Engineering Failure Analysis 122 (2021) 105265

Contents lists available at ScienceDirect

Engineering Failure Analysis

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

In situ mathematically simulation for CO2 internal corrosion in

wet natural gas gathering pipelines system by HYSYS
Min Dai *
Xinjiang Petroleum Engineering Co., Ltd., China Petroleum Engineering & Construction Corporation (CPECC), Karamay 834000, Xinjiang, China

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

Keywords: The rate of corrosion due to the presence of carbon dioxide in a carbon steel pipe is an important
CO2 internal corrosion design consideration for gathering pipelines, particularly in the transport of wet natural gas. To
ASPEN HYSYS evaluate the corrosion rate in a gathering pipeline, in situ simulation study using HYSYS program
Wet natural gas
employed to predict CO2 corrosion in natural gas gathering pipelines system. The effects of
Gathering pipelines
NORSOK M− 506 Standard
operational conditions, inhibitors, pipe parameters and flow regime on the CO2 corrosion are
studied in this paper. Increasing operating pressure leads to increase the pH and decrease the CO2
corrosion rate when the operating pressure over 10 MPa and the solution contents 0.85kmol/m3
iron carbonate. The CO2 corrosion rate does not always increase with temperature, and which will
reach a maximum value at 45 ◦ C. A high gas velocity will lead to a high corrosion rate. Less free
water present in pipeline may lead to more serious corrosion than that of a large amount of free
water. The corrosion rate may be inhibited by liquid hydrocarbon and glycol, and the glycol is not
an economic inhibitor, especially when there is a large amount of free water in pipeline. The large
diameter and flat pipe are useful for decreasing CO2 corrosion rate. The shear stress and corrosion
rate will increase sharply when the slug flow forms in pipeline. Through comparing the data of
simulation with field corrosion rate data, feasibility of this numerical simulation method is
proved.

1. Introduction

With the increasing demand for energy in the world, natural gas has been the fossil energy resource with the fastest growth
recently. Natural gas is a widely used clean fossil fuel which can reduce the greenhouse emissions. Some agencies have estimated that
the consumption of natural gas in the world will increase by 1.7% per year [1].
Because of economic reasons most of the pipes used in natural gas gathering industries were made from different types of carbon
steel instead of stainless steel, which has a lower corrosion resistance against CO2 corrosion. This class of material is susceptible to CO2
corrosion in wet natural gas services. CO2 is a naturally existing compound in natural gas fields in quantities ranging from traces to
about 50% presented in produced water, oil, and gas processing. In the presence of water, CO2 dissolves to form carbonic acid which is
corrosive causing different forms of corrosion [2–3]. Approximately 60% of failures in the oilfield are caused by CO2 corrosion due to
the inadequate prediction methods and the low resistance of steels to this type of corrosion [4–5]. The cost for corrosion problem
worldwide is estimated to be more than 3.3 trillion dollars [6].
The mechanism of CO2 corrosion is complicated and is affected by various factors such as temperature, CO2 partial pressure, flow

* Corresponding author.
E-mail address: dai_min@126.com.

https://doi.org/10.1016/j.engfailanal.2021.105265
Received 11 March 2020; Received in revised form 19 January 2021; Accepted 25 January 2021
Available online 2 February 2021
1350-6307/© 2021 Elsevier Ltd. All rights reserved.
M. Dai Engineering Failure Analysis 122 (2021) 105265

