archives

of thermodynamics
Vol. 35(2014), No. 1, 117–140
DOI: 10.2478/aoter-2014-0008

Analysis of pipeline transportation systems

for carbon dioxide sequestration

ANDRZEJ WITKOWSKI∗
MIROSŁAW MAJKUT
SEBASTIAN RULIK

Silesian University of Technology, Institute of Power Engineering and

Turbomachinery, Konarskiego 18, 44-100 Gliwice, Poland

Abstract A commercially available ASPEN PLUS simulation using a

pipe model was employed to determine the maximum safe pipeline distances
to subsequent booster stations as a function of carbon dioxide (CO2 ) inlet
pressure, ambient temperature and ground level heat flux parameters under
three conditions: isothermal, adiabatic and with account of heat transfer.
In the paper, the CO2 working area was assumed to be either in the liq-
uid or in the supercritical state and results for these two states were com-
pared. The following power station data were used: a 900 MW pulverized
coal-fired power plant with 90% of CO2 recovered (156.43 kg/s) and the
monothanolamine absorption method for separating CO2 from flue gases.
The results show that a subcooled liquid transport maximizes energy effi-
ciency and minimizes the cost of CO2 transport over long distances under
isothermal, adiabatic and heat transfer conditions. After CO2 is compressed
and boosted to above 9 MPa, its temperature is usually higher than ambient
temperature. The thermal insulation layer slows down the CO2 temperature
decrease process, increasing the pressure drop in the pipeline. Therefore in
Poland, considering the atmospheric conditions, the thermal insulation layer
should not be laid on the external surface of the pipeline.

Keywords: Carbon dioxide; Dense phase; Pipeline transportation; Energy efficiency;

Thermal insulation

∗
Corresponding Author. E-mail: [Link]@[Link]
118 A. Witkowski, M. Majkut and S. Rulik

1 Introduction

Of several approaches to carbon dioxide (CO2 ) transport, pipeline trans-

portation is the most economical one to transport large amounts of CO2
over long distances. Zhang et al. [18] studied the pressure drop behavior of
supercritical CO2 as well as the CO2 dense phase along the pipeline. Their
results show that the pressure along the pipeline keeps dropping until CO2
evaporates and the pipeline may eventually be blocked. This means that
there is a maximum safe transport distance. If there is a need to transport
CO2 farther than over this maximum distance, boosting pump stations are
needed along the pipeline. The CO2 can be transported over long distances
in two states: either as a supercritical fluid or as a subcooled liquid. Gas-
phase transport is disadvantageous due to the low density and high pressure
drops. Generally, CO2 transportation in the subcooled liquid state has some
advantages over the supercritical state transport, most importantly because
of the lower compressibility and higher density of the liquid within the pres-
sure range considered here, which permits smaller pipe sizes and generates
lower pressure losses. In the present work, the state of existence of CO2 was
assumed to be either in the liquid or in the supercritical state. The aim of
this paper is to analyze CO2 transport via a pipeline from the capture site
to the disposal site under isothermal, adiabatic and heat transfer conditions.
According to the conclusions reached at [18], the currently available versions
of the equations of state (EOS) to predict properties of supercritical CO2
under conditions close to the critical point are not reliable enough to design
a precise compression system. In the present work, in order to compare the
results, two Lee-Kesler-Plocker (LKP) and Peng-Robinson-Boston-Mathias
(PRBM) equations of state were used. In the following discussion, the same
power station data as those reported in [4, 9] were assumed. ASPEN PLUS
V 7.0 [2], a commercially available design process simulator with an exten-
sive thermodynamic library, was used to simulate the CO2 transportation
process. The calculations were carried out assuming pure carbon dioxide.
However, captured CO2 does include a series of impurities depending on
the capture technology which affects the CO2 phase diagram. Moreover, in
real pipeline transportation of CO2 , the pipeline may go through changes in
elevation, which has not been considered until now. But it has to be noted
that a change in elevation has a great impact on the hydrodynamic perfor-
mance. This work can be used as reference for the design and construction
of CO2 pipelines in Poland in the future.
Analysis of pipeline transportation systems. . . 119

