http://pubs.acs.

org/journal/acsodf Article

A New Model for Calculation of Arrest Toughness in the Fracture

Process of the Supercritical CO2 Pipeline
Qihui Hu, Nan Zhang,* Yuxing Li, Wuchang Wang, Jianlu Zhu, and Jiyu Gong

Cite This: ACS Omega 2021, 6, 16804−16815 Read Online

ACCESS Metrics & More Article Recommendations

ABSTRACT: A new model based on a decompression wave prediction model and

an improved BTC model has been developed to investigate the arrest toughness in
the fracture process of the supercritical CO2 pipeline. The comparison of the
decompression wave velocity and the fracture propagation velocity was carried out
to identify whether the pipe can prevent fracture propagation relying on its own
toughness. If not, the minimum Charpy V-notch energy and the minimum wall
thickness of steel pipes required for arrest fracture can be calculated using the
improved BTC model. The results show that the working conditions with an initial
pressure for the fracture of 11.7 MPa and a temperature of 323.15 K are the most
difficult conditions to stop the fracture. The minimum wall thickness calculated
only according to the strength design cannot meet the toughness requirements for
ductile fracture arrest in the most difficult conditions in some cases. Then, the
minimum wall thickness of the supercritical CO2 pipeline required for ductile
fracture arrest in these cases will be obtained. For instance, the minimum wall
thicknesses of X65, X70, and X80 steel pipes for fracture arrest with a pipe diameter of 610 mm at a design pressure of 13.2 MPa are
17.28, 14.58, and 12.81 mm, respectively, and when the pipe diameter is 1016 mm at a design pressure of 20.4 MPa, the minimum
wall thicknesses of X70 and X80 pipes can meet the requirements of arrest toughness. The model established in this study can
quickly and accurately calculate the minimum wall thickness and minimum Charpy energy required to stop fracture in the
supercritical CO2 pipeline, which is suitable for engineering applications. The findings of this study can help in better understanding
of the fracture process of supercritical CO2 pipelines.

■ INTRODUCTION
For a long time, fossil energy has been an important driving force
transported as a gas, liquid, dense phase, and supercritical phase.
Since the density of CO2 in the supercritical phase is comparable
for the world’s economic development. However, the to that of the liquid phase and the viscosity is comparable to that
combustion of fossil energy sources such as coal and oil of the gaseous phase (so, supercritical CO2 has large density and
produces large amounts of greenhouse gases, CO2. In recent low viscosity), the supercritical state is the best phase state for
decades, the emission of CO2-based greenhouse gases into the pipeline transportation of CO2 in terms of economic efficiency.3
atmosphere has become an increasing concern for the world. The critical temperature of CO2 is 304.13 K, and the critical
Currently, Carbon Capture, Utilization, and Storage (CCUS) pressure is 7.38 MPa.4 In order to ensure safe and stable
is the only technology that can significantly reduce CO2 operation, no phase change should be allowed during the CO2
emissions from electricity and industry and is one of the most pipeline transportation. The operating pressure of the super-
potential and effective solutions to the greenhouse effect in the critical CO2 pipeline needs to be greater than the critical
coming decades.1 Usually, the distance between the CO2 pressure, and the pipeline maintains high-pressure operation,
capture site and the storage or use site is long, and how to which may make the pipe wall material sensitive to defects and
transport the CO2 from the capture site to the target site safely prone to leakage or even fracture accidents. Leaking of CO2
and efficiently is a key aspect that restricts the development of accumulates in low-lying areas and may cause environmental
this technology.
Commonly used CO2 transportation methods include ship,
truck, rail, and pipeline transportation. At present, there is less Received: March 14, 2021
experience on large-scale ship transportation, and the cost of Accepted: June 7, 2021
high-pressure transport vessels is too high. When the trans- Published: June 23, 2021
portation distance exceeds 160 km and the transportation
volume exceeds 79,000 m3/d, pipeline transportation is more
economical than truck and trailer transportation.2 CO2 can be
© 2021 The Authors. Published by
American Chemical Society https://doi.org/10.1021/acsomega.1c01360
16804 ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

damage and casualties,5−7 so measures must be taken to inhibit the natural gas pipeline, and more in-depth understanding is
pipeline cracking and fracture propagation. needed for ductile fracture in CO2 pipelines. Currently, the
The criteria for determining whether a pipeline fracture is relationship between decompression behavior and pipeline
expanding are the velocity criterion and the energy criterion.8 rupture is still not very clear, and the previous models for
The energy criterion compares the magnitude of the fracture calculating the ductile fracture propagation of the CO2 pipeline
driving force generated by the energy release inside the during the decompression process are complex. In addition, to
supercritical CO2 with the resistance of the pipe toughness to the best of the authors’ knowledge, there is no research on the
prevent fracture propagation. When the driving force is greater most difficult conditions and the minimum wall thickness and
than the resistance, fracture expands; otherwise, the driving minimum Charpy V-notch energy required for toughness to
force is insufficient and fracture propagation stops. The velocity ductile fracture arrest for high-pressure supercritical CO2
criterion is to compare the propagation velocity of the pipelines, which are very important for CO2 pipeline design
decompression wave and the fracture propagation. If the
velocity of the decompression wave is always greater than the
fracture propagation velocity under the same pressure, the
fracture propagation will stop after a certain period of time;
otherwise, the crack will continue to expand until the pipe
geometry or toughness changes. These two criteria are unified in
nature, but the velocity criterion is more intuitive.
To avoid continued axial fracture propagation, common
measures are to improve the toughness of the pipe and to install
external crack arresters. Improving the toughness of the pipe can
increase the arrest pressure and reduce the fracture propagation
velocity. The main methods to increase the toughness of the
pipeline are increasing the wall thickness and improving the
Charpy impact toughness of the pipe material.9 Measures to
improve Charpy impact toughness of the pipe material include
improving the heat treatment process and adding suitable
alloying elements.10 If the pipeline can rely on its own toughness
to stop fracture propagation, the Charpy energy should be
improved to meet the requirement of the fracture propagation
velocity between the operating pressure and the arrest pressure
just below the decompression wave velocity of the supercritical Figure 1. General sketch of this work.
CO2 decompression process and to make it minimum, required
for the pipe to arrest fracture by its own toughness.
To determine the Charpy energy required to arrest fracture,
the V-notch Charpy impact test,11,12 the drop hammer tear
test,13 the crack tip opening angle,14 and the full-scale burst test
are usually required. The most commonly used method is to
determine the arrest toughness of the pipeline using the Battelle
two-curve model (BTC) modified by full-scale tests. The BTC
model investigates the fluid−solid coupling between the
decompression process and the fracture propagation process
by constructing a fracture propagation curve and a decom-
pression wave velocity variation curve.2 At lower steel strengths,
the BTC model can predict the Charpy V-notch energy required
to arrest fracture of pipes better, but when applied to high-
strength (X70 and above) steels, the Charpy V-notch energy
calculated by the BTC model has significant deviation.15,16 Leis
et al.17 tested the Charpy V-notch impact specimens of different
steel materials with the Charpy V-notch energies ranging from
24 to 352 J. Based on the test results, the BTC model was
corrected and successfully applied to the Canada−US X70 gas
pipeline. In addition, Eiber18 modified the BTC model for
application to the X80 steel pipeline.
Currently, there are many studies on the fracture toughness of
natural gas pipelines.19−21 However, it has been found that
pipelines transporting CO2 in the dense phase are more prone to
running-ductile fracture than natural gas pipelines.22 This is due
to the high saturation pressure reached from the dense phase or
supercritical phase state and because CO2 is prone to phase
change during the decompressing process, which will result in a
dramatic decrease in decompression wave velocity. Therefore, Figure 2. Calculation model for the toughness of the supercritical CO2
the crack arrest of the CO2 pipeline is more difficult than that of pipeline to arrest fracture.
16805 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