rate, salt concentration, etc. [7–8]. From literature, it is evident that in the absence of protective scales, corrosion rate increases with
increasing CO2 partial pressure and in high values of pH, increasing the CO2 partial pressure enhances the protective scale precipitation
which decreases corrosion rate [9]. Furthermore, increasing temperature accelerates the CO2 corrosion rate in low pH, but in higher
pH, as iron carbonate solubility decreases, formation of protective scales decreases the corrosion rate [10–11]. In the absence of in­
hibitor and non-scaling conditions (pH < 6, T < 60 ◦ C), the increase in flow rate increases corrosion rate. On the other hand, in case of
scaling condition (pH ≥ 6, T ≥ 60 ◦ C) or presence of inhibitor film, the effect of flow rate is insignificant [12]. Several correlations have
been modified to estimate the corrosion rate for the internal flow based on the main mentioned parameters. DeWaard et al. [13]
developed semi empirical correlations to calculate CO2 corrosion rate based on CO2 partial pressure with velocity up to 1 m/s at the
surface of the pipelines. Jepson et al. [14] developed a fully empirical correlation to estimate the corrosion rate in horizontal pipelines
with multiphase slug flow. The model relates the corrosion rate with carbon dioxide partial pressure, water cut, pressure gradient
across the mixing zone, and temperature which modified later to including the effects of oil type and slug frequency. Wang et al. [15]
presented correlation covering different parameters such as reactions on the pipeline surface, water chemistry, and mass change
between pipeline surface and the bulk of fluid. Nesic et al. [16–17] developed a comprehensive correlation to predict corrosion rate
into mild steel pipelines includes the effects of hydrogen sulfide, water entrainment in multiphase flow, and corrosion inhibition by
crude oil. Muhammad Hashim Abbasa et al. [18] developed a neural network modelling of high pressure CO2 corrosion in pipeline
steels to CO2 corrosion rate. In 2005, Norwegian petroleum industry modified new correlation to predict the CO2 corrosion based on
different parameters [19]. NORSOK M− 506 Standard models applied which is widely accepted model to predict the CO2 corrosion rate
developed with support from Norwegian Oil Industry Association (OLF) and the Federation of Norwegian Manufacturing Industries
(TBL) [20].
Although a large amount of works on the theoretical and experimental studies about CO2 corrosion in the most recent 20 years, the
presence of CO2 in a hydrocarbon pipeline can cause an undesirable deterioration of the pipe wall. This corrosion can cause serious
flow problems that may ultimately lead to a catastrophic failure of pipe. The rate of corrosion due to the presence of carbon dioxide in a
carbon steel pipe is an important design consideration for pipelines, particularly in the transport of wet natural gas. The prediction of
this corrosion rate is critical when deciding if carbon steel was an acceptable material or if anti-corrosion measures such as the addition
of inhibitor are required. To evaluate the corrosion rate in a pipeline, in situ simulation study using HYSYS program employed to
predict CO2 corrosion in wet natural gas gathering pipelines system in this paper.

2. Process description

Gas wells are connected to network of gathering pipelines where these pipes gathering natural gas and send it out to the man
gathering pipeline for longer distance transport as shown by Fig. 1. The diameter of gathering pipelines is usually small, and which has
a high operating pressure. The gas gathering pipelines system work on principles of moving gas from gas well to natural gas process
plant. In order to prevent the gas gathering pipelines from being blocked by the hydrates formed in the transportation process, the
produced gas must be heated or added inhibitor. The produced gas from well is called wet gas which contained gas hydrocarbon, liquid
hydrocarbon (gas condensate), water and some other impurities such as CO2. Commonly, the operating pressure and temperature of
gathering pipelines are under 20 MPa and 60 ◦ C, respectively. The material primary selection of gathering pipelines is mild steel, unless
there is a large amount of chloride and hydrogen sulfide in produced water. The CO2 corrosion considered as one of the common
problems that facing gas gathering process which is known as sweet corrosion.

Fig. 1. The schematic diagram for gas gathering pipelines system and process.

