fluid. SPE Annual Technical Conference and Exhibition, 5-8 Oct. 1997 Texas, USA, SPE 38623. 561-572.
McKee, C.R., Bumb, A.C., Koenig, R.A. 1988. Stress-dependent permeability and porosity of coal and other geologic
formations. SPE Formation Evaluation, SPE 12858-PA, 3 (1), 81-91.
NIST Ref. http://webbook.nist.gov/chemistry/fluid/, accessed 16 August 2010.
Palmer, I. and Mansoori J. 1998. How permeability depends on stress and pore pressure in coalbeds : a new model. SPE
52607-PA, SPE Reservoir Evaluation & Engineering, 1 (6), 539-544. doi : 10.2118/52607-PA
Palmer, I. 2009. Permeability changes in coal : analytical modelling. International Journal of Coal Geology, 77, 119-126.
Pan, Z. and Connell, L. D. 2007. A theoretical model for gas adsorption-induced coal swelling. International Journal of
Coal Geology, 69, 243-252.
Pashin, J. C. and McIntyre M.R. 2003. Temperature - pressure conditions in CBM reservoirs of Black Warroir basin :
implications for carbon sequestation and ECBM recovery. Coal Geology, 54, 167-183.
Perkins, T. K. and Gonzales, J. A. 1985. The effect of thermoelastic stresses on injection well fracturing. SPE Journal
11332-PA, 25 (1), 78-88.
Reeves, S. R. 2001. Geologic sequestration of CO
2
in deep, unmineable coalbeds : an integrated research and commercial-
scale field demonstration project. Proceedings of the SPE Annual Technical Conference and Exhibition, September 30-
October 3, 2001 New Orleans, SPE 71749.
Reeves, S.R., Clarkson, C., Erickson, D. 2002. Selected field practices for ECBM recovery and CO
2
sequestration in coals
based on experience gained at the Allison and Tiffany units,. DOE Topical Report, DE-FC26-00NT 40924.
Reeves, S.R., Taillefert, A., Pekot, L., Clarkson, C. 2003. The Allison Unit of San Juan basin, CO
2
– ECBM Pilot : A
Reservoir Modelling Study. DOE Topical Report, DE-FC26-0NT40924.
Reeves, S.R. and Oudinot, A. 2004. The Tiffany Unit N
2
- ECBM Pilot : a reservoir modelling study. Advanced Resources
International, Incorporated, DOE FC26-00NT40924.
Reiss, L. H. 1980 edition. The reservoir engineering aspects of fractured formations, Technip Publication.
Ross, H., Hagin, P., Zoback, M. 2009. CO
2
storage and ECBM recovery : reservoir characterisation and fluid flow
simulations-Powder River Basin, USA. International Journal of Greenhouse Gas Control, 3, 773-786.
Saghafi, A., Faiz, M., Roberts, D. 2007. CO
2
storage and gas diffusivity properties of coals from Sydney Basin, Australia.
International Journal of Coal Geology, 70 (1-3), 240-254.
16 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

