You are on page 1of 8

Available online at www.sciencedirect.

com


Energy Procedia 00 (2010) 000000

Energy
Procedia

www.elsevier.com/locate/XXX

GHGT-10
CO
2
Transport Depressurization, Heat Transfer and Impurities
Gelein de Koeijer
a
, Jan Henrik Borch
a
, Michael Drescher
b
, Hailong Li
b
, ivind Wilhelmsen
b
, Jana
Jakobsen
b

a Statoil ASA, Research & Technology,
NO-7005 Trondheim, Norway
b SINTEF Energy Research,
NO-7465 Trondheim, Norway
Elsevier use only: Received date here; revised date here; accepted date here
Abstract
Detailed knowledge on depressurization, heat transfer and impurities in CO
2
pipelines has shown to be important for safe and cost effective
design of CCS chains. The aim of this paper is to present the latest results on the modelling and experimental verification of these aspects. A
newly developed CO
2
module in the flow engineering software OLGA from SPT Group has been used to model experimental results on
depressurization, which is a two-phase transient flow with evaporation and Joule-Thomson cooling. A visual comparison of model and
experiments in PT-diagrams showed that the new CO
2
module in OLGA captures the depressurization behaviour reasonably well, but still has
room for improvement. Heat transfer coefficients for heat transfer from aqueous surroundings to cold CO
2
in a pipeline segment was measured
in a dedicated rig. A thin layer of ice formation was observed on the bottom of the pipeline. The overall heat transfer coefficient at the
experimental conditions has been calculated to be 45.0 W/m
2
K, where the outer heat transfer coefficient was 162.3 W/m
2
K, and the inner heat
transfer coefficient 166.2 W/m
2
K. 46 Experiments and modelling of H
2
O solubility in vapour and liquid CO
2
were performed for verifying the
VLHE-behaviour at low H
2
O concentrations down to -50 C. Hydrafacts model HWHyd 2.2 model predicted the experimental result with
average 8.2% deviation. Hence, the model can be used for setting hydrate formation limited H
2
O specifications in transport of pure CO
2
.
2010 Elsevier Ltd. All rights reserved
Keywords: CO2 transport, Depressurization, Heat Transfer, H2O solubility
1. Introduction
This paper will discuss progress on CO
2
transport R&D within the CO
2
IT IS project [1], which stands for CO
2
Interface-
Transport-Interface-Storage. This project is lead by Statoil and has SINTEF Energy Research as partner. It is partially funded by
the CLIMIT program from the Research Council of Norway.
Statoil is operator of 3 of the 4 commercial CCS projects in the world which all have CO
2
transport. The Snhvit project
(Northern Norway) operates a 153 km offshore pipeline transporting liquid CO
2
from an LNG plant to one subsea well. At the
Sleipner field (North Sea) the CO
2
is transported a short distance near the critical point [2] between two connected offshore
platforms. The CO
2
capture unit is on one platform, while the wellhead is connected to the other. The In Salah operations in
Algeria transport CO
2
in dense phase to three injection wells. Statoil has also business development ongoing for future CCS
projects where transport plays an important role connecting capture and storage. The R&D in this project is motivated by
improving the HSE and reducing the costs in existing and future CCS chains. This project focuses on acquiring general
knowledge that can be used for all combinations of capture, transport and storage technologies and can support current
operations and business development.
Detailed knowledge on depressurization, heat transfer and impurities in CO
2
pipelines have shown to be important for safe
and cost effective design of CCS chains. The aim of this research project is to experimentally verify and model these aspects of
CO
2
transport. The aim of the paper is to present the latest results on:

Corresponding author. Tel.:+47 51990000; fax: +47 73584282.