2
M. Dai Engineering Failure Analysis 122 (2021) 105265

3. Simulation basis

The natural gas gathering process was simulated by the Aspen HYSYS V10.0 program which was a chemical process program
widely applied to mathematically simulate gas gathering and transportation processes. Aspen HYSYS offered different types of
thermodynamic property fluid packages to deal with hydrocarbon systems such as Peng-Robinson, SRK, Glycol, Acid Gas et al.. The
Acid Gas fluid package was recommended for hydrocarbons associated with CO2 due to its large binary interaction parameters. As
shown in Fig. 2, the calculation model selected the Pipe Segment which not only can calculate pipeline hydraulics but also offer CO2
corrosion analysis.
The Aspen HYSYS employed different models to estimate the CO2 corrosion such as NORSOK standard M− 506, and de Waard
Model 1991 and 1995. NORSOK M− 506 was developed with the support of the Norwegian Oil Industry Association (OLF) and the
Federation of Norwegian Manufacturing Industries (TBL), and de Waard Model 1991 and 1995 was developed by de Waard and
colleagues at Shell Oil. The NORSOK M− 506 correlation is developed as a purely empirical corrosion rate model, focused on most
important parameters (temperature, pH, CO2 partial pressure, and wall shear stress). The de Waard models are developed with more
theoretical basis, the temperature-dependency terms are based on a kinetic model, and experimental data was used to fit some of the
constants. The NORSOK M− 506 is widely accepted, and works best when the conditions are tried to model are near the conditions used
to develop the correlations. This model presents are commended practice for calculation of corrosion rates in hydrocarbon production
and process systems where the corrosive agent is CO2. The NORSOK M− 506 model is an empirical corrosion rate model for carbon steel
in water containing CO2 at different temperatures, pH, CO2 fugacity and wall shear stress. It is based on flow-loop experiments at
temperatures from 5 ◦ C to 160 ◦ C. A large amount of data at various temperatures, CO2 fugacity, pH and wall shear stresses are used to
develop the model. To calculate the pH, NORSOK M− 506 model employ powerful Newton-Raphson iterative technique to find sus­
pected root of the very complex hydrogen ion cation concentration equation. In this simulation, NORSOK M− 506 Standard model was
applied. The general equations for the corrosion rate [19] were given by:

(1)
◦ 0.36
T = 5 C, CRt = Kt fCO2 ft (pH)
( )0.146+0.0324logf CO2
S
0.36
T = 15oC, CRt = Kt fCO2 ft (pH) (2)
19
( )0.146+0.0324logf CO2
S
0.62
20oC ≤ T ≤ 150oC, CRt = Kt fCO2 ft (pH) (3)
19

where CRt is the corrosion rate (mm/year). fCO2 is the fugacity of carbon dioxide, which related to the CO2 partial pressure (bar). Kt is a
temperature-dependent corrosion rate constant. Table 1 of Reference [19] gives values for temperatures between 5 ◦ C and 150 ◦ C.
ft(pH) is the pH function, which is given in Table 2 of Reference [19]. The function has been evaluated for temperatures between 5 ◦ C
and 150 ◦ C and for pH values between 3.5 and 6.5. S is the shear stress at the pipe wall (Pa), which is determined with the pipe flow
correlation used to calculate the pressure drop across the pipe, and which can be calculated by Eq. (4).

S = 0.5 × ρm × f × u2m (4)

where the friction factor, f, can be expressed as:

Fig. 2. Process flow scheme of the simulated gathering pipeline by Aspen HYSYS.

3
M. Dai Engineering Failure Analysis 122 (2021) 105265

Table 1
Simulation basis and feed conditions.
Property Condition
Temperature 30 ~ 55 ◦ C
Pressure 0.2 ~ 17 MPa
Iron carbonate 0.85kmol/m3
Components Fluid, Mol.%
Natural gas Condensate Water

Methane 92.759 0.000 –

Ethane 3.130 1.082 –
Propane 0.565 2.265 –
i-Butane 0.103 1.292 –
n-Butane 0.118 2.930 –
i-Pentane 0.049 2.362 –
n-Pentane 0.030 2.436 –
n-Hexane 0.038 5.825 –
n-Heptane 0.030 17.784 –
n-Octane 0.000 16.471 –
n-Nonane 0.000 10.082 –
n-Decane 0.000 7.544 –
n-C11 0.000 5.725 –
n-C12 0.000 5.130 –
n-C13 0.000 4.012 –
n-C14 0.000 3.952 –
n-C15 0.000 2.738 –
n-C16 0.000 1.869 –
n-C17 0.000 1.577 –
n-C18 0.000 1.213 –
n-C19 0.000 0.930 –
n-C20 0.000 0.705 –
n-C21 0.000 0.515 –
n-C22 0.000 0.421 –
n-C23 0.000 0.333 –
n-C24 0.000 0.230 –
n-C25 0.000 0.184 –
n-C26 0.000 0.120 –
n-C27 0.000 0.091 –
n-C28 0.000 0.066 –
n-C29 0.000 0.052 –
n-C30 0.000 0.066 –
CO2 1.500 0.000 –
H2O 0.061 0.000 100.000
N2 1.618 0.000 –
Flow rate/cut 0.5 ~ 10.0 m/s 0 ~ 0.180 wt% 0 ~ 0.018 wt%