Seidle, J.P. and Huitt, L.G. 1995. Experimental measurement of coal matrix shrinkage due to gas desorption and
implications for cleat matrix increase. SPE International Meeting on Petroleum Engineering, 14–17 November 1995
Beijing, China., SPE 30010-MS. pp. 575, doi : 10.2118/30010-MS
Shi, JQ. and Durucan, S. 2004. Drawdown induced changes in permeability of coalbeds: A new interpretation of the
reservoir response to primary recovery. Transport in Porous Media, 56, 1-16.
Shi, JQ. and Durucan, S. 2005. A model for changes in coalbed permeability during primary and enhanced methane
recovery. SPE Reservoir Evaluation and Engineering, SPE 87230-PA, 8 (4), 291-294. doi : 10.2118/87230-PA
Shi, JQ., Durucan S., Fujioka M. 2008. A reservoir simulation study of CO
2
injection and N
2
flooding at the Ishikari
coalfield CO2 storage pilot project, Japan, International Journal of Greenhouse Gas Control, 2, 47–57.
Siedle, J., Jeansonne, M.W., Erickson, D.J. 1992. Application of matchstick geometry to stress dependent permeability in
coals. 1992 SPE Rocky Mountain regional meeting Casper, Wyoming, 18-21 May. SPE 24361-PA, doi :
10.2118/24361-PA
Van Bergen, F., Pagnier, H., Krzystolik. 2006. Field experiment of enhanced coalbed methane-CO
2
in the upper Silesian
basin of Poland, Environmental Geosciences 13 (3), 201–224.
Walsh, J. B. 1965a. The effects of cracks on compressibility of rocks. Journal of Geophysical Research, 70 (2), 381-389.
Walsh, J.B. 1965b. The effects of cracks on the uniaxial elastic compression of rocks. Journal of Geophysical Research, 70
(2), 399-411.
Zhu, W.C., Liu, J., Sheng, J.C. and Elsworth, D. 2007. Analysis of coupled gas flow and deformation process with
desorption and Klinkenberg effects in coal seams. Rock Mechanics and Mining Sciences, 44, 971-980.


Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 17


CO
2
injection : preferential
sorption
Matrix swelling induced
stresses
Permeability reduction
near wellbore
Cold N
2
injection : thermal
stimulation
Improved elastic properties
Enhanced permeability, injectivity
and CO
2
storage
Appendix-A : Figures for background information






















Fig. A-2 : Process flow of the model



Fig. A-1 : Cleat networks of coal seams with dual-
porosity structure consisting micro-pores within matrix
and macro-cleats / fractures (Cui et al. 2007)
Fig. A-3 : Measured gas isotherms comparison
(Shi et al. 2008)
18 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

Appendix-B : Governing equation for total stress


Solid deformation :

Defining the solids deformation on plane strain, the governing equation is :

) H g ) u .(
) 2 1 )( 1 ( 2
E
u
) 1 ( 2
f B
2
V = V V
÷ +
+ V
÷
µ o
v v v
E
.......................................................... (B-1)

where; E is Young’s modulus (kg m
-1
s
-2
); v is Poisson’s ratio, while u is the total displacement and
consists of directional displacements u and v (m). The term α
b
ρ
f
g∇H amounts to the fluid pressure gradient in
the x- and y-directions multiplied by the Biot’s poro-elastic constant (

and it is often described as the
fluid-to-structure coupling expression. The plane-strain problem is converted to a convenient system of
equations, using Navier’s equation that relates stress, strain and displacement in solids. At equilibrium
conditions, the stresses are given by :

| | F = Vo ........................................................................................................................ (B-2)

where; [o] is the stress tensor and F is the vector that contains the fluid pressure gradient (fluid to
structure coupling term) and other body loads. The stress-strain relationship for an isotropic solid under plain
strain conditions is represented by :


(
(
(
¸
(

¸

(
(
(
(
¸
(

¸

÷
÷
÷
÷ +
=
(
(
(
¸
(

¸

xy
y
x
xy
y
x
2
2 1
0 0
0 1
0 1
) 2 1 )( 1 (
E
¸
c
c
u
u u
u u
u u
t
o
o
...........................................................................(B-3)

where; t and ¸ denote the shear stress and shear strain, respectively.

Parallel to the flow model, the Solid Mechanics model uses equation to describe change in stresses (σ),
strains (ε) and displacement (u), from an initial state. This focus on change in displacement is standard in solids
modelling and greatly simplifies specifying loads, boundaries, and initial conditions.


Thermal stress :

Materials change volume with temperature causing thermal strains to develop in the materials. If the material
cannot freely deform in response to the volume change, thermal stresses will develop in the material depending
on the thermal expansion and heat capacity, modulus of elasticity and the degree of freedom. Therefore, in a
mechanical system under structural loads with varying temperatures the total stress will be combination of
structural and thermal stress for elastic material.
The thermo-elasticity has been accounted for the effect of changes in temperature on the stress and
displacement in the TPE model. Under the assumption of linearity, with the change in reservoir temperature (T
0
)
effect of injection gas temperature (T) induced strains are :

( )
0 T
T T ÷ ÷ = o c ...................................................................................................... (B-4)

where; o
T
is coefficient of (linear) thermal expansion of coal.