and pipe material selection. Since there is always a balance process. The state equations used in the literature to
between accuracy and efficiency, it is necessary to develop a predict the thermodynamic behavior of the CO 2
simpler and faster model to efficiently calculate the wall decompression process include GERG-2008, EOSCG-
thickness and Charpy energy required for toughness to stop GERG, PR, Peng−Robinson−Stryjek−Vera, etc.24−28 At
fracture of supercritical CO2 pipelines. present, calculation models used to predict the decom-
The main objective of the present paper is to develop a pression wave characteristics of the pipeline include
simplified model to predict the toughness required for ductile GASEDECOM,24 DECOM,26 PipeTech,29 and other
fracture arrest of the supercritical CO2 pipeline. The calculated models. However, these models are too complicated in
wall thickness and Charpy energy can provide recommendations calculating the decompression wave velocity of CO2
for supercritical CO2 pipeline design and pipe material selection. mixtures. In order to improve the calculation efficiency,
The organization of this work is as follows. First, the calculation Gu et al.23 developed a simple model based on the GERG-
methodology of the model is presented. Second, the most 2008 equation of state to predict the decompression wave
difficult working conditions for fracture arrest are determined behavior at the two-phase state.
according to the supercritical CO2 decompression wave model. Gu et al.23 assumed that the decompression process was
Then, the decompression wave velocity and fracture prop- a one-dimensional horizontal flow, the fluid inside the
agation velocity during the decompression process of the pipeline was in thermodynamic equilibrium, and there
supercritical CO2 pipeline are compared to determine whether was no slip between the gas and liquid phases in the
the pipeline toughness can arrest the fracture. Finally, the adiabatic flow process. Then, they compared the model
minimum wall thickness and minimum Charpy energy for calculation data and the experimental data to verify the
toughness to fracture arrest of the supercritical CO2 pipeline are model in predicting the decompression wave character-
obtained. The general sketch of this work is shown in Figure 1.

■
istics of pure CO2 and CO2 with impurities. The detailed
calculation process of decompression wave is shown in
CALCULATION METHODOLOGY the Appendix A.
The calculation model proposed in this work is based on the (3) The arrest pressure calculation
improved BTC model and a decompression wave prediction According to the BTC model, the pressure interval for
model developed by Gu etal.23 The key to analyzing the ductile calculating the fracture propagation velocity should be
fracture arrest of the supercritical CO2 pipeline is to determine from the initial pressure of supercritical CO2 pipeline
the decompression wave velocity curve and fracture propagation fracture propagation down to the arrest pressure. To
velocity curve of the decompression process. When the pressure obtain the arrest pressure of the supercritical CO2
inside the pipeline is greater than the arrest pressure, the pipeline, the arrest stress should be confirmed first.
decompression wave velocity is always greater than the fracture The calculation of the arrest stress was initially
propagation velocity and the pressure at the front end of fracture proposed by Maxey and his colleagues31 on the basis of
propagation keeps decreasing. When the pressure at the front the critical through-wall flaw size equation for initiation as
end of the fracture propagation decreases to the arrest pressure, shown in eqs 1−3. They deduced that the fracture arrest
the fracture propagation stops. Figure 2 shows the diagram of the process is static due to the fact that the fracture velocity
calculation model for ductile fracture arrest by toughness of the rapidly decreases to zero at this point, which is the reverse
supercritical CO2 pipeline decompression process established in process of crack initiation. Therefore, the arrest stress can
this work. be calculated by the fracture initiation equation and the
The solution process of the calculation model is as follows: equivalent half-length of the extended crack and then
corrected according to the results of full-scale experi-
(1) Steel parameter input
ments. They found that half of the effective crack length
The calculation for the ductile fracture arrest of the
supercritical CO2 pipeline requires the input of the can be expressed as 3 rDt , with a corresponding Folias
Young’s modulus, flow stress, pipe outer diameter, and factor MT of 3.33, which is brought into eq 1 to yield the
wall thickness. For different grades of steel, the modulus of modified arrest stress eq 4, and then the arrest pressure of
elasticity does not differ much, so the Young’s moduli of the pipe is found from the relationship between stress and
X65, X70, and X80 are taken to be 210 GPa. According to pressure, as shown in eq 5.
steel pipelines for use in pipeline transportation systems in 2σ ̅ 2 2
σT = arccos(e−πKc /8cσ̅ )
the petroleum and natural gas industries, the minimum

ÅÄÅ ÑÉÑ0.5
MT π (1)

ÅÅ jij c zyz jij c zyzÑÑÑ

yield strengths of X65, X70, and X80 steel are 450, 485,

Å
MT = ÅÅ1 + 1.255jj zz − 0.0135jj 2 2 zzÑÑ
and 555 MPa, respectively, and the flow stress is 69 MPa
ÅÅ k r Dt {ÑÑÑÖ
2 4