k μ
f = 0.001375 × [1 + (2000 + 106 m )0.33 ] (5)
D ρm um D

where um is the mixed velocity of gas and liquid (m/s), ρm is the mixed density of gas and liquid (kg/m3), μm is the mixed viscosity of gas
and liquid (Ns/m2), k is the pipe roughness (m), D is the pipe diameter (mm).
The pH of the solution can either be entered by a fixed valve or calculated using the method outlined in Reference [19]. The re­
ported pH in water analyses is usually measured after depressurization and atmospheric exposure of the sample. Such pH gives no
information about the in situ pH which is a requisite for the corrosion rate evaluation. In order to calculate the corrosion rate accu­
rately, the pH is chosen in situ calculating by HYSYS. Using this calculation method, the initial bicarbonate concentration, the ionic
strength, and whether or not the water is saturated with iron carbonate should be specified.
The presence of inhibitor such as glycol in the fluid can reduce the corrosion rate by a factor given by:
{ [ ] }
xgly < 0.95, logF gly = 1.6 log 100 × (1 − xgly ) − 2 (6)

xgly > 0.95, Fgly = 0.008 (7)

where xgly is the weight fraction of glycol. If liquid hydrocarbon is present in the pipeline, the corrosion rate may be inhibited.
The operation conditions used in the simulation of gathering process and the wet natural gas compositions is shown in Table 1. The
parameters of gathering pipelines employed in the simulation are listed in Table 2.

4
 M. Dai
Table 2
The parameters of simulated gathering pipelines.
specification Pipe coating
Outside diameter, Wall thickness, Length, Elevation, m Material Thermal conductivity, W/ Roughness, Material Wall thickness, Thermal conductivity, W/
mm mm km m.oC mm mm m.oC

DN50 60 6 3.6 0, Flat Mild 45 0.04572 Urethane 40 0.018

steel Foam
DN80 89 8 3.6 0, Flat Mild 45 0.04572 Urethane 40 0.018
steel Foam
DN100 114 10 3.6 0, Flat Mild 45 0.04572 Urethane 40 0.018
5

steel Foam
DN100 114 10 3.6 − 90, Concave Mild 45 0.04572 Urethane 40 0.018
steel Foam
DN100 114 10 3.6 20 ~ 100, Mild 45 0.04572 Urethane 40 0.018
Uphill steel Foam
DN150 168 15 3.6 0, Flat Mild 45 0.04572 Urethane 40 0.018
steel Foam
DN200 219 18 3.6 0, Flat Mild 45 0.04572 Urethane 40 0.018
steel Foam

Engineering Failure Analysis 122 (2021) 105265

M. Dai Engineering Failure Analysis 122 (2021) 105265

4. Results and discussion

Connecting with the actual construction situation and the influence factors described above, the effects of operational conditions,
inhibitors, pipe parameters and flow regime on the CO2 corrosion are studied in this paper.

4.1. Effects of operational conditions

4.1.1. Effects of operating pressure

The operating pressure influences CO2 corrosion directly by affecting CO2 partial pressure, pH, and shear stress. Fig. 3 shows the
CO2 partial pressure, pH, shear stress, and CO2 corrosion rate distribution curves for the variation of operating pressure. As shown in
Fig. 3(a), the CO2 partial pressure increases with increasing of the operating pressure, but decreases with the pipeline length. The
reason for this can be attributed to the pipeline pressure drop increases with the increasing of pipeline length. Fig. 3(b) shows the in
situ pH calculated by HYSYS increases with the operating pressure, and decreases with the pipeline length. The in situ pH has to be
determined by calculations which take into account the controlling buffer systems CO2 /HCO–3 and CH3COOH/CH3COO–. Fig. 3(c)
shows the results of shear stress along the pipeline with the different operating pressure. Shear stress is particularly associated with gas
velocity, density, and viscosity. The higher operating pressure usually leads to the higher density, and viscosity under given velocity,
hence the shear stress increases with operating pressure. The CO2 corrosion rate is a function of the shear stress, pH, temperature, and
CO2 partial pressure as described in equation (3). Due to the pH increasing with operating pressure, the CO2 corrosion rate curves along
the pipeline decrease with the increasing of operating pressure are observed in Fig. 3(d), although the shear stress increases with
operating pressure should increase CO2 corrosion rate. This indicates that pH play a central role in controlling the CO2 corrosion rate. It