The thermally induced stress are caused by combination of temperature change and a mechanical restraint
that inhibits free expansion or contraction of the rock. The thermal stress is therefore represented as under :

( ) | |
0 T T
T T 3 E ÷ ÷ = o o ........................................................................................................ (B-5)

where, (3o
T
) is the volumetric thermal expansion coefficient. (Jaeger et al. 2007)

The equation coefficients can be spatially varying depending on the temperature T and can be linked to other
physics in the model.



Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 19


Total body load :

The overall body load, using the above basic total stress equations on swelling and thermal analogy, was
formulated for total body load / stress simulations :


(
¸
(

¸

+
÷ ÷ ÷ = V V
÷ +
+ V
÷ p P
p V
K ) T E 3 ( p ) u .(
) 2 1 )( 1 ( 2
E
u
) 1 ( 2
E
L
L
s T
2
o A o
u u u
...................................................(B-6)

where; terms on the right hand side are the body loads for the Navier’s equation, first term represents pore
pressure load, second term represents the thermal stress and third term represents the swelling stress.

20 Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds

0.0
5.0
10.0
15.0
20.0
25.0
30.0
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
R
a
d
i
a
l

a
n
d

T
a
n
g
e
t
i
a
l

S
t
r
e
s
s
e
s
,

M
P
a

Radial distance / Wellbore radius ratio, r/R
w
Model validation using analytical equations for stress at wellbore wall
Solid.sx (x-direction)
Radial (from analytical eqn.)
Solid.sy (y-direction)
Tangential (from analytical eqn.)
Appendix-C : Model validation

Circular cross-sectional holes are usually drilled as wells inside the rocks which cause displacement and
distribution of in-situ stresses near the wellbore. The principles of geomechanics would be applied in the
thermo-poro-elastic model by using the COMSOL Multiphysics® Solid Mechanics interface. Therefore, simple
redistribution and concentration of radial stress (o
rr
) and tangential stress (o
uu
) near to the wellbore considering
a “borehole in uniaxial reservoir” with identical geometry and far-field principal stresses (o
A
& o
B
) was chosen
for model validation purposes.
Considering the typical initial state situation of no pore pressure in the rock and no injection pressure in the
wellbore, axi-symmetric stresses are distributed as illustrated in Fig. C-1 and Fig. C-2. In case the far field
stresses are different (o
A
> o
B
), the stresses at point marked 1 would be greater than and 2, however for
simplicity, the far field stress are assumed as 15 MPa each. The numerical model results of stress tensors were
matched with analytical equations as under : (Jaeger et al. 2007) :

( ) ( ) u o o o o o 2 Cos
r
R
3
r
R
4 1
2
1
r
R
p
r
R
1
2
1
4
w
2
w
2 1
2
w
2
w
2 1 rr
(
(
¸
(

¸

|
.
|

\
|
+ |
.
|

\
|
÷ ÷ + |
.
|

\
|
+
|
|
.
|

\
|
|
.
|

\
|
÷ + =
· · · ·
.............(C-1)
( ) ( ) u o o o o o
uu
2 Cos
r
R
3 1
2
1
r
R
p
r
R
1
2
1
4
w
2 1
2
w
2
w
2 1
(
(
¸
(

¸

|
.
|

\
|
+ ÷ ÷ |
.
|

\
|
÷
|
|
.
|

\
|
|
.
|

\
|
+ + =
· · · ·
......................... (C-2)
where;

: far field stresses at 15 MPa are equal in X and Y directions, u=0 along X-axis, R
w
: wellbore
radius of 0.0762 m (6” dia.), r : radial distance from wellbore for a square field of 1000 m sides, p : injection
pressure at 6 MPa. The initial mean stress state is given by : o
0
= (o
rr
+ o
uu
)/2. The radial stress for u=0
would be same as stress tensor in x-direction and the tangential stress would be same as in y-direction, which
has been considered later in extracting the data from simulation. The plot indicated a good match thus validating
the model in its intended conditions and configuration as per Fig. C-3.





