2 Physical properties of carbon dioxide

The properties of carbon dioxide are considerably different from those of
other fluids commonly transported by pipeline. Thus, in pipeline design, it
is necessary to use accurate representations of the phase behavior, density,
and viscosity of pure CO2 and of CO2 containing impurities. For the mul-
tiphase flow, models or correlations are currently available for predicting
the pressure profile in a well. An important characteristic of CO2 that dis-
tinguishes it from other substances typically bulk-transported in pipelines
is its low critical temperature, namely 31.1 o C. Technically, CO2 can be
transported through pipelines as a gas, as a supercritical fluid or as a sub-
cooled liquid, depending on the pressure and temperature conditions in the
pipeline system (Fig. 1). Since CO2 is a highly corrosive medium, the water
content must be reduced to less than 60% of the saturation state [6]. In
the case of intercooled compression, a portion of the moisture is removed
through condensation. However, it is still necessary to provide a further
drying stage after the final compressor stage.

Figure 1. A phase diagram for CO2 .

The method of controlling the system temperature and pressure under

a particular condition directly determines significant aspects of the design of
the system processes, the pressure losses, the mechanical structure and, ul-
timately, the energy and cost efficiency. The results presented in this paper
120 A. Witkowski, M. Majkut and S. Rulik

are based on the physical properties of CO2 containing impurities and were
obtained using real gas equations of state (EOS) with the Lee and Kesler
equation modified by Plocker, Knapp, and Prausnitz (the LKP equation of
state) [2,8] and the Peng-Robinson equation of state with Boston-Mathias
modifications (PRBM) [2,14]. All these equations are included in the AS-
PEN PLUS V 0.7 design process simulator [2].
Moreover, Fig. 2 shows that CO2 compressibility is nonlinear in the
range of pressures common for pipeline transport and is highly sensitive to
any impurities as predicted by the Peng-Robinson equation of state [10].
Thus, it is necessary to use accurate representations of the phase behavior,
density, and viscosity of CO2 while designing a pipeline. To reduce difficul-
ties in design and operation, it is generally recommended that the pressure
in a CO2 pipeline should be higher than 8.6–10 MPa because the abrupt
changes in CO2 compressibility can then be avoided across the range of
temperatures that may be encountered in the pipeline system [5,10].

Figure 2. Nonlinear compressibility of CO2 in the range of pressures common for pipeline
transport as predicted by the Peng-Robinson equation of state [10].
Analysis of pipeline transportation systems. . . 121

3 Pipeline configuration
A complete CO2 capture and sequestration (CCS) system requires safe, reli-
able and cost-efficient solutions for transmission of CO2 from the capturing
facility to the permanent storage site. The CO2 pipeline facilities from the
power plant flue stack, through the separation, compression, transportation
and pumping systems to the injection well head are shown schematically
in Fig. 3. For sequestration purposes, CO2 is generally injected to depths
exceeding 1000 m [11]. At greater depths, CO2 increases in density and
below 800 m it becomes a supercritical fluid. A large compressing system
is required to compress the source CO2 under nearly atmospheric pressure
conditions. Figure 3 provides an example of an eight-stage integrally geared
compressor system, which was selected in [16] as the most available, reliable
and efficient compression technology and injection configuration. Compres-
sor stations in a CO2 pipeline system can be subdivided into two classes:
stations positioned at the pipeline inlet, and booster stations located along
the pipeline to compensate for the pressure drop due to friction and eleva-
tion losses.

4 Properties of CO2 in pipeline transport

After CO2 is captured at emission sources, it has to be transported to the
storage site. At present, pipelines are the most common means of trans-
portation of large quantities of the gas. Two states can be used to transport
CO2 over long distances: either as a supercritical fluid or as a subcooled
liquid. The physical condition which is suitable for pipeline transportation
in terms of pressure and temperature, is the supercritical/dense phase. This
phase is preferable due to the fact that it is relatively stable compared to
the liquid state, which minimizes cavitation problems in the system compo-
nents such as booster stations and pumps. Since the critical point for CO2
is 31.1 o C and 7.38 MPa, the system pressure higher than 7.5 MPa results
in transportation at supercritical parameters, as long as the temperature
remains above 31.1 o C. According to Fig. 1, if pressure drops below the
critical pressure, the phase may be liquid or gaseous, or both, depending
on the local temperature. Gas-phase transport is disadvantageous due to
the low density and the high pressure drop. Since critical temperature is
higher than normal ground temperature, either thermal insulation is needed
to keep CO2 in the supercritical state or CO2 needs to be heated at every
122 A. Witkowski, M. Majkut and S. Rulik