Fig. 3. The results of CO2 partial pressure, pH, shear stress, and corrosion rate for gathering pipeline at different operating pressure (Flat,
DN100,45 ◦ C, Fgas = 5 m/s, Fcondensate = 0.0150 wt%, Fwater = 0.006 wt%).

6
M. Dai Engineering Failure Analysis 122 (2021) 105265

was also shown previously both experimentally and computationally by Tanupabrungsun et al. [21] that pH had a strong influence on
the corrosion rate. Higher pH leads to a reduction of the corrosion rate, which may be an attractive way of managing CO2 corrosion.
The drawback of this way is that it can lead to excessive scaling and can be rarely used with formation water systems. Furthermore, the
CO2 corrosion rate starts increasing sharply to reach maximum valve, subsequently the corrosion rate decreases with increasing
pipeline length. The reason for this may be concluded that the CO2 partial pressure decreasing sharply in the latter part of pipeline
leads to decreasing of CO2 corrosion rate.
It is noteworthy that in the case of scale-free CO2 corrosion, an increase of CO2 partial pressure typically leads to an increase in the
corrosion rate [22–23]. The commonly accepted explanation is that with the increase of CO2 partial pressure, the concentration of
H2CO3 increases, which will lead to a low pH and accelerate the cathodic reaction, and ultimately the corrosion rate [24]. However, a
different result is observed when there is 0.85kmol/m3 iron carbonate in solution as shown in Fig. 3(b), and increasing of CO2 partial
pressure can have a beneficial effect on decreasing of corrosion rate. At a high pH, higher CO2 partial pressure leads to an increase in
bicarbonate and carbonate ion concentration and a higher supersaturation, which accelerates precipitation and scale formation
[12,25]. To further study the pH and corrosion rate vary with CO2 partial pressure, the operating pressure is extended to a range from
0.2 MPa to 17.0 MPa. The effect of extended operating pressure on the corrosion rate in the presence of iron carbonate is illustrated in
Fig. 4. The Fig. 4(a) shows the pH decreases initially and then increases with the increase of CO2 partial pressure, and the minimum pH
appears at the operating pressure of about 2.0 MPa. For the results as shown in Fig. 4(b), a phenomenon of the corrosion rate increases
initially and then decreases with the increase of CO2 partial pressure is observed, but the maximum corrosion rate appears at the
operating pressure of about 10.0 MPa, not at 2.0 MPa. May be the shear stress causes the maximum corrosion rate move to high
operating pressure, because high operating pressure can enlarge shear stress, and which ultimately accelerates corrosion rate.

4.1.2. Effects of operating temperature

Operating temperature accelerates the corrosion rate in several ways, such as electrochemical, chemical, transport, etc. Temper­
ature accelerates the chemical reactions in the bulk solution and the electrochemical reactions at the steel surface by increasing re­
action rates. Temperatures can also speed up the mass transfer process by decreasing the viscosity of the solution. Previous studies
reported by literatures showed that the corrosion rate steadily increases with temperature [23,26], but these cases were studied at
static state and did not consider the pH and shear stress changed with operating temperature. In this paper, the results of pH, shear

Fig. 4. The effects of extended operating pressure on pH and corrosion rate (Flat, DN100, 45 ◦ C, Fgas = 5 m/s, Fcondensate = 0.0150 wt%, Fwater =
0.006 wt%).