ÅÇ k rDt {
more than the minimum yield strength,2 so the flow
stresses of X65, X70, and X80 steels are taken to be 519, (2)
554, and 624 MPa, respectively.
Kc = (12.5 × CVN × E)0.5 (3)

ji i −12.5π × CVN × E yzzzyz

(2) Decompression wave velocity calculation

arccosjjjexpjjjj zzzzz
Since the leak of supercritical CO2 will produce a
j
k k {{
decompression wave at the fracture to both ends of the 2σ ̅
−σa =
pipeline, comparing the decompression wave velocity and 3.33π 24σ ̅ 2 rDt
the fracture propagation velocity can determine whether (4)
the pipeline fracture propagation stops or continues to
σaDt
expand.2 Pa = −
The equation of state is mainly used to describe the r (5)
thermodynamic behavior of the CO2 decompression (4) Calculation of the fracture propagation velocity
16806 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

When the pressure at the crack tip is higher than the

arrest pressure, the calculation formula for fracture
velocity of the supercritical CO2 pipeline is based on
the propagation speed of the plastic strain field, and
according to the existing data of the steady ductile fracture
propagation, the equation is shown in eq 6:
ij Pd y
jj − 1zzz
jj P zz
1/6
σ̅
k a {
Vf = C
CV (6)
(5) Comparison of decompression wave velocity and fracture
propagation velocity
According to the velocity criterion, it is necessary to
compare the decompression wave velocity and fracture
propagation velocity at different pressures during the
isentropic decompression process of supercritical CO2
leakage to determine whether the pipeline toughness can
arrest the fracture. When the pressure inside the pipeline
is higher than the arrest pressure, the decompression wave
velocity is always greater than the fracture propagation Figure 3. Decompression velocities under different initial pressures.
velocity, and the toughness of the pipeline can meet the
fracture arrest requirement.
(6) Calculation of the pipe wall thickness and the minimum If the pipeline can rely on its own toughness to arrest fracture,
Charpy V-notch energy the fracture propagation velocity should always be lower than
First, the minimum wall thickness of the pipeline is the decompression wave velocity eqs 7 and 8, and the complete
determined by the strength design. Without changing the calculation process of the required minimum wall thickness of
pipe toughness parameters, the Charpy V-notch energies the pipeline or the minimum Charpy V-notch energy is shown in
Figure 2.

■
per unit area of X65, X70, and X80 pipelines are 0.96, 1.59,
and 2.01 J/mm2, respectively. Then the decompression
RESULTS AND DISCUSSION
wave velocity and fracture propagation velocity are
calculated under the conditions of minimum wall Since the critical pressure of CO2 is 7.38 MPa and the critical
thickness and minimum Charpy V-notch energy to temperature is 304.13 K, the operating pressure of the
judge whether the toughness of the pipeline can satisfy supercritical CO2 pipeline should generally be maintained at
the ductile fracture arrest requirement of the supercritical 8−20.4 MPa and the operating temperature at 305.15−323.15 K
CO2 pipeline. Otherwise, it is necessary to increase the to ensure that phase change does not occur during the
wall thickness or improve the toughness of the pipeline for supercritical CO2 pipeline transportation.30 The decompression
arresting fracture. wave transmission law of the supercritical CO2 pipeline after
leakage is first analyzed according to the decompression wave
As mentioned above, the calculated deviation of Charpy V-
calculation models to determine the most difficult arrest
notch energy for high-strength steel calculated using the BTC
conditions of the pipeline. In the most difficult arrest conditions,
model is large. Therefore, it is necessary to improve the accuracy
the arrest pressure of the supercritical CO2 pipeline is higher
of Charpy V-notch energy calculation for X70 and X80 pipelines
than the saturation pressure of CO2, which means that the
by using full-scale experimental data.17,18 The modified Charpy
supercritical CO2 pipeline can rely on its own toughness to arrest
V-notch energy calculation equations for X70 and X80 pipelines
fracture.2 Otherwise, other methods such as crack arresters need

l
are shown in eqs 7 and 8.
o
o
to be used for arrest fracture.

m
o
When the supercritical CO2 pipeline cannot meet the ductile
o
CvArrest = CvBTCM Cv ≤ 95 J;
o
nCvArrest = CvBTCM + 0.002CvBTCM − 21.98
2.04
fracture arrest toughness condition, according to eqs 4−5, it can
Cv ≤ 95 J be seen that increasing the wall thickness of the pipeline and the

l
Charpy V-notch energy of the pipeline material can effectively
o
(7)

o
increase the arrest pressure, but the cost of laying the pipeline
m
o
o
CvArrest = CvBTCM Cv ≤ 95 J;
o
will increase by increasing the wall thickness or improving the

nCvArrest = CvBTCM + 0.003CvBTCM − 21.98

2.04 Cv ≤ 95 J toughness parameter of the pipeline. Therefore, by studying the
fracture propagation process of the pipeline with different wall
(8) thicknesses and Charpy V-notch energies, it can help in selecting
The purpose of calculating the wall thickness and the the appropriate wall thickness and toughness parameters of
minimum Charpy V-notch energy required to arrest fracture is supercritical CO2 pipelines.
to determine the toughness of a particular pipeline. From eqs 4 Decompression Wave Characteristics of the Super-
and 6, it can be seen that the wall thickness and the Charpy V- critical CO2 Leakage Process. Analyzing the decompression
notch energy of the pipeline material are related to the arrest wave characteristics of the CO2 leakage process with the initial
pressure and the fracture propagation velocity of the pipeline. By pressure and temperature in the supercritical region, it can be
increasing the wall thickness and the Charpy V-notch energy of determined that the most difficult working conditions for the
the pipeline material, it can not only increase the arrest pressure pipeline to arrest fracture is when the decompression wave
of the pipeline but also reduce the fracture propagation velocity. velocity is the smallest and the saturation pressure is the largest.
16807 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