Figure 3. Diagram of a CO2 pipeline transportation configuration (processing, compres-

sion, and injection). B-boiler, HP-high pressure part, LP-low pressure part,
IP-intermediate pressure part of the steam turbine, LP1, LP2, LP3, LP4 low
pressure feed water heater, HP1, HP2, HP3-high pressure feed water heater,
DSH-desuperheater.
Analysis of pipeline transportation systems. . . 123

certain distance. Otherwise, it will transform into the liquid state. Mean-
while, some authors propose that transporting liquid CO2 at a relatively
low temperature is preferred in terms of a reduction in the pressure drop
along the pipeline [19]. In the present paper the CO2 working area is as-
sumed to be either in the liquid or in the supercritical state and the results
of these two states are compared.
For subcooled liquid CO2 transportation, a facility cooling CO2 to 15 o C
or less is needed so that it can be kept below its critical temperature down
the line. Conceptually, refrigeration could be added along the pipeline using
CO2 as the working medium, but this will obviously increase the capital and
operating costs as well as reduce the overall energy efficiency. Generally,
CO2 transportation in the subcooled liquid state has some advantages over
the supercritical state transport, most importantly because of the liquid
lower compressibility and higher density within the pressure range consid-
ered here, which permits smaller pipe sizes and generates lower pressure
losses. To reduce difficulties in design and operation, it is generally recom-
mended that the pipeline should operate at pressures higher than 8.6 MPa.
Then, the abrupt changes in CO2 compressibility and specific heat can be
avoided across the range of temperatures (Fig. 2) that may be encountered
in the pipeline system [5,10].
Most new pipelines are laid underground, despite the higher initial costs,
for environmental, security and safety reasons. Underground temperatures
are much more stable than surface temperatures. Hence the operating tem-
peratures of CO2 pipelines are generally dictated by the temperature of the
surrounding soil.
The pressure keeps dropping along the pipeline until CO2 evaporates and
the pipeline may eventually be blocked. This means that there is a max-
imum safe transport distance. If there is a need to transport CO2 farther
than over this maximum distance, boosting pump station are required along
the pipeline.
A pipeline segment is defined as a length of pipeline for which the inlet
pressure and the minimum outlet pressure values are specified, e.g., a length
of pipeline between two compressor stations.

5 Pressure loss correlation

There are a number of fluid correlations, derived empirically, that account
for the hydrostatic and frictional fluid losses in a wellbore under varied flow
124 A. Witkowski, M. Majkut and S. Rulik

conditions. The presence of a multiphase flow greatly complicates pressure

drop calculations. Many flow models or correlations are currently available
for the multiphase flow making it possible to predict the pressure profile
in a well. The reasonably good performance of multiphase flow models,
within the context of [10], is considered to be burdened with a relative
error, between the measured and predicted values of the pressure profile,
which is less than or equal to 20%. The Beggs and Brill method [3] is
suitable for the multiphase flow and for the horizontal or vertical flows as
well as intermediate cases. It also takes account of the general mechanical
energy balance and the average in situ density to calculate the pressure
gradient.

6 Modeling CO2 transport by pipeline for the case

study
In the following discussion, the same power station data as those reported
in [4,9] are used. They include: a 900 MW pulverized hard coal-fired power
plant with 90% of CO2 recovered and the MEA (monoethanolamine) amine
based absorption method for separating CO2 from the flue gases. The
transport mass flow rate was assumed at 4.93 Mt/year (156.43 kg/s) and
the distance – at 400 km. The pipeline inlet pressure was assumed to be
153 MPa, and the inlet temperature – 35 o C at the supercritical state.
Once the CO2 pressure drops to below 9 MPa, a boosting station would be
installed to increase the pressure back to 153 MPa. The pipeline need not
be insulated when advantage can be taken of cold ground conditions, which
helps to maintain liquid conditions. The typically long length of a CO2
pipeline segment coupled with the lack of insulation on buried pipelines can
be treated as an isothermal system, where CO2 features the temperature
of the earth surrounding the pipeline. Considering that the flow velocity of
CO2 inside the pipeline is usually between 1 and 2 m/s, an internal diameter
of 0.45 m was assumed based on a trial and error calculation.
The pipe wall thickness, t, in meters is given as [13]

pmop d2
t= , (1)
2S E F
in which pmop is the maximum operating pressure of the pipeline (Pa), d2
is the outside pipe diameter (m), S is the specified yield stress for the pipe
material (Pa), E is the longitudinal joint factor, and F is the design factor.
Analysis of pipeline transportation systems. . . 125