Fig. C-3 : Validation of gas injection case using analytical equations for stress distribution at wellbore region

Fig. C-1: Typical stresses around wellbore
(Marsden, 1997)
Fig. C-2 : Stress tensors from COMSOL model, solid.sx
in X-direction (in red) and solid.sy : Y-direction
Numerical modelling of thermo-poro-elastic behaviour of injection wellbore and its impact on CO2 storage in coalbeds 21



Thermal properties of coalbed Symbol Units Value Reference
Reservoir temperature - initial T
o
K 313
(Pashin and McIntyre
2003)
Volumetric heat capacity C
p
J / Kg
o
K 1330
Engineering Toolbox
website 2010
Thermal conductivity k
T
W/m
o
K 0.3-0.8
(Herrin and Deming
1996), (Zhanxue 2005)
Coefficient of thermal expansion
o
T

1/
o
K 2.00E-05 (Durucan et al. 2005)

Elastic properties Symbol Units Value Reference
Coal density µ kg/m
3
1250-1330 (Zhu et al. 2007)
Young's modulus E G Pa 2.14 (Durucan et al. 2009)
Poisson's ratio v -- 0.30 (Durucan et al. 2009)
Langmuir's volume (CO2) VL CO2 m
3
/ kg 0.022 (Durucan et al. 2009)
Langmuir's volume (N2) VL N2 m
3
/ kg 0.009 (Durucan et al. 2009)
Langmuir's pressure (CO2) PL CO2 M Pa 0.77 (Durucan et al. 2009)
Langmuir's pressure (N2) PL N2 M Pa 6.80 (Durucan et al. 2009)
Cleat volume compressibility Cf 1/ Pa 2.9E-08 (Ross, et al. 2009)
Matrix shrinkage/swelling
coefficient (CO2)
oS
Kg/m
3
0.5 (Durucan et al. 2009)
Matrix shrinkage/swelling
coefficient (N2)
oS
Kg/m
3
0.1 (Durucan et al. 2009)
Fluid properties Symbol Units Value Reference
Density of CO2 (6 MPa, 313
o
K) µ CO2 kg/m
3
149 (NIST website 2010)
Density of N2 (6 MPa, 248
o
K) µ N2 kg/m
3
70 (NIST website 2010)
Dynamic viscosity CO2 q CO2 Pascal sec 1.668e-5 (NIST website 2010)
Dynamic viscosity N2 q N2 Pascal sec 1.650e-5 (NIST website 2010)
Heat capacity CO2 C
p
CO2 Joule/Kg/
o
K 2220 (NIST website 2010)
Heat capacity N2 C
p
N2 Joule/Kg/
o
K 1175 (NIST website 2010)
Thermal conductivity CO2 k
T
CO2 W/m
o
K 0.0245 (NIST website 2010)
Thermal conductivity N2 k
T
N2 W/m
o
K 0.0252 (NIST website 2010)
Cleat porosity | -- 0.10 (Ross et al. 2009)
Cleat permeability k m
2
5 E-15 (Ross et al. 2009)

Thermally cycled values
Appendix-D : Input data tables

Table D-1 : Fluid flow properties



















Table D-2 : Elastic properties



















Table D-3 : Thermal properties









Laboratory tested properties of coal at low temperature (Durucan et al. 2005)

The elastic and flow properties for coal tested over a range of low temperature up to 248
o
K in the laboratory
while at further lower temperatures of 223
o
K and 173
o
K are not known. Therefore, for modelling purpose
similar properties have been maintained for all temperature ranges. The coals were subjected to the temperatures
mentioned for more than 24 hours. The following plot (Fig. C-1) depicts the variation of the elastic modulus
with time, wherein solid symbols denote tests conducted on coal samples that have undergone 3 thermal cycles :













Fig. C-1 : Imperial College, London lab. experimental findings for elastic properties of coal at low temperatures
(Durucan et al. 2005)