If the arrest pressure under these conditions is higher than the

saturation pressure of CO2, it can be considered that the
supercritical CO2 pipeline can arrest fracture by its own
toughness.2
In order to determine the most difficult fracture arrest
conditions of the pipeline, the initial temperature of the
supercritical CO2 pipeline is first assumed to be a fixed value.
At this initial temperature (for example, 313.15 K), the leakage
process of supercritical CO2 is reduced under different initial
pressure conditions. The relationship between the decom-
pression wave velocity and the CO2 saturation pressure is shown
in Figure 3.
When the supercritical CO2 pipeline operates in a certain
pressure range, the initial pressure at the pipeline fracture
initiation may be any pressure in the interval. In this research, the
pressures at the crack initiation are chosen to be 20.4, 16, 12, 9.2,
and 8 MPa. It can be seen from Figure 3 that when the leakage
occurs in the supercritical CO2 pipeline, the temperature and
pressure inside the pipeline keep decreasing. When the pressure
Figure 4. Decompression wave velocities at different initial temper-
decreases to the saturation pressure, the decompression wave
atures.
velocity decreases sharply due to the phase change of CO2 inside
the pipeline. Moreover, it can be seen from Figure 3 that the
lower the saturation pressure is, the lower the pressure when the
decompression wave is equal to zero, and the greater the overall
decompression wave velocity is, which is beneficial to arrest
fracture according to the BTC model. However, as shown in
Figure 3, it is not that the higher the initial pressure is, the higher
the saturation pressure is and the more difficult it is to arrest
fracture. In the initial pressure range analyzed in this work, when
the initial pressure is 9.2 MPa (initial temperature is 313.15 K),
the saturation pressure of CO2 fracture-induced during the
leakage process is the largest and the pipeline cannot arrest
fracture under these working conditions according to the
requirement that the arrest pressure should be higher than the
saturation pressure.
It is also known from Figure 3 that the saturation pressure is
related to the initial pressure at the fracture port under certain
temperature conditions, and the saturation pressure is maximum
at an initial pressure of 9.2 MPa. Taking the initial pressure of the Figure 5. Saturation pressure change curves.
fracture to be 9.2 MPa, the decompression wave velocities at
different initial temperatures are shown in Figure 4.
As shown in Figure 4, when the initial pressure at the fracture
of the supercritical CO2 pipeline is 9.2 MPa, the saturation fracture are 307.65 K and 8 MPa (point 1), 313.15 K and 9.2
pressure of CO2 decreases with increasing temperature when the MPa (point 2), and 323.15 K and 11.7 MPa (point 3), the
initial temperature is in the range of 313.15 to 323.15 K. maximum saturation pressure of CO2 during the leakage process
However, when the initial temperature is in the range of can reach the critical pressure (7.38 MPa). If the pipe relies on its
305.15−313.15 K, the saturation pressure of CO2 rises with the own toughness to arrest fracture, it is required that the cracking
increase of temperature. From Figures 3 and 4, it can be seen pressure of the pipeline must be higher than the saturation
that the saturation pressure is not only related to the initial pressure of CO2, so the conditions of initial pressure and
pressure of the supercritical CO2 pipeline but also related to the temperature for these three points are the more difficult
initial temperature, and the saturation pressure is not linear with conditions for the pipeline to arrest fracture. When the operating
the initial temperature−pressure change at the fracture. So, the temperature is lower than 307.65 K, the saturation pressure
influence of initial pressure and temperature at the fracture on decreases as the initial pressure increases, i.e., the lower the
the saturation pressure should be considered comprehensively. initial pressure for fracture propagation, the more difficult it is to
Figure 5 presents the relationship between the saturation arrest fracture.
pressure of CO2 and the initial pressure at the fracture during the When calculating the decompression wave velocity, it is
leakage process of the supercritical CO2 pipeline when the initial assumed that the entire decompression process is an isentropic
temperatures are at 305.15, 307.65, 313.15, and 323.15 K. As depressurization process, which means that when the initial
shown in Figure 5, when the operating temperature of the entropy of the supercritical CO2 is consistent with the initial
supercritical CO2 pipeline is higher than 307.65 K, the saturation entropy at the critical point, the CO2 will inevitably undergo
pressure during the fracture propagation process increases and phase change at the critical point during the isentropic
then decreases with the increase of the initial pressure at the decompression process, and the decompression wave velocity
fracture port. When the initial temperature and pressure at the also drops sharply at the critical pressure.
16808 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

According to the equation of state, the entropy of CO2 at the from the gas phase or the liquid phase, the saturation pressure is
critical point (a temperature of 304.13 K and a pressure of 7.38 lower than the critical pressure, so when the initial entropy at the
MPa) is S = 1.4392 J/kg/K. The entropies of points 1, 2, and 3 in fracture port is the same as the critical point entropy, the
Figure 6 (also the three points in Figure 6) are 1.4395, 1.4250, saturation pressure during the decompression process is the
and 1.4343 J/kg/K, which are all similar to the entropy at the largest and the required arrest pressure is also the largest and
critical point. Therefore, when the initial entropy at the fracture higher than the critical pressure.
It is also known from Figure 6 that when the initial entropy of
the supercritical CO2 is the same, the saturation pressure is also
the same, and the requirements for the pipeline arrest pressure
are the same, but the decompression wave velocity change
process is different due to the difference of initial temperature
and pressure. The following analysis is performed for the
decompression process of different working conditions with
saturation pressure near the critical point (as shown in Figure 7)
to determine the most difficult working conditions for the
pipeline to arrest fracture.
According to the BTC model, the smaller the decompression
wave velocity is, the smaller the allowable rupture velocity for the
supercritical CO2 pipeline is, and the more difficult it is to arrest
fracture. It can be seen from Figure 7 that when the operating
pressure is in the range of 8−20.4 MPa and the temperature is in
the range of 305.15−323.15 K, the decompression wave velocity
is the smallest and the saturation pressure is 7.38 MPa under the
conditions of an initial pressure of 11.7 MPa and temperature of
323.15 K, which are the most difficult working conditions for the
supercritical CO2 pipeline to arrest fracture by its own
Figure 6. Supercritical CO2 isentropic pressure drop process. toughness.
The Influence of Charpy V-Notch Energy on Fracture
Propagation. In order to determine the ductile fracture arrest
port is consistent with the entropy at the critical point, the parameters and analyze the influence of Charpy V-notch energy
supercritical CO2 will enter the two-phase region at the critical on fracture propagation of the supercritical CO2 pipeline, the
point during the isentropic decompression process. At this time, decompression process of supercritical CO2 leakage of different
the saturation pressure is the critical pressure, as shown in the steel grades under certain pipe diameter and pressure conditions
curve of S = 1.4392 J/kg/K in Figure 7. When the initial entropy is analyzed. Three groups of supercritical CO2 pipelines with
at the fracture is higher than the entropy at the critical point, the 610 mm diameter and 13.2 MPa design pressure, 762 mm
supercritical CO2 will first become gaseous and then enter the diameter and 18.6 MPa, and 1016 mm and 20.4 MPa were
gas−liquid two-phase region. When the initial entropy at the selected. First, the wall thickness determined according to the
fracture is lower than the entropy of the critical point, the strength design criteria and without any toughness improvement
supercritical CO2 will change to a dense phase and liquid phase of pipe materials, whether the pipeline can complete the crack
successively and then enter the gas−liquid two-phase region. arrest process relying on its own toughness, was studied. When
However, whether CO2 enters the gas−liquid two-phase state the minimum wall thickness of the supercritical CO2 pipeline is
determined according to the pipeline strength design, the
pipeline design factor is 0.72. The X65, X70, and X80 pipes will
be analyzed below (Table 1).
In the supercritical CO2 decompression process, an initial
pressure is 11.7 MPa and a temperature is 323.15 K are the most
difficult conditions for complete arrest fracture by pipeline
toughness. The relationship between the supercritical CO2