7
M. Dai Engineering Failure Analysis 122 (2021) 105265

stress, and corrosion rate for gathering pipeline at different operating temperature are shown in Fig. 5. As shown in Fig. 5(a), the pH
increases with increasing of operating temperature, and the reason for this is increase of temperature will result in decrease of CO2
solubility, which will lead to a high pH. Fig. 5(b) shows that the shear stress also increases with the increasing of operating tem­
perature, especially at the 45 ~ 50 ◦ C, the shear stress increases sharply. This indicates that the sensitivity of operating pressure for
shear stress is around 45 ~ 50 ◦ C. Fig. 5(c) shows that CO2 corrosion starts increasing with increasing of operating temperature to reach
maximum corrosion rate of 2.35 mm/year at 45 ◦ C, subsequently the corrosion rate decreases with temperature increases to at 55 ◦ C.
For temperature below 45 ◦ C, corrosion rate increases with temperature because the effect of shear stress plays a leading role. But for
temperature over 45 ◦ C, the effect of pH increase will play a central role. Recent studies reported by literatures also found that at higher
pH, the precipitation kinetics would be accelerated by temperature increasing which aids the protective film formation, and which
could decrease the corrosion rate, the peak in the corrosion rate depended on water chemistry and flow conditions [27–28]. Hence, it
can conclude that the CO2 corrosion rate does not always increase with temperature, and which will reach a maximum value, and
subsequently decrease with temperature further increase.

4.1.3. Effects of gas velocity

Flow regime is another factor affects the process of corrosion rate in natural gas gathering pipelines. Flow regime shows the manner
in which fluid flows through a pipeline where flow type has direct impact on the wall shear stress. In this paper, the gas velocity varies
from 1 m/s to 8 m/s is simulated, and the results are shown in Fig. 6. The change of gas velocity will lead to changes in CO2 partial
pressure, pH, and shear stress. As shown in Fig. 6(a), the pipe pressure drop increases with increasing of gas velocity, and which causes
the CO2 partial pressure decrease. The Fig. 6(b) shows that the pH decreases with increasing of gas velocity, because the reduction in
CO2 partial pressure can increase the pH as mentioned above. The variation of shear stress with gas velocity is shown in Fig. 6(c). When
the gas velocity below 4 m/s, the shear stress increases slowly, and when the gas velocity above 4 m/s, the shear increases fast. The
Fig. 6(d) shows the variation of corrosion rate with increasing of gas velocity is similar to that of shear stress, and also a sensitive
increase is observed when the gas velocity above 4 m/s. The reason for corrosion rate increase with gas velocity can attribute to that
the increasing of gas velocity makes the pH decrease, and the shear stress increase, and low pH and high shear stress can lead to a high

Fig. 5. The results of pH, shear stress, and corrosion rate for gathering pipeline at different operating temperature (Flat, DN100, 17 MPa, Fgas = 5
m/s, Fcondensate = 0.0150 wt%, Fwater = 0.006 wt%).

8
M. Dai Engineering Failure Analysis 122 (2021) 105265

Fig. 6. The results of CO2 partial pressure, pH, shear stress, and corrosion rate for gathering pipeline at different gas velocity (Flat, DN100, 45 ◦ C,
17 MPa, Fcondensate = 0.0150 wt%, Fwater = 0.006 wt%).

corrosion rate.
Moreover, higher flow velocity usually means high turbulence and effective mixing in the solution [29]. A large number of
experimental studies showed that increased turbulent flow accelerates the corrosion species both toward and away from the metal
surface, and this may result in an increase in the corrosion rate when the mass transfer was the rate controlling factor and no corrosion
film was formed at the metal surface [12]. Zhang [30] investigated the effect of varies flow velocity on the corrosion rate of L360QS
steel in H2S/CO2 environments with sulphur deposition and concluded that the corrosion rate increased by about 10 times when
velocity increased from 3 to 7 m/s. In this paper, the results also indicate that high gas velocity in favor of high corrosion rate.

4.1.4. Effects of water cut

The water cut of gas–water flow has a significant effect on the corrosion rate. The results of pH and corrosion rate for gathering
pipeline at different water cut are shown in Fig. 7. No corrosion is observed in Fig. 7(b), when no free water present in pipe. The
corrosion rate increases with the water cut increases from 0.003 wt% to 0.006 wt%, and then decreases with the water cut further
increase. This may be due to the CO2 solubility varies with the water cut. As shown in Fig. 7(a), the CO2 partial pressure and pH
increase with the water cut. It can conclude from the Fig. 7 that to avoid the CO2 corrosion, the best way is separating the free water
from the wet natural gas before it gets into the gathering pipelines, and less free water present in pipeline may lead to more serious
corrosion than that of a large amount of free water.