E-mail address: gdek@statoil.com
c 2011 Published by Elsevier Ltd.
Energy Procedia 4 (2011) 30083015
www.elsevier.com/locate/procedia
doi:10.1016/j.egypro.2011.02.211
2 Author name / Energy Procedia 00 (2010) 000000
x pipeline depressurization by comparing results from the newly developed CO
2
module in the flow engineering software
OLGA from SPT Group to experiments
x heat transfer from pipeline surroundings to cold evaporating liquid CO
2
in the pipeline
x experiments and modeling of H
2
O solubility in vapour and liquid CO
2
performed by Hydrafact
2. Depressurization
Uncontrolled depressurization of liquid or dense phase CO
2
can lead to temperatures below the design temperature of the
pipeline. Operations below design temperature should be avoided for maintaining pipeline integrity and guarantees.
Depressurization of liquid CO
2
is a two-phase transient phenomenon with Joule-Thomson cooling and vaporization. Improved
accuracy of models describing this phenomenon is important for designing procedures for transient operations and improved
design. For this purpose an R&D rig is constructed, and experimental campaigns and modelling efforts have been performed
earlier [1]. A new experimental campaign has been executed on the depressurization rig. Experiments of steady state gas and
liquid single phase flow, steady state 2-phase flow and various transient flows were performed. The most important part of the
rig is a 139 m CO
2
pipeline with 1 cm inner diameter and pressure and temperature measurements at 4 locations along the
pipeline: 0 m, 50 m, 100 m and 139 m.
A study has been performed to model these results from the depressurization rig with a new CO
2
module in the commercial
flow engineering simulator OLGA from SPT group [3]. This module is now commercially available, but lacks experimental
verification. The aim of this study was to perform an assessment of the tool as a first step in the experimental verification. After
this study the modelling results were compared with the experimental results. An example of one of the best results is given in
Figure 1 and Figure 2. The experiment was a quick depressurization from a liquid at ~-10 C with a valve closed at 0 m and
opened fully at 139 m. The modelling and experimental results are plotted in a PT-diagram at the 4 different locations along the
pipeline, together with the boiling line of pure CO
2
.

0
5
10
15
20
25
30
35
40
-70 -60 -50 -40 -30 -20 -10 0
Temperature (C)
P
r
e
s
s
u
r
e

(
B
a
r
a
)
exp 100m
exp 139m
OLGA 100m
OLGA 139m
Boiling line

Figure 1 Comparison of experimental and modelling results of pressure
and temperature during depressurization at 100 and 139 m (end) along
pipeline

0
5
10
15
20
25
30
35
-60 -50 -40 -30 -20 -10 0
Temperature (C)
P
r
e
s
s
u
r
e

(
B
a
r
a
)
exp 0m
exp 50m
OLGA 0m
OLGA 50m
Boiling line

Figure 2 Comparison of experimental and modelling results of pressure
and temperature during depressurization at 0 (start) and 50 m along
pipeline

Both experiment and model show a short initial liquid depressurization until the boiling line is met. The depressurization
continues as a boiling liquid following the boiling line. The moment all liquid is vaporized gives a bend in the PT-diagram,
where vapour depressurization continues until atmospheric pressure is obtained. A visual comparison shows that the new CO
2

module in OLGA captures the depressurization reasonably well, and that there is still has room for improvement. It is likely that
better experimental accuracy, more experimental verification and empirically adapted models for liquid slip and heat transfer can
reduce the discrepancies.
In some CCS chains the CO
2
contains other components with several mole percent, which may have an effect on the
depressurization behaviour. Depressurization of multi-component CO
2
mixtures is a topic of ongoing fundamental R&D in other
projects like CO
2
Dynamics

[4].
G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015 3009
Author name / Energy Procedia 00 (2010) 000000 3
3. Heat transfer
During depressurization or any other transient operation (e.g. start-up, load change, performance check of valves etc.) in
liquid CO
2
pipelines, heat transfer will mostly occur from the surroundings into the pipeline. The CO
2
cools down due to Joule-
Thomson effect and vaporization of CO
2
. This type of heat transfer is an uncommon situation in natural gas pipeline operation
and not studied extensively. Below 0 C even ice can form on the outside of the pipeline. This may change the heat transfer
coefficient significantly over time, and creates another transient phenomenon. In order to measure and model these effects
accurately, experiments have been performed on a new rig with a piece of the Snhvit CO
2
pipeline suspended in a temperature
controlled insulated box with multiple temperature sensors, which is described in earlier work [1]. Experiments are performed by
adding cold liquid CO
2
in the bottom of the pipeline segment. The CO
2
vaporizes due to heat transfer from the surroundings, and
the vapour exits at the top. Mass flow and temperatures are measured to give the transferred heat and associated heat transfer
coefficients. Figure 3 below shows the process layout of the test facility.