Table 1. Toughness Parameters and Minimum Wall

Thicknesses of Different Pipelines

pipe OD design pressure CV minimum wall

material (mm) (MPa) ((J/mm2)) thickness (mm)
X70 610 13.2 0.96 12.23
X80 1.59 11.36
X80 2.01 9.95
X65 762 18.6 0.96 21.38
X70 1.59 19.87
X80 2.01 17.41
X65 1016 20.4 0.96 31.20
Figure 7. Decompression wave velocity diagrams at different points of X70 1.59 29.00
the isentropic curve. X80 2.01 25.41

16809 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

decompression wave velocity and fracture propagation velocity

in this situation is analyzed below to determine the ductile
fracture arrest of the pipeline.
It can be seen from Figures 8−10 that the decompression
wave velocity of the natural gas pipeline is much higher than
fracture propagation velocity and CO2 decompression wave
velocity, and the fracture arrest process can be completed by its

Figure 10. Fracture and decompression wave velocity curves of

different material pipelines with 1016 mm diameter.

not meet the requirements of ductile fracture arrest for the

supercritical CO2 pipeline. When the pipe diameter is 762 mm
and the design pressure is 18.6 MPa, and the pipe diameter is
1016 mm and the design pressure is 20.4 MPa, the minimum
Figure 8. Fracture and decompression wave velocity curves of different wall thickness of the pipeline to meet the strength design is larger
material pipelines with 610 mm diameter. due to the high design pressure of the pipeline. When using the
velocity criterion to calculate the pipeline toughness, it is found
that the minimum wall thicknesses of X70 and X80 pipelines
meet the ductile crack arrest conditions, but the X65 pipeline
cannot meet the ductile crack arrest conditions.
For the pipeline that cannot meet the requirement of ductile
fracture arrest, improving the toughness of the pipeline material
or increasing the wall thickness is generally adopted so that it can
rely on its own toughness to stop the crack. First, the influence of
improving the toughness of the pipeline material (increasing the
Charpy energy of the pipeline material) on the fracture process
of the supercritical CO2 pipeline is analyzed.
According to eqs 4−6, it is known that increasing the Charpy
V-notch energy of supercritical CO2 pipelines can not only
increase the arrest stress of the pipeline but also reduce the
fracture propagation velocity. When the Charpy V-notch energy
is taken to infinity, the arrest stress reaches the maximum value

jij ijj yzy

as shown in eq 9.

j jj −12.5π × CVN × E zzzzzzz

j
j
arccosjjexpjj j zzzzz
jj jj zzzz
zz
2σ ̅
j
−σamax =
k k {{
3.33π 24 σ 2 DwDt
Figure 9. Fracture and decompression wave velocity curves of different ̅ 2
material pipelines with 762 mm diameter.
2σ ̅ σ̅
= arccos(0) =
3.33π 3.33 (9)
own toughness. However, for the supercritical CO2 pipeline,
when the pipe diameter is 610 mm and the design pressure is When the diameter and wall thickness of the pipeline are not
13.2 MPa, the arrest pressures of the X65, X70, and X80 pipes changed, the arrest pressure corresponding to the maximum
are less than the saturation pressure of the CO2 decompression arrest stress reaches the maximum, and the maximum arrest
process when they meet the minimum wall thickness of strength pressure is calculated according to the following formula.
requirements. During the fracture propagation process, the 2σ ̅Dt
fracture propagation velocity is higher than the decompression Pa =
3.33Dw (10)
wave propagation velocity, the pressure at the front of crack
propagation remains unchanged, and the speed of crack The ductile fracture arrest of the supercritical CO2 pipeline
propagation remains unchanged, which is about 100−150 m/ with 610 mm outer diameter and 13.2 MPa design pressure is
s, so the fracture arrest process cannot be completed, which does analyzed. When the outer diameter of the pipe is 610 mm and
16810 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