4.2. Effects of inhibitors

4.2.1. Corrosion inhibition by liquid hydrocarbon

It is generally believed that the corrosion rate may be inhibited by the liquid hydrocarbon present in the pipeline. The first is a

9
M. Dai Engineering Failure Analysis 122 (2021) 105265

Fig. 7. The results of pH and corrosion rate for gathering pipeline at different water cut (Flat, DN100, 45 ◦ C, 17 MPa, Fgas = 5 m/s, Fcondensate =
0.0150 wt%).

Fig. 8. The result of corrosion rate for gathering pipeline at different condensate cut (Flat, DN100, 45 ◦ C, 17 MPa, Fgas = 5 m/s, Fwater = 0.006
wt%).

10
M. Dai Engineering Failure Analysis 122 (2021) 105265

wettability effect and relates to a hydrodynamic condition where liquid hydrocarbon entrains the water and prevents it from wetting
the steel surface (continuously or intermittently) [12]. The second effect is corrosion inhibition by certain components of the liquid
hydrocarbon that reaches the steel surface either by direct contact or by first partitioning into the water phase [12]. In order to study
the effect of liquid hydrocarbon on the corrosion rate, different condensate cut was simulated. As shown in Fig. 8, when the condensate
cut below 0.015 wt%, the increasing of condensate cut can help decrease the corrosion rate. But it is worth noting that too much
condensate cut may increase corrosion instead, because the shear stress will increase obviously with a large number of condensate
causes the liquid velocity increase.

4.2.2. Corrosion inhibition by glycol

Methanol, ethylene glycol and triethylene glycol are often added to gathering pipelines system in order to prevent hydrates from
forming. In the very few studies available it has been assumed that the main ‘‘inhibitive’’ effect of ethylene glycol and methanol on
corrosion comes from dilution of the water phase, which leads to a decreased activity of water [12]. As shown in Fig. 9, the low cut of
ethylene glycol cannot inhibit corrosion obviously, and the corrosion rate will rapidly reduce to 0 when the cut of ethylene glycol
reaches to 72%. Fig. 9 shows the comparison of methanol, ethylene glycol and triethylene glycol inhibit corrosion, and the ethylene
glycol is best, the methanol and triethylene glycol need cuts of 92% and 87% respectively. The result of corrosion inhibition by glycol
shows glycol is useful in inhibition, but it is not an economic way, especially when there is a large amount of free water in pipeline. This
also indicates that it is necessary to develop high efficient inhibitors for CO2 corrosion.

4.3. Effects of pipe parameters

As discussed in above, the corrosion rate is related to shear stress. The pipe parameters usually have significant effect on the shear
stress of pipe at the same operating conditions. Hence, it is necessary to study the pipe parameters such as diameter and elevation.

4.3.1. Effects of pipe diameter

The results of corrosion rate and shear stress for gathering pipeline at different pipe diameter are shown in Fig. 10. As shown in
Fig. 10(a), the shear stress decreases with increasing of pipe diameter, and the value changes sensitively when the pipe diameter at the
range from DN50 to DN 100. The corrosion rate vary with pipe diameter is similar to that of the shear stress, and a sharp increase of
corrosion rate is also observed in Fig. 10(b) when the pipe diameter at the range of DN50 to DN 100. The results indicate that a large
pipe diameter is beneficial for decreasing of corrosion rate, but too large pipe diameter is uneconomical.