3.1. Experimental Investigations
Since the physical properties of water are well known, tap water was chosen as the initial surrounding substance. This is
essential in the initial calibration of the test facility and for the calculation of the initial heat transfer coefficients before applying
substances, whose physical properties vary a lot, such as sand and gravel.
For the initial testing the surrounding water in the container was adjusted to +6 C by an auxiliary glycol circuit. The CO
2

pressure was controlled by the cooling unit for the tank, which was set to a constant cooling duty. At start-up the measured mass
flow of the CO
2
vapour was relatively high, due to the temperature of the pipeline, which initially was the same as the
surrounding water. However, after a test operation of 12 hours the temperatures and mass flow were reaching significantly more
stable conditions. After a total of 48 hours of operation very stable values were attained and a constant pressure of 28.0 barg was
measured. This corresponds to a saturation temperature of -6.7 C. The mass flow was stable at ca. 6.5 kg/h and a stable
temperature profile in the pipeline wall had been established. Figure 4 illustrates the temperatures measured at different positions
in the test facility, and reveals the temperature gradients in the pipeline wall for the time period of one hour. Furthermore the
mass flow and pressure are shown. The heat transfer and respective heat transfer coefficients calculated in the following section
are based on this time period. After a total operation time of approximately 72 hours a thin layer of ice formation was observed
on the bottom of the pipeline. Formation of ice will be subject to future investigations.
3.2. Calculation background
The overall heat transfer coefficient from the pipeline to the surrounding container may be defined as:

T A
Q
h
tot
'


(1)
Here, Q

is the thermal energy transfer in [W], h is heat transfer coefficient in [W/Km


2
] and A is surface area of the heat being
transferred in [m
2
]. T is the temperature difference between the interior of the pipeline and the surrounding media. As heat is
Figure 3 Sketch of experimental set-up for measuring heat transfer coefficient of cold vaporizing CO2 in warmer surroundings
3010 G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015
4 Author name / Energy Procedia 00 (2010) 000000
transferred from the surroundings to the pipeline, CO
2
liquid will start to boil and the vaporized amount m is measured [kg
CO
2
/h]. The heat transferred to the CO
2
fluid is hence proportional to the vaporized mass of CO
2
:

m H Q
vap

' (2)


Here, 'H
vap
is the specific heat of vaporization in [J/kg]. The heat is transferred from the surrounding media to the pipeline
with an outer heat-transfer coefficient, h
out
. The heat is further conducted through an insulation layer and the pipe wall. Inside the
pipeline, the boiling-regime has an inner heat transfer coefficient, h
in
. The overall heat transfer coefficient through the tubular
part of the pipeline can be decomposed into four terms taking into account the different heat transfer phenomena:
out s iso in tot
h k
r r r
k
r r r
r h
r
h
1 ) / ln( ) / ln( 1
3 1 2 3 2 3
1
3
(3)
Here, r
3
, r
2
and r
1
denote the radii of the outer isolation, the outer and the inner steel walls respectively. k
iso
and k
s
denote the
thermal conductivity of the insulation and the steel. Applying the principle for resistances in series, the resistances to heat
transfer may be estimated.
3.3. Results
Figure 4 and Figure 5 show the measured steady-state results after a total operation time of 72 hours. Based upon the
measured flow rate of vaporized CO
2
part and the temperature difference between the water and the interior of the pipeline, the
overall heat transfer coefficient was calculated to be 45.0 W/m
2
K. The saturation temperature corresponding to a pressure of 28
barg inside the pipeline, matched well with the measured CO
2
temperature at the pipeline inlet and outlet.
The outer heat-transfer coefficient was calculated to be 162.3 W/m
2
K, and the inner heat transfer coefficient to 166.2 W/m
2
K.
Using empirical correlations for free convection over a horizontal cylinder [5] gives an estimate of 130 W/m
2
K for the outer heat
transfer coefficient. This shows that the most likely heat transfer mechanism at the outside of the pipeline is free convection. The
estimate also shows that over 20% deviation in the heat transfer coefficient should be expected if simplified theoretical models
-10
-8
-6
-4
-2
0
2
4
6
8
1100 1110 1120 1130 1140 1150 1160
Time unit [minute]
T
e
m
p
e
r
a
t
u
r
e