the wall thicknesses of X65, X70, and X80 pipelines are 12.23,
10.94, 9.58 mm, respectively, the maximum arrest pressures
achieved by increasing the Charpy V-notch energy are 6.25,
6.20, and 6.11 MPa, respectively, which are all less than the
saturation pressure of 7.38 MPa for the most difficult fracture
arresting conditions. Therefore, when the wall thickness of the
supercritical CO2 pipeline is taken to meet the minimum value
of strength design, the fracture arrest cannot be accomplished by
increasing the Charpy V-notch energy of the pipeline material in
this condition. When the outer diameter of the pipe is 1016 mm
and the wall thickness of X65 is 31.20 mm, the maximum arrest
pressure is 9.57 MPa, so it can arrest fracture by increasing the
Charpy V-notch energy of the pipeline material. Moreover, the
energy of the Charpy V-notch cannot be increased infinitely for
the actual pipeline material. Therefore, the minimum Charpy V-
notch energies for the different wall thicknesses of X65, X70, and
X80 pipelines to meet the crack-stopping toughness will be
analyzed below. The specific calculation process is referred to
Figure 2.
It can be seen from Figure 11 that the minimum Charpy V-
notch energy required for the same wall thickness of pipelines
with different strength classes is almost the same. When the
pipeline wall thickness is less than dc, increasing the pipeline wall
thickness can significantly reduce the minimum Charpy V-notch
energy required to arrest fracture. However, when the pipeline
wall thickness is greater than dc, it is not feasible to reduce the
requirements for pipeline material toughness by increasing the
pipeline wall thickness.
The minimum Charpy V-notch energy is only the basic
requirement that the pipeline can arrest fracture by its own
toughness. To reduce the fracture propagation velocity, the
Charpy V-notch energy of the pipeline can be further increased
by changing the heat treatment process of the pipeline or adding
alloys. As shown in Figure 12, the effect of Charpy V-notch
energy on the fracture propagation process is analyzed for the
X65 pipeline with a wall thickness of 18 mm as an example.
As can be seen from Figure 12, increasing the Charpy V-notch
energy of the pipeline can not only increase the arrest pressure of
the pipeline and stop the fracture propagation process in
advance but also can effectively reduce the fracture propagation
velocity, shrink the pipeline fracture length, and effectively Figure 11. Minimum Charpy V-notch energies of different pipeline wall
reduce the economic loss caused by pipeline rupture. thicknesses: (a) OD = 610 mm; (b) OD = 1016 mm.
The Influence of Pipeline Wall Thickness on Crack
Expansion of the Fracture Propagation Process of the thickness of the pipeline can be appropriately increased to
Supercritical CO2 Pipeline. From eq 10, it can be seen that improve the arrest pressure of the pipeline. The following will
increasing the arrest pressure by raising the Charpy V-notch combine the supercritical CO2 decompression process to
energy will eventually be limited by the pipeline wall thickness determine the minimum wall thicknesses of different material
condition. Therefore, in some cases, the pipeline wall thickness pipelines to meet the toughness requirement.
must be appropriately increased to meet the toughness In the supercritical CO2 decompression process, the
requirement for arrest. saturation pressure of CO2 is the highest under the initial
Under the condition that the Charpy V-notch energy per unit conditions that the initial pressure is 11.7 MPa and the initial
area remains constant, it is known from eqs 4 and 5 that temperature is 323.15 K. This saturation pressure is close to the
increasing the wall thickness of the pipeline can increase the critical pressure and the decompression wave velocity is the
arrest pressure of the pipeline and reduce the fracture slowest, which are the most difficult conditions to complete
propagation velocity, which is beneficial to the pipeline arrest. fracture arrest by relying on the toughness of the pipeline. The
The influence of the X65, X70, and X80 pipeline wall thicknesses relationship between the supercritical CO2 decompression wave
on the arrest pressure will be analyzed in the following, as shown transfer and fracture propagation in this most difficult situation
in Figure 13. is analyzed below to determine the ductile fracture arrest of the
It can be seen from Figure 13 that the arrest pressure increases pipeline.
with the increase of pipe wall thickness under the condition of As shown in Figure 8, when the outer diameter of the
certain Charpy V-notch energy, which is basically linear. supercritical CO2 pipeline is 610 mm and the design pressure is
Therefore, when the toughness of the supercritical CO2 pipeline 13.2 MPa, the minimum wall thickness of the pipeline
does not meet the ductile fracture arrest requirements, the wall determined according to the strength design cannot meet the
16811 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

Figure 12. Effect of X65 pipeline Charpy V-notch energy on fracture

velocity.

requirements of the ductile fracture arrest of the pipeline, and if

the pipeline is required to arrest fracture by its own toughness,
the arrest pressure can be increased and the fracture propagation
velocity can be reduced by increasing the wall thickness. Figures
14 and 15 show the analysis of the fracture propagation process
and the decompression process of the supercritical CO2 pipeline
with different wall thicknesses for X65 and X80 pipelines as
examples.
It can be seen from Figure 14 that the wall thickness of the
X65 pipeline should not be less than 17.28 mm to prevent
fracture propagation of the supercritical CO2 pipeline without
changing the Charpy V-notch energy of the pipeline material. In
order to prevent ductile fracture of the supercritical CO2
pipeline, the wall thickness of the X65 pipeline should not be
less than 17.28 mm. It can also be seen from Figures 14 and 15
that increasing the pipeline wall thickness can improve the arrest
pressure, but the effect of reducing the fracture propagation
velocity at higher pressures is not obvious. According to this
model, when the pipeline diameter is 610 mm, the minimum Figure 13. Relationship between fracture arrest pressure and pipeline
wall thickness: (a) OD = 610 mm; (b) OD = 1016 mm.
wall thicknesses of X70 and X80 pipelines to meet the ductile
fracture arrest are 14.58 and 12.81 mm, respectively.

■ SUMMARY AND CONCLUSIONS

Based on the improved BTC model combined with the
(3) The arrest pressure of the pipeline can be improved by
increasing the pipeline wall thickness or increasing the
Charpy V-notch energy of the pipeline material. Mean-
decompression wave calculation model of the CO2 pipeline while, the method of increasing the Charpy V-notch
leakage process, the model for arrest toughness of the energy of the pipeline will ultimately be limited by the
supercritical CO2 pipeline is established and the following pipeline geometry, and within the scope of this paper;
conclusions are obtained. when the pipeline wall thickness is greater than dc, it is
difficult to reduce the minimum Charpy V-notch energy
(1) During the decompression process of the supercritical required for arrest toughness of the pipeline by increasing
CO2 pipeline, when the entropy at the fracture port is the the pipeline wall thickness. The wall thickness of the
same as that of the critical point, the saturation pressure of pipeline is linearly related to the arrest pressure if other
CO2 can reach the critical pressure, and the arrest pressure conditions are unchanged, but increasing the wall
must be higher than critical pressure. The most difficult thickness will affect the economic efficiency. Therefore,
working conditions for the supercritical CO2 pipeline to a suitable wall thickness and Charpy V-notch energy
arrest fracture by its own toughness are an initial pressure should be selected to reduce the fracture propagation
of 11.7 MPa and temperature of 323.15 K. velocity of the pipeline.
(2) When the strength and design pressure of the supercritical
CO2 pipeline are high, the minimum wall thickness The present work has presented a new simplified model to
obtained according to the strength design criterion can predict the minimum wall thickness and minimum Charpy
meet the requirements of ductile fracture arrest. energy required to stop fracture in the supercritical CO2
16812 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

a= λZRT (A2)

CO2 changes from the single phase to the gas−liquid two-

phase with the decrease of temperature and pressure. Assuming
that the pressure and temperature of both gas−liquid phases are
in equilibrium, the equation for the sound velocity a2 in pressure
and temperature equilibrium is shown as eq A3
2
1 1 ρ Cp , gCp , l(ξl − ξg )
= +
a 22 a12 T Cp , g + Cp , l (A3)

where k ∈ {g, l}, a1 is the sound velocity in pressure equilibrium

as shown in eq A4, and ξk and Cp, k are determined using eq
A5−A7 respectively.
ij x 1 − xg yzz
2jjj g + zz
j
jj ρ2 a 2 2 2 z zz
1

k g {
= ρ
a12 g ρ l
a l (A4)

ji ∂T zy
ξk = jjj zzz
j ∂p z
Figure 14. Fracture and decompression velocity curves of the X65

k {sk
pipeline.
(A5)

Cp , k = ρk xkck (A6)

i ∂s y
ck = T jjj zzz
k ∂T { (A7)

The outflow velocity at the fracture of pipeline can be

determined by eq A8.
pi − 1 − pi
Ui = Ui − 1 +
aiρi (A8)

After the calculation of sound velocity and outflow velocity at

the fracture, the decompression wave velocity can be obtained
by eq A1.