For the purpose of estimating the pipe wall in this case study, the maximum
operating pressure is assumed at 153 MPa, the longitudinal joint factor is
1.0, and the design factor is 0.72 (US Code of Federal Regulations – CFR,
[13]). The minimum yield stress is specified as 483 MPa, which corresponds
to the API (American Petroleum Institute) for 5L X-70 steel line pipe [1].
Alternatively, a thermal insulation layer was assumed on the pipeline exter-
nal surface. The flow process of CO2 along the pipeline is mainly influenced
by three factors: friction forces, heat exchange through the pipe wall be-
tween the soil and the thermal insulation layer, and the change in elevation.
Based on the data given above, thickness of the pipe wall, t, was calculated
to be 10 mm. Since the critical temperature of CO2 is predominately higher
than the normal temperature of the soil or the ambient temperature, after
CO2 is compressed and boosted to above 10 MPa, the thermal insulation
will slow down the CO2 temperature decrease process, increasing the pres-
sure drop in the pipeline. Therefore, the thermal insulation layer should
not be laid on the pipeline external surface in Poland. In the present work,
the CO2 working area was assumed to be either in the liquid or in the
supercritical state.

7 Results and discussion

7.1 Comparison between adiabatic and isothermal
transmission
7.1.1 Maximum safe transport distance
For fluid flow through pipelines, the transmission process can be generally
found between isothermal and adiabatic conditions. For typical lengths of
the underground CO2 pipeline segment with no insulation, the segment may
be treated as an isothermal system, where CO2 is at the temperature of the
soil surrounding the pipeline. If CO2 enters the pipeline in the supercritical
state, it goes into the gaseous state at some point along the pipeline because
of the pressure drop. At a constant diameter pipeline, the CO2 velocity
increases along the pipeline, causing a very big pressure drop or ‘choking
conditions’ at a certain distance.
Figure 4 shows a comparison of the pressure drop and the fluid density
change resulting from it depending on the distance for different initial CO2
temperature values, for adiabatic and isothermal flow conditions. There is
not much difference between the adiabatic and isothermal conditions when
126 A. Witkowski, M. Majkut and S. Rulik

a)

b)

Figure 4. Comparison of the pressure drop (a) and density changes (b) along the pipeline
for adiabatic and isothermal conditions at different inlet CO2 temperatures.

CO2 at the inlet is in the liquid state because the liquid is incompressible. It
also shows that a negligible difference exists between adiabatic and isother-
mal transmission when the CO2 inlet state is liquid at 15 o C. The difference
becomes significant when CO2 reaches the supercritical fluid state (35 o C).
Under isothermal conditions and with an initial subcooled liquid, there is
a sufficient heat transfer to the ground, and, therefore, the transmission dis-
tance is shortened. In adiabatic conditions, there is no heat transfer across
the pipe wall and the transmission distance is longer. As it was accepted
earlier, the lowest transport pressure of the CO2 pipeline is 9 MPa. In this
Analysis of pipeline transportation systems. . . 127

case, the maximum safe transport distance depends on the thermal condi-
tions, changing from 469 km at the CO2 subcooled temperature of 0 o C to
339 km at the supercritical temperature of 35 o C, and under isothermal con-
ditions. For the adiabatic case, there is a discontinuity in the density profile.
This discontinuity corresponds with the saturation state when CO2 changes
density quickly, e.g., from the subcooled liquid state to a two-phase state.
Even for supercritical fluids where there is no phase change, the density
variation has a very nonlinear region. With initial temperatures above the
supercritical point, the CO2 density changes abruptly within the pipeline
once the temperature reaches the saturation point, and reaches two-phase
flow conditions. Moreover, Fig. 4 also indicates that if the inlet temperature
is higher than critical, CO2 repressurization is needed after a much shorter
distance, whether adiabatic or isothermal conditions prevail.