[

C
]
Temperature Pipeline out Temperature Pipeline in
Average water temperature Average temperature on outer insulation
Average temperature on outer steel
Figure 4 Measured temperatures at various radial positions as function of time
20
21
22
23
24
25
26
27
28
1100 1110 1120 1130 1140 1150 1160
Time unit [minute]
P
r
e
s
s
u
r
e

[
b
a
r
g
]
0
1
2
3
4
5
6
7
8
M
a
s
s

f
l
o
w

[
k
g
/
h
]
Pressure Flow
Figure 5 Measured vapour flow out of the pipe and pressure in the tank as function of time
G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015 3011
Author name / Energy Procedia 00 (2010) 000000 5
are used to estimate the heat transfer through the pipeline. It is reasonable to assume that heat is transferred by free convection in
the current experimental set-up given the relatively small temperature gradients at the inside and the outside of the pipeline (4.2
K and 3.6 K).
4. Impurities VLHE experiments on water solubility
4.1. Introduction
Every CO
2
stream has impurities. All of the 4 commercial CCS projects (Sleipner, Weyburn, In Salah, Snhvit) have dealt
with risks related to these impurities before. Currently, R&D is ongoing on the preferred and/or acceptable concentrations. In
these discussions, it is recommended to distinguish between guidelines, specifications and regulation. De Visser et al. [6] and
other R&D give guidelines. Specifications are unique to each CCS project, while the OSPAR convention gives the most updated
regulation. OSPAR is the legal instrument guiding international cooperation on the protection of the marine environment of the
North-East Atlantic, has approved a regulation stating that geologically stored CO
2
should contain overwhelmingly CO
2
[7].
The four large scale CCS value chains in operation have already experienced a broad range of impurity concentrations. For
example, at Sleipner the CO
2
is nearly saturated with water, while at Snhvit the CO
2
is very dry. From an operator point of
view, flexibility on purity is preferred for best HSE and lowest cost for every CCS project. This means that every single CCS
project will have its own specifications settled on knowledge-based optimization to local boundary conditions (CO
2
source,
climate, regulations, geography etc), fulfilling regulations like OSPAR.
Water has received first priority in the CO
2
IT IS project, since it is the most common impurity. The Vapour-Liquid-Hydrate
equilibria of CO
2
-H
2
O mixtures are known to be complex, which can be seen from the 3D PTz-diagram in Diamond [8]. For
CCS chains the behaviour at low water concentration (<3000 ppm) is of interest. Most available data can be found in Song [9]
and Austegaard [10]. A gap in available experimental data has been identified for the VLH-equilibria, especially at temperatures
below -20 C. As shown in Figure 1 and Figure 2 on depressurization, such temperatures are attainable in transient operations.
Hence, depending on the relevant impurity concentrations, detailed knowledge of the VLHE-behaviour may be important for
setting safe and cost effective H
2
O specifications for each CCS project. Considering the available literature data, a set of 46
equilibrium measurements of CO
2
and water was chosen. The experiments were executed by Hydrafact [11]. The chosen
pressures and temperatures duplicate some of the earlier measurements and expand to cold regions around the boiling line of
pure CO
2
.
4.2. Model
A general phase equilibrium model based on the uniformity of component fugacities in all phases has been used to model the
phase behaviour of the carbon dioxide water system. This model is implemented in Hydrafacts HWHyd 2.2 which has been used
in this work. In summary, the statistical thermodynamics model uses the Valderrama modification of the Patel and Teja equation
of state (VPT-EoS) (Valderrama [12]) and non-density-dependent mixing rules (NDD) [13] for fugacity calculations in all fluid
phases. The hydrate phase is modelled using the solid solution theory of van der Waals and Platteeuw [14], as developed by
Parrish and Prausnitz [15]. The equation recommended by Holder et al. [16] is used to calculate the heat capacity difference
between the empty hydrate lattice and pure liquid water. The Kihara model for spherical molecules is applied to calculate the
potential function for compounds forming hydrate phases [17].
A little studied VLHE-behaviour was found when decreasing the H
2
O concentration in CO
2
. The most commonly published
hydrate line in a PT-diagram is the one with excess water. It has a VHL
W
-part and L
CO2
HL
W
-part meeting at the boiling line of
Figure 6 Qualitative PT-diagram of VLH equilibria at one low H2O concentration in CO2 including hydrate equilibrium line with excess
water and boiling line of pure CO2
Temperature
P
r
e
s
s
u
r
e
V
L