■ AUTHOR INFORMATION
Corresponding Author
Nan Zhang − Shandong Key Laboratory of Oil & Gas Storage
Figure 15. Fracture and decompression wave velocity curves of the X80 and Transportation Safety, China University of Petroleum
pipeline. (East China), Qingdao 266580, China; Email: s18060071@
s.upc.edu.cn
pipeline. This model is relatively simple and the program is less
computationally intensive, which is suitable for engineering Authors
applications. However, the fracture velocity equation may not be Qihui Hu − Shandong Key Laboratory of Oil & Gas Storage
accurate when the diameter of the pipeline is less than 457 mm, and Transportation Safety, China University of Petroleum
so this model is not suitable for small-diameter pipelines. (East China), Qingdao 266580, China

■
Yuxing Li − Shandong Key Laboratory of Oil & Gas Storage
APPENDIX and Transportation Safety, China University of Petroleum
(East China), Qingdao 266580, China
A. Decompression Wave Velocity Calculation Wuchang Wang − Shandong Key Laboratory of Oil & Gas
The supercritical CO2 decompression wave velocity is calculated Storage and Transportation Safety, China University of
as follows: Petroleum (East China), Qingdao 266580, China
W=a−U (A1) Jianlu Zhu − Shandong Key Laboratory of Oil & Gas Storage
As can be seen from eq A1, the decompression wave velocity is and Transportation Safety, China University of Petroleum
related to the sound velocity and the leakage outflow velocity, (East China), Qingdao 266580, China
and the calculation of CO2 sound velocity is divided into two Jiyu Gong − Shandong Key Laboratory of Oil & Gas Storage
cases, single phase and two phases. and Transportation Safety, China University of Petroleum
The sound velocity that can be obtained from the equation of (East China), Qingdao 266580, China
state (eq A2) is related to the temperature when CO2 is in the Complete contact information is available at:
single phase. https://pubs.acs.org/10.1021/acsomega.1c01360
16813 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

Notes (8) Aursand, E.; Dumoulin, S.; Hammer, M.; Lange, H. I.; Morin, A.;
The authors declare no competing financial interest. Munkejord, S. T.; Nordhagen, H. O. Fracture propagation control in

■
CO2 pipelines: Validation of a coupled fluid-structure model. Eng.
Struct. 2016, 123, 192−212.
ACKNOWLEDGMENTS (9) Jiang, X.; Qu, D. R.; Liu, X. H. Supercritical CO2 pipeline
This work was supported by the Fundamental Research Funds transportation and safety. Oil Gas Storage Transp. 2013, 32, 809−813.
for the Central Universities, China (20CX02405A), the (10) Zhang, Y.; Shuai, J.; Lv, Z. Y. Investigation of the effects of
Development Fund of Shandong Key Laboratory of Oil & Gas material parameters on the relationship between crack tip constraint
Storage and Transportation Safety, and the National Science and CTOD fracture toughness. Theor. Appl. Fract. Mech. 2020, 108,
20621−20629.
and Technology Special Project (2016ZX05016-002).

■
(11) Landrein, P.; Lorriot, T.; Guillaumat, L. Influence of some test
parameters on specimen loading determination methods in instru-
NOMENCLATURE mented Charpy impact tests. Eng. Fract. Mech. 2001, 68, 1631−1645.
σ̅ , flow stress (MPa); E, Young’s modulus (MPa); Dw, pipeline (12) Toshiro, K.; Isamu, Y.; Mitsuo, N. Evaluation of dynamic fracture
diameter (mm); P, pressure (MPa); T, temperature (K); W, toughness parameters by instrumented Charpy impact test. Eng. Fract.
Mech. 1986, 24, 773−782.
decompression wave velocity (m/s); Dt, wall thickness of the (13) Kang, M.; Kim, H.; Lee, S.; Shin, S. Y. Effects of Dynamic Strain
pipeline (mm); CVN, Charpy V-notch energy (J); Pa, arrest Hardening Exponent on Abnormal Cleavage Fracture Occurring
pressure (MPa); Vf, ductile fracture propagation velocity (m/s); During Drop Weight Tear Test of API X70 and X80 Linepipe Steels.
a, sound speed (m/s); U, leakage flow velocity (m/s); λ, Metall. Mater. Trans. A 2014, 45, 682−697.
isentropic expansion coefficient (−); Z, compressibility factor (14) Horsley, D. J. Background to the Use of CTOA for Prediction of
(−); R, universal gas constant (8.314 J/kg/K); ρ, density (kg/ Dynamic Fracture Arrest in Pipelines. Eng. Fract. Mech. 2003, 70, 547−
m3 ); C p,k , extensive heat capacity (J/K/m3); x, molar 552.
composition (−); cp, heat capacity (J/kg/K); s, specific entropy (15) Bai, Y.; Wierzbicki, T. A new model of metal plasticity and
(J/kg/K); σT, hoop stress at failure (MPa); MT, Folias factor for fracture with pressure and Lode dependence. Int. J. Plast. 2008, 24,
a through-wall flaw (−); Kc, critical (plane stress) stress intensity 1071−1096.
factor (N/mm1.5); c, half-length of a through-wall crack (mm); r, (16) Barlat, F.; Lege, D. J.; Brem, J. C. A six-component yield function
for anisotropic materials. Int. J. Plast. 1991, 7, 693−712.
nominal radius of the pipeline (mm); σa, arrest stress (MPa); Pa, (17) Leis, B. N.; Eiber, R. J.; Carlson, L. Relationship Between
arrest pressure (MPa); Vf , fracture propagation velocity (m/s); Apparent (Total) Charpy Vee-Notch Toughness and the Correspond-
C, constant, the backfilled pipe is 0.275 and the unbackfilled pipe ing Dynamic Crack-Propagation Resistance. International Pipeline
is 0.379; CV, Charpy V-notch energy per unit area (J/mm2); Pd, Conference; American Society of Mechanical Engineers 2018, 2, 723−
pressure at the crack tip (MPa) 731.
Subscripts (18) Eiber, R. Fracture-arrest prediction requires correction factors.
Oil Gas J. 2008, 106, 52−54+58.
g, gaseous phase; l, liquid phase