7.1.2 Influence of the pipe diameter

The pipeline diameter is incorporated as a crucial parameter in the cost
estimation of CO2 pipeline transport. Figure 5 shows the maximum safe
transport distance of CO2 , (a) and the influence of different internal pipe
diameters under adiabatic and isothermal conditions on CO2 density (b),
and velocity (c). Actually, as it was assumed earlier, the lowest transport
pressure is 9 MPa. In this case, the maximum safe transport distance
is 110 km, 220 km, and 405 km at the pipe inner diameters of 0.35 m,
0.4 m, and 0.45 m, respectively. It can be seen from Fig. 5 that the inner
diameter of the pipeline has a strong influence on the number of boosting
pump stations in the long distance transport and, consequently, on the total
pipeline transportation costs.

7.2 Energy balance with surroundings

7.2.1 General remarks
Analysis of CO2 transport by pipeline must take a most realistic account of
the influence of the ambient temperature on heat exchange between carbon
dioxide in the pipe and the surroundings along the pipeline. Some studies
proposed that CO2 should be transported in the supercritical state. For
a subcooled liquid transmission and a temperature higher than critical to
minimize heat gains, the pipeline may be placed underground and/or in-
sulated. An underground and insulated pipeline reduces the pressure drop
and, therefore, the energy losses in the system. However, this results in
128 A. Witkowski, M. Majkut and S. Rulik

a)

b)

c)

Figure 5. Maximum safe transport distances (a), density (b) and velocity (c) with differ-
ent internal pipe diameters in adiabatic and isothermal conditions.
Analysis of pipeline transportation systems. . . 129

the increase in capital and maintenance costs. The pipeline needs not be
insulated when advantage can be taken of cold ground conditions, which
maintain liquid conditions. In the present work, the CO2 working area was
assumed to be either in the liquid or in the supercritical state. In the latter
case, thermal insulation was assumed on the external surface of the pipe
(Fig. 6). Engineering experience indicates that long distance pipelines are
usually placed underground at a depth of 1.2–1.5 m. It is also found that the
annual lowest and highest soil temperatures at a depth of 1.5 m in Poland
are between 5 and 16 o C (Fig. 7).

Figure 6. Cross section of the underground and insulated pipeline. r1 – internal radius
of pipe, r2 – external radius of pipe, re – external wall radius of the thermal
insulation layer, t – pipewall thickness, tsurf ace – ground surface temperature,
tsoil – soil temperature.

Two cases – with the lowest and the highest soil temperatures – are
considered in the pipeline design to ensure that the pipeline can operate
well over the whole year.
A two-dimensional heat conduction formula [7] can be used to calculate
the overall heat transfer coefficient between the ground and the CO2 in the
pipeline (Fig. 6):

1
k= r1 r2 r1 re r1 r1 1 , (2)
λpw ln r1 + λti ln r2 + λsoil ln 2z
re + z αag

where heat conductivities of the pipe wall and of the insulation layer mate-
rials are λpw = 25 W/mK and λti = 0.058 W/mK, respectively.
The thermal conductivity of the soil is assumed at λsoil = 1.21 W/(mK),
and the distance between the ground surface and the pipe center is 1.225 m.
The air convection heat transfer coefficient is αg = 5 W/m2 K. According
130 A. Witkowski, M. Majkut and S. Rulik

Figure 7. Annual soil temperatures at different depths and ambient temperatures, tsoil
– soil temperature.

to [18] and [19], the thermal resistance of the convective thermal transfer
between CO2 and the inner pipe wall is much smaller than that between the
pipe wall and the heat insulation layer, so it is assumed that the tempera-
ture of the inner pipe wall is equal to the temperature of CO2 on the same
cross section. The value of the overall heat transfer coefficient k between
the ground and the CO2 calculated from Eq. (2) for a pipeline with a 0.05 m
and 0.03 m heat insulation layer is 0.7387 and 0.912 W/m2 K, respectively,
and for a pipeline without insulation – 2.11 W/m2 K. Resistance to the heat
transfer from the tubing and casing is ignored, as the conductivity of the
steel used in the tubing and casing is at least an order of magnitude larger
than any other conductivity in the system.