lin
e

p
u
r
e

C
O
2
V
V+H
L+H
L
H
y
d
r
a
t
e
l
i
n
e

a
t

e
x
c
e
s
H
2
O
Temperature
P
r
e
s
s
u
r
e
V
L

lin
e

p
u
r
e

C
O
2
V
V+H
L+H
L
H
y
d
r
a
t
e
l
i
n
e

a
t

e
x
c
e
s
H
2
O
3012 G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015
6 Author name / Energy Procedia 00 (2010) 000000
pure CO
2
, which is actually the upper quadruple point when water is present. Figure 5 shows a simplified qualitative description.
The hydrate phase boundary in the presence of free water is the right most vertical dashed line. The boiling line of pure CO
2
is
given in the most horizontal dashed line. Below a certain water concentration the models showed that these lines split due to
different H
2
O-solubility in CO
2
-rich vapour phase and liquid phase CO
2
. One such equilibrium line is shown in a PT-diagram in
Figure 6 at one low H
2
O concentrations. This means for example that at certain total H
2
O concentrations and pressures the phase
transitions VoVHoLoLH can occur during temperature decrease (shown with an arrow in Figure 6). This behaviour was
previously established by Chapoy et al [16]. The aim of this work was to execute a more extensive experimental verification of
this behaviour, focusing on temperatures below -20 C.
4.3. Experiments
The equipment used by Hydrafact comprised an equilibrium cell and a set-up for measuring the water content of equilibrated
fluids passed from the cell. A schematic of the rig and a more detailed description of the method is shown in Chapoy et al [16].
The equilibrium cell is a 300 ml titanium piston vessel rated to 690 bar. The cell is surrounded by a jacket connected to a
circulator with temperature control, keeping the temperature of the fluid within 0.1 C of the set-point. It can operate at
temperatures between -90 and 100 C. The cell temperature is measured using a PRT (Platinum Resistance Thermometer)
located in the jacket. The moisture content set-up comprises a heated line, a chilled mirror hygrometer and a flow meter. The
hygrometer is an EdgeTech DewMaster that measures the dew/frost point at temperatures ranging from -75 to 100 C, at
pressures up to 20 bar, with a resolution of 0.1 C and a stated accuracy of 0.2 C.
At the start of the test approximately 5 ml of de-ionised water was placed in a cup shaped depression in the top of the piston.
The cell was closed and the temperature reduced to -10 C and evacuated before injecting CO
2
. Cell temperature and pressure
were then adjusted to achieve the desired test conditions. The cell temperatures were cycled between temperatures above and
below the set point for at least 20 hours. A series of water content measurements were conducted during a period spanning
several days. The water content was found to be stable, indicating that equilibrium had been established. During the
measurements, the top cell valve was opened to fill the section of heated line up to the valve prior to the hygrometer. At the same
time nitrogen was introduced into the base of the cell in order to maintain a constant pressure. Following this the valve prior to
the hygrometer was opened sufficiently to achieve a flow rate through the hygrometer between 0.5 and 1 L/min. The water
content reading from the hygrometer was then monitored until it was stable for at least 10 minutes. This was taken as the
moisture content of the equilibrated fluid passing from the cell.
4.4. Results and discussion
The water content in equilibrium with hydrates was measured in the temperature range of -50 to +5
o
C and pressure range of
0.7-13 MPa, covering the range of conditions that can be encountered during depressurization of CO
2
. In general, the predicted
results are in good agreement with the measured data, with an average absolute deviation of 8.12%. The regression coefficient
(R
2
) between measured and modelled was 0.9973 (with intercept = 0), see Figure 7. It was concluded that this was accurate
enough and that the parameters in the HWHyd 2.2 model did not need to be retuned. Thus means that the qualitative VLHE
behaviour in Figure 6 is confirmed and can be quantified. However, it is known that other components like CH
4
[9] may have a
significant effect on the H
2
O solubility in CO
2
. The results in form of measured concentration and the temperatures and pressures
are considered to be a commercial advantage of Statoil and SINTEF.
R
2
= 0.9973
0
200
400
600
800
1000
1200
1400
1600
1800
2000
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Modelled water solubility (ppmv)
M
e
a
s
u
r
e
d