■
(19) Gu, X.; Guo, Y. Crack Arrest Toughness of High Grade Gas
Pipeline. Materials Science Forum; Trans Tech Publications Ltd 2017,
REFERENCES 898, 758−765, DOI: 10.4028/www.scientific.net/MSF.898.758.
(1) Martynov, S.; Brown, S.; Mahgerefteh, H.; Sundara, V. Modelling (20) Maruschak, P. O.; Panin, S. V.; Chausov, M. G.; Bishchak;
choked flow for CO2 from the dense phase to below the triple point. Int. Polyvana, U. V. Effect of long-term operation on steels of main gas
J. Greenhouse Gas Control 2013, 19, 552−558. pipeline. Reduction of static fracture toughness. J. Nat. Gas Sci. Eng.
(2) Mohitpour, M.; Seevam, P.; Botros, K. K. Pipeline Transportation 2017, 38, 182−186.
of Carbon Dioxide Containing Impurities. 2012. ISBN: 978−0−7918- (21) Lee, J. S.; Ju, J. B.; Jang, J. I.; Jang, J. I.; Kim, W. S.; Kwon, D. Weld
5983-4. crack assessments in API X65 pipeline: Failure assessment diagrams
(3) Chong, F. K.; Lawrence, K. K.; Lim, P. P.; et al. Planning of carbon with variations in representative mechanical properties. Mater. Sci. Eng.,
capture storage deployment using process graph approach[J]. Energy A 2004, 373, 122−130.
2014, 76, 641−651. (22) Eskil, A.; Torodd, B.; Cato, D. CO2 Pipeline Integrity: A Coupled
(4) Sim, S.; Cole, I. S.; Choi, Y. S.; Birbilis, N. A review of the Fluid-structure Model Using a Reference Equation of State for CO2.
protection strategies against internal corrosion for the safe transport of Energy Proc. 2013, 37, 3113−3122.
supercritical CO2 via steel pipelines for CCS purposes. Int. J. Greenhouse (23) Gu, S. W.; Li, Y. X.; Teng, L.; Hu, Q.; Zhang, D.; Ye, X.; Wang,
Gas Control. 2014, 29, 185−199. C.; Wang, J.; Iglauer, S. A new model for predicting the decompression
(5) Evans, W. C.; Kling, G. W.; Tuttle, M. L.; Tanyileke, G.; White, L. behavior of CO2 mixtures in various phases. Process Saf. Environ. Prot.
D. Gas buildup in Lake Nyos, Cameroon: The recharge process and its 2018, 120, 237−247.
consequences. Appl. Geochem. 1993, 8, 207−221. (24) Cosham, A.; Eiber, R. J.; Clark, E. B. Gasdecom: Carbon Dioxide
(6) Woolley, R. M.; Fairweather, M.; Wareing, C. J.; Proust, C.; and Other Components. Proceedings of the 8th International Pipeline
Hebrard, J.; Jamois, D.; Narasimhamurthy, V. D.; Storvik, I. E.; Skjold, Conference; IPC 2010, 2, 777−794.
T.; Falle, S. A. E. G.; Brown, S.; Mahgerefteh, H.; Martynov, S.; Gant, S. (25) Elshahomi, A.; Lu, C.; Michal, G.; Liu, X.; Godbole, A.; Venton,
E.; Tsangaris, D. M.; Economou, I. G.; Boulougouris, G. C.; P. Decompression wave speed in CO2 mixtures: CFD modelling with
Diamantonis, N. I. An integrated, multi-scale modelling approach for the GERG-2008 equation of state. Appl. Energy. 2015, 140, 20−32.
the simulation of multiphase dispersion from accidental CO2 pipeline (26) Jie, H. E.; Xu, B. P.; Wen, J. X. Predicting the Decompression
releases in realistic terrain. Int. J. Greenhouse Gas Control 2014, 27, 221− Characteristics of Carbon Dioxide Using Computational Fluid
238. Dynamics. Proceedings of the 2008 7th International Pipeline Conference;
(7) Woolley, R. M.; Fairweather, M.; Wareing, C. J.; Falle, S. A. E. G.; American Society of Mechanical Engineers 2012, 3, 585−595.
Mahgerefteh, H.; Martynov, S.; Brown, S.; Narasimhamurthy, V. D.; (27) Munkejord, S. T.; Hammer, M. Depressurization of CO2-rich
Storvik, I. E.; Sælen, L.; Skjold, T.; Economou, I. G.; Tsangaris, D. M.; mixtures in pipes: Two-phase flow modelling and comparison with
Boulougouris, G. C.; Diamantonis, N.; Cusco, L.; Wardman, M.; Gant, experiments. Int. J. Greenhouse Gas Control. 2015, 37, 398−411.
S. E.; Wilday, J.; Zhang, Y. C.; Chen, S.; Proust, C.; Hebrard, J.; Jamois, (28) Teng, L.; Li, Y.; Zhao, Q.; Wang, W.; Hu, Q.; Ye, X.; Zhang, D.
D. CO2 PipeHaz: Quantitative hazard assessment for next generation Decompression characteristics of CO2 pipelines following rupture. J.
CO2 pipelines. Energy Proc. 2014, 63, 2510−2529. Nat. Gas Sci. Eng. 2016, 36, 213−223.

16814 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815
ACS Omega http://pubs.acs.org/journal/acsodf Article

(29) Mahgerefteh, H.; Brown, S.; Denton, G. Modelling the impact of

stream impurities on ductile fractures in CO2 pipelines. Chem. Eng. Sci.
2012, 74, 200−210.
(30) Seevam, P. N.; Race, J. M.; Downie, M. J.; Hopkins, P.
Transporting the Next Generation of CO2 for Carbon, Capture and
Storage: The Impact of Impurities on Supercritical CO2 Pipelines.
Proceedings of the 2008 7th International Pipeline Conference; ASME
2008, 1, 39−51.
(31) Maxey, W. A.; Kiefner, J. F.; Eiber, R. J.; Duffy, A. R. Ductile
Fracture Initiation, Propagation and Arrest in Cylindrical Vessels. STM
Special Technical Publication 1972, 70−81.

16815 https://doi.org/10.1021/acsomega.1c01360
ACS Omega 2021, 6, 16804−16815