7.2.2 Choking conditions

An increase in the pressure drop means higher operating costs and possibly
the need to introduce compressor stations. The pressure drop along the
pipeline is dependent on the flow velocity, ambient temperature, as well as
on geometric characteristics of the pipeline such as length, elevation changes
etc. The pressure drop and temperature changes along the pipeline reduce
the CO2 density and increase velocity, which, in turn, increases the pressure
drop and ultimately leads to a choking condition at a certain distance. The
maximum safe pipeline length to prevent choking is preliminarily selected
at a value about 10% smaller than the choking point [18]. The dependence
Analysis of pipeline transportation systems. . . 131

Figure 8. Dependence of the safe pipeline length and of the choking point on CO2 pipe
inlet temperatures (20 o C, 30 o C, 35 o C, 45 o C) at energy balance with sur-
roundings.

Figure 9. Dependence of the safe pipeline length and of the choking point on ambient
temperatures (0 o C, 10 o C, 20 o C, 30 o C).
132 A. Witkowski, M. Majkut and S. Rulik

of the safe distance on the pipeline inlet and ambient temperatures for the
reference case study is presented in Figs. 8 and 9, respectively. It can be
seen that both the CO2 temperature (Fig. 8) and the ambient temperature
(Fig. 9) have a significant impact on the pipeline maximum safe length.
In order to avoid the choking condition, recompression of CO2 becomes
necessary.

7.2.3 Influence of ambient temperature and the thermal

insulating layer
In order to understand the impact of the thermal insulating layer, the
pipeline operational parameters were calculated with and without the in-
sulating layer at different thickness and ambient temperature values. Fig-
ures 10 and 11 show four parameters of CO2 (temperature, pressure, density
and velocity) along the pipeline with and without insulation, at different
ambient temperatures calculated with the use of two different real gas equa-
tions of state: the LKP and the RPBM equations. In the simulation, the
inlet conditions for CO2 are fixed. It is transported to the injection site in
a straight line over flat ground. Figures 10a and 11a show that CO2 pressure
drops linearly along the pipeline. It can be seen that the maximum differ-
ence in the maximum safe transport distance up to assumed pressure drops
to below 9 MPa (about 91 km at the LKP calculations and 78.4 km at the
PRBM calculations) occurs for CO2 transmission at maximum differences
between ambient temperatures and without thermal insulation.
With initial CO2 temperatures above the supercritical point and with
calculations performed using the LKP equations, CO2 density changes ab-
ruptly within the pipeline once the temperature reaches the saturation
point, and a two-phase flow commences. It can be seen from Fig. 10 that
as the ambient temperature gets lower, the liquid phase flow appears at the
shorter distance. However, the safe flow distance gets considerably longer.
It can be seen that the pressure drop increases significantly with ambient
temperature, which confirms our estimation that low temperature is prefer-
able for pipeline transport. It can also be seen from Fig. 9 that as the
ambient temperature gets lower, the CO2 temperature drops more quickly
to a level lower than the critical point within a very short distance. Figures
12 and 13 show the maximum safe transport distance of the CO2 pipeline at
different ambient temperatures with and without thermal insulation layer
even more clearly. It can be seen that if the inlet temperature keeps con-
stant, the pressure drop decreases as the ambient temperature decreases
Analysis of pipeline transportation systems. . . 133

a)

Figure 10. Comparison of the pressure drop (a), the change in temperature (b), in density
(c), and in velocity (d) along a pipeline with and without thermal insulation
for the conditions of energy balance with surroundings at different ambient
temperatures tamb : 0o C, 15o C, 20o C, 30o , tCO2 – inlet CO2 temperature, k –
overall heat transfer coefficient; LKP equation of state.

since velocity then decreases too.