w
a
t
e
r

s
o
l
u
b
i
l
i
t
y

(
p
p
m
v
)
Figure 7 Modelled vs measured H2O solubility in CO2
G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015 3013
Author name / Energy Procedia 00 (2010) 000000 7
4.5. Other discussions
The discussion on impurities in CO
2
has started relatively recently. Several other aspects related to impurities that have not
been discussed in depth and may be topics for more detailed studies are:
x It is stated that H
2
O concentration should be low enough to avoid formation of free/liquid water [6]. However, H
2
O
concentration limit is closely related to the material philosophy. If stainless steel is the material of choice formation of free
water is not a big risk. Sleipner is a good example where the margin to free water formation is small and corrosion resistant
materials are used. Snhvit has a large margin to free water and carbon steel is used. In the future other combinations of H
2
O
concentration and material may well be used.
x O
2
may give geochemical and geo-biological reactions in the storage reservoir. The oil and gas industry has ample experience
with O
2
injection and air injection (see e.g. [19]), in situ combustion, fire flooding and produced water injection with
dissolved O
2
(see e.g. [20]). This shows that experience exists with a large span of O
2
concentrations potentially useful for
CCS. More experimental R&D is needed for quantifying the specific effects for CO
2
storage and EOR.
x The main reason for controlling the amount of so-called non-condensables (O
2
, N
2
, Ar, H
2
) above 4-5% is reducing the
minimum miscibility pressure in EOR. Moreover, less non-condensables reduce compressor costs. But these arguments
should not be used to exclude CCS chains up front where transporting more non-condensables than 4-5% may be safer and/or
cheaper.
5. Conclusions
Progress has been made on modelling and experimental verification on depressurization, heat transfer and impurities in CO
2

pipelines.
A newly developed CO
2
module in the flow engineering software OLGA from SPT Group has been used to model
experimental results on depressurization, which is a two-phase transient flow with evaporation and Joule-Thomson cooling. A
visual comparison of model and experiments in PT-diagrams showed that the new CO
2
module in OLGA captures the
depressurization behaviour reasonably well, but still has room for improvement.
Experiments have been conducted with a piece of the Snhvit CO
2
pipeline suspended in a temperature controlled insulated
box with multiple temperature sensors. Even though the experiments are still at an early stage, the initial results reveal a stable
operation of the test facility and are in good agreement with results found in literature. The overall heat transfer coefficient at the
experimental conditions has been calculated to be 45.0 W/m
2
K, where the outer heat transfer coefficient was 162.3 W/m
2
K, and
the inner heat transfer coefficient 166.2 W/m
2
K. However, to minimize uncertainty, a more thorough calibration of the test
facility in terms of determining additional heat leakage into the CO
2
besides at the pipeline needs to be conducted. In future
experiments, the effect of ice formation and the effect of filling the container around the pipeline with other substances such as
clay, soil, and rock will be investigated. Future experiments will also investigate the effect of having a larger thermal gradient
across the pipe wall, which will possibly change the governing heat transfer mechanisms compared to the current experimental
set-up.
New experimental measurements for the water content of CO
2
in equilibrium with hydrates have been successfully measured
at a wider range of conditions than previously reported using a high pressure equilibrium cell combined with a chilled mirror
hygrometer. The experimental data from this work are in acceptable agreement with predictions made using HWHyd 2.2 and
based upon this outcome it was not considered necessary to regress the model parameters to the experimental values as discussed
above. Hence, the model can be used for setting hydrate formation limited H
2
O specifications in transport of pure CO
2
.
6. Acknowledgements
The Research Council of Norway and its CLIMIT program are acknowledged for their financial contribution to the CO
2
IT IS
project. SPT Group is acknowledged for the execution of the study with the new CO
2
module in OLGA. Hydrafact is
acknowledged for modelling and performing experiments on H
2
O solubility in CO
2
. Both were valuable suppliers to the CO
2
IT
IS project.
7. References
-[1] De Koeijer GM, Borch JH, Jakobsen J, Drescher M, Experiments and modeling of two-phase transient flow during CO
2

pipeline depressurization, Energy Procedia, 2009;1;1:1683-1689
-[2] Hansen H, Eiken O, Aasum TO, The path of a carbon dioxide molecule from a gas-condensate reservoir, through the amine
plant and back down into the subsurface for storage. Case study: The Sleipner area, South Viking Graben, Norwegian North Sea,
SPE 96742-MS, 2005
-[3] SPT Group website, http://www.sptgroup.com/en/Company/News/OLGA-as-first-commercially-available-transient-
simulator-for-CO
2
-transport/ (accessed August 2010)
-[4] Munkejord ST, Jakobsen JP, Austegard A, Mlnvik MJ, Thermo- and fluid-dynamical modelling of two-phase multi-
component carbon dioxide mixtures, International Journal of Greenhouse Gas Control, 2010;4:589-596
3014 G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015
8 Author name / Energy Procedia 00 (2010) 000000
-[5] Cenge YA, Heat and Mass Transfer A pratical approach, Third Edition, 2006
-[6] De Visser E, Hendriks C, Barrio M, Mlnvik MJ, De Koeijer G, Liljemark S, Le Gallo Y, Dynamis CO
2
quality
recommendations, International Journal of Greenhouse Gas Control, 2008;2;4:478-484
-[7] OSPAR commission, OSPAR Decision 2007/2 on the Storage of Carbon Dioxide Streams in Geological Formations, 2007,
http://www.ospar.org/documents/DBASE/DECRECS/Decisions/07-02e_Decision%20CO2%20storage.doc (accessed August
2010)
-[8] Diamond L, Review of the systematics of CO
2
-H
2
O fluid inclusions. Lithos 2001;55:69-99
-[9] Song KY, Kobayashi R.. Water content of CO
2
in equilibrium with liquid water and/or hydrates. SPE Form. Eval.
1987;2:500508.
-[10] Austegard A, Solbraa E, De Koeijer G, Mlnvik MJ, Thermodynamic Models for Calculating Mutual Solubilities in H
2
O
CO
2
CH
4
Mixtures, Chemical Engineering Research and Design, 2006;84;9:781-794
- [11] Hydrafact company website, http://www.hydrafact.com/ (accessed August 2010)
-[12] Valderrama JO, A generalized Patel-Teja equation of state for polar and non-polar fluids and their mixtures. J. Chem.
Engng Japan 1990;23:87-91
-[13] Avlonitis, D, Danesh, A, Todd, AC, Prediction of VL and VLL Equilibria of Mixtures Containing Petroleum Reservoir
Fluids and Methanol with a Cubic EOS, J. Fluid Phase Equilibria, 1994;94:181-216
-[14] Van der Waals JH, Platteeuw JC, Clathrate solutions. Adv. Chem. Phys. 1959;2:2-57.
-[15] Parrish WR, Prausnitz JM, Dissociation pressures of gas hydrate formed by gas mixture, Ind. Eng. Chem. Process. Des.
Develop. 1972;11:26-34
-[16] Holder GD, Corbin G, Papadopoulos KD, Thermodynamic and molecular properties of gas hydrate from mixtures
containing methane, argon and krypton, Ind. Eng. Chem. Fundam. 1980;19:282-286.
-[17] Kihara, T., Virial Coefficient and Models of Molecules in Gases, Reviews of Modern Physics, 1953, 25(4), 831
-[18] Chapoy A, Burgass R, Tohidi B, Austell JM, Eickhoff C, Effect of Common Impurities on the Phase Behaviour of Carbon
Dioxide Rich Systems: Minimizing the Risk of Hydrate Formation and Two-Phase Flow, SPE 123778, 2009
-[19] Stokka S, Oesthus A, Frangeul J, Evaluation of Air Injection as an IOR Method for the Giant Ekofisk Chalk Field
SPE97481, 2005
-[20] Scoppio L, Nice PI, Material Selection for Turnaround Wells - An Evaluation of the Impact upon Downhole Materials
when Mixing Produced Water and Sea Water, SPE08089, 2008
G. de Koeijer et al. / Energy Procedia 4 (2011) 30083015 3015