At lower ambient temperatures the pressure drop in the pipeline without
thermal insulation is lower than in the insulated pipeline. It can be seen
that the maximum difference in the maximum safe transport distance up to
assumed pressure drops to below 9 MPa (about 32.8 km both for the LKP
and PRBM calculations), for the two cases with and without insulation,
occurs for CO2 transmission at the ambient temperature of 0 o C. However,
this difference drops to nearly zero as the ambient temperature increases to
about 27 o C. Only at the ambient temperature of 30 o C is the safe trans-
port distance longer for the case with insulation if the LKP equations of
state are used; the distance obtained by means of the PRBM equations
of state is nearly the same. This verifies our estimation that a pipeline
without thermal insulation is preferable for pipeline transport in the Polish
climate. Generally, safe transportation distances calculated with the use of
the PRBM equation of state are longer than those calculated with the LKP
equation of state.
134 A. Witkowski, M. Majkut and S. Rulik

b)

c)

d)
Figure 10b-d (continued).
Analysis of pipeline transportation systems. . . 135

a)

b)

c)
Figure 11a-c. For caption see next page.
136 A. Witkowski, M. Majkut and S. Rulik

d)

Figure 11. Comparison of the pressure drop (a), the change in temperature (b), in density
(c), and in velocity (d) along a pipeline with and without thermal insulation
for the conditions of energy balance with surroundings at different ambient
temperatures: 0o C, 15o C, 20o C, 30o ; PRBM equation of state.

Figure 12. Maximum safe length of a pipeline with (k = 0.738) and without (k = 2.11) an
insulating layer; distance to subsequent booster station at different ambient
temperatures; LKP equation of state.

8 Conclusions
The paper presents the analysis of the influence of multiple factors, including
pipe diameter, pipeline inlet temperature, ambient temperature or thermal
Analysis of pipeline transportation systems. . . 137

Figure 13. Maximum safe length of a pipeline with and without an insulating layer; dis-
tance to subsequent booster station at different ambient temperatures; PRBM
equation of state.

insulation layer on the thermodynamic performance of the CO2 flow in the

pipeline, and proposes common guidelines for the design of CO2 transport
pipelines in terms of the pressure drop minimization, such as the choice of
a proper pipe diameter, the necessity to use or not the thermal insulation
layer on the external surface of the pipe in the Polish atmospheric conditions
and the reduction in the transport temperature. To mitigate difficulties in
design and operation, in this case CO2 pipelines are operated at a pressure
greater than 9 MPa or under dense phase conditions, where abrupt changes
in CO2 compressibility can be avoided across the range of temperatures
that may be encountered during the pipeline system operation. The results
show that the pressure along the pipeline keeps dropping until CO2 evapo-
rates and the pipeline may eventually be blocked. This means that there is
a maximum safe transport distance. If CO2 should be transported farther
than the maximum distance, boosting pump stations are needed along the
pipeline.
Booster stations are placed at discrete step intervals and their number
and the cost they involve may be minimized by a careful analysis. Three
transportation conditions – adiabatic, isothermal and with heat exchange
between CO2 and the surroundings along the pipeline – were considered.
In the analysis of adiabatic and isothermal energy losses, at the lowest final
138 A. Witkowski, M. Majkut and S. Rulik

transport pressure of 9 MPa and at the highest initial CO2 temperature

of 35 o C, the maximum safe transmission distance to a booster station
amounts to 365 km under adiabatic conditions, which is by up to 7.7%
longer compared to the distance under isothermal conditions (339 km).
The most realistic approach in the analysis of CO2 transport by pipeline
must take account of the influence of ambient temperature on the heat ex-
change between CO2 and the surroundings along the pipeline. An increase
in ambient temperature reduces CO2 density and increases the velocity
along the pipeline, which in turn increases the pressure drop and leads to
building up choking conditions. A bigger pressure drop means higher oper-
ating costs and possibly the need to introduce recompression stations. For
the purposes of this study, the maximum value of ambient temperature,
which in Poland may be as high as 30 o C, was assumed. This significantly
shortens the maximum distance at the inlet pressure of 15.3 MPa to 310 km,
compared to the maximum safe distance, which amounts to 403 km at the
ambient temperature of 0 o C. So, for a CO2 stream in the inlet supercritical
state (35 o C), the maximum safe transmission distance is 310 km, which is
by up to 9.3% shorter than the distance calculated for isothermal conditions.
To sum up, when designing the pipeline, the extreme case with the highest
ambient temperature should be considered to ensure that the pipeline can
work well all through the year. For a CO2 stream in the inlet supercritical
state (35 o C), the maximum safe transmission distance is 310 km.
Since CO2 temperature is usually higher than the soil temperature after
CO2 is compressed and boosted to above 15 MPa, the thermal insulation
layer will slow down the CO2 temperature decrease process, increasing the
pressure drop in the pipeline. Therefore in Poland, considering the atmo-
spheric conditions, the thermal insulation layer should not be laid on the
external surface of the pipeline.
The calculations were carried out assuming pure carbon dioxide. How-
ever, captured CO2 includes a series of impurities depending on the capture
technology which affects the CO2 phase diagram. Even small amounts of
impurities in CO2 change the location of the supercritical line so larger
margins of safety may have to be applied in CO2 pipeline design.

Acknowledgments The results presented in this paper were obtained

from research work cofinanced by the National Centre of Research and
Development in the framework of Contract SP/E/1/67484/10, Strategic
Research Programme (Advanced Technologies for Energy Generation: De-
Analysis of pipeline transportation systems. . . 139

velopment of a Technology for Highly Efficient Zero-Emission Coal-Fired

Power Units Integrated with CO2 Capture).

Received 3 January 2013

References
[1] American Petroleum Institute: Spec 5L-Specification for Line Pipe, 43rd Edn. Wash-
ington 2004.
[2] Aspen Plus, Version 7.0. Computer program, 2008.
[3] Beggs H.D., Brill J.P.: A Study of two-phase flow in inclined pipes. J. Petrol
Technol., 25(1973), 607–617.
[4] Chmielniak T., Łukowicz H.: Condensing power plant cycle – assessing possibil-
ities of improving its efficiency. Arch. Thermodyn. 31(2010), 3, 105–113.
[5] Farris C.B.: Unusual design factors for supercritical CO2 pipelines. Energy
Progress 3(1983), 3, 150–158.
[6] Gottlicher Gerold: The Energetics of Carbon Dioxide Capture in Power Plants.
VDI Verlag, Dusseldorf 1999.
[7] Incropera F.P., DeWitt D.P.: Introduction to heat transfer, 3rd Edn. John
Wiley & Sons, Inc., New York 1996.
[8] Ludtke K.H.: Process Centrifugal Compressors. Springer, 2004.
[9] Łukowicz H., Dykas S., Rulik S., Stępczyńska K.: Thermodynamic and
economic analysis of a 900 MW ultra-supercritical power unit. Arch. Thermodyn.
32(2011), 3, 231–244.
[10] McCoy S.T., Rubin E.S.: An engineering-economic model of pipeline transport of
CO2 with application to carbon capture and storage. Int. J. Greenhouse Gas Contr.
2(2008), 2, 219–229.
[11] Mohitpour M., Seevam P., Botros K.K., Rothwell B., Ennis C.: Pipeline
Transportation of Carbon Dioxide Containing Impurities. ASME, New York 2012.
[12] Polish Committee for Standardization: Thermal Insulation of Ducts, Fittings and
Installations, PN-B-02421, Jul. 2000.
[13] Transportation. Title 49 Code of Federal Regulations Pt. 195, 2005 Edn. 170–171.
[14] Rao B.: Multiphase Flow Models Range of Applicability. CTES, L.C, May 18, 1998.
[15] Wayne C., Edmister Byung Ik Lee: Applied Hydrocarbon Thermodynamics Vol.
1. Gulf Publishing Company, London Houston, London, Paris, Tokyo 1984.
[16] Witkowski A., Majkut M.: The impact of CO2 compression systems on the com-
pressor power required for a pulverized coal power plant in post-combustion carbon
sequestration. Arch. Mechanical Eng., LIX(2012), 3, 343–360.
[17] Witkowski A., Rusin A., Majkut M., Rulik S., Stolecka K.: Comprehensive
Analysis of the Pipeline Transportation Systems for CO2 Sequestration. Thermody-
namics and Safety Problems. In: Proc. 3rd Int. Conf. Contemporary Probl. Thermal
Eng. CPOTE 2012, 18-20 September, 2012, Gliwice.
140 A. Witkowski, M. Majkut and S. Rulik

[18] Zhang Z.X., Wang G.X. Massarotto P., Rudolph V.: Optimization of
pipeline transport for CO2 sequestration. Energ. Conver. Manage. 47(2006), 702–
715.
[19] Zhang D., Wang Z., Sun J., Zhang L., Zheng L.: Economic Evaluation of
CO2 pipeline transport in China. Energ. Conver. Manage. 55(2012), 127–135.