4.3.2. Effects of pipe elevation

Pipe elevation can cause static pressure difference in the pipeline, which will affect the mass density and velocity of fluid in
pipeline. As described in the Eq. (4), the shear stress is related to mixed mass density and velocity, hence, pipe elevation is another
factor that affecting the corrosion rate of pipe. In this paper, the elevation difference of pipe with range from 0 m to 100 m is studied,
and the results are shown in Fig. 11. The Fig. 11(b) shows that the corrosion rate of pipe with 0 m elevation (flat pipe) is the lowest, and
which increases with the pipe elevation is not obvious. The reason for this can be attributed to the shear stress increases with elevation
difference is insensitive as shown in Fig. 11(a). The Fig. 11(c) also shows the mixed mass density and velocity of gas and liquid have no
significant change along with the elevation increasing. It should be noted that the effect of elevation difference on corrosion rate is less
than that of pipe diameter.

Fig. 9. The results of corrosion inhibition by glycol (Flat, DN100, 45 ◦ C, 17 MPa, Fgas = 5 m/s, Fcondensate = 0.0150 wt%, Fwater = 0.006 wt%).

11
M. Dai Engineering Failure Analysis 122 (2021) 105265

Fig. 10. The results of corrosion rate and shear stress for gathering pipeline at different pipe diameter (Flat, 45 ◦ C, 17 MPa, Fgas = 5 m/s, Fcondensate
= 0.0150 wt%, Fwater = 0.006 wt%).

4.4. Effects of flow regime

One of the more complex problems is the effect of flow regime on CO2 corrosion since the pipelines carrying gas, condensate and
water are operating under three-phase flow conditions. The flow patterns can be found in wet natural gas gathering pipelines are
usually the slug and annular flow. In the liquid phase, water and condensate can flow separated or mixed with either phase being
continuous with the other flowing as a dispersed phase. Different flow patterns lead to a variety of steel surface wetting mechanisms
which greatly affect corrosion [12]. In the absence of protective scales multiphase flow can lead to very high fluctuating mass transfer
rates (particularly in slug flow) which, in turn, can affect corrosion. Associated with this are even larger fluctuations of the surface
shear stress which can lead to removal of protective scale and/or inhibitors. In order to study how the flow regime affects the corrosion
rate, a concave pipeline with − 90 m elevation is simulated. The slug flow is observed when the condensate cut reaches to 0.168 wt%,
and the flow pattern is always annular flow when the condensate cut below 0.168 wt%. Hence, the condensate cuts of 0.168 wt% and
0.167 wt% are used to simulate slug and annular flow respectively, and the number difference is very small, which can be ignored the
effect of liquid velocity difference on the shear stress. As shown in Fig. 12(a), the slug flow appears in the pipeline range from 2800 m to
3600 m, and the max slug length is about 0.5 m. The Fig. 12(b) and Fig. 12(c) show that the shear stress and corrosion rate will increase
sharply when the slug appears, and because the slug forms as pulse, the increase of shear stress and corrosion should be a region instead
of an exact value. The reason for the increase of shear stress and corrosion rate can be attributed to the gas velocity will increase sharply
with the slug forms, and which will cause the shear stress increase. It can conclude that the slug flow will promote the CO2 corrosion,
and trying to avoid forming the slug flow in gathering pipeline is very necessary.

4.5. Simulation validation with field corrosion rate data

The corrosion rate simulated by HYSYS in this study was validated with the field data from a natural gas gathering pipeline located
in Tarim oilfield, Xinjiang, China. The operating conditions of the natural gas gathering pipeline are as follow: operating pressure 15.5

12
M. Dai Engineering Failure Analysis 122 (2021) 105265

Fig. 11. The result of corrosion rate and shear stress for gathering pipeline at different pipe elevation (DN100, 45 ◦ C, 17 MPa, Fgas = 5 m/s,
Fcondensate = 0.0150 wt%, Fwater = 0.006 wt%).

~ 16.1 MPa, operating temperature 30 ~ 45℃, inlet gas velocity 2.8 ~ 4.2 m/s, condensate cut 0.012 wt%, water cut 0.005 wt%, and
CO2 in natural gas is 0.89 mol.%. The natural gas gathering pipeline parameters are as follow: outside diameter 219 mm, wall thickness
18 mm, mild steel material, 12560 m long, and 67 m elevation difference. The coating material and thickness are urethane foam and
40 mm, respectively. The field corrosion rate data was obtained by ultrasonic inspection for the natural gas gathering pipeline through
an intelligent pig during annual maintenance. As the operating conditions were changing with the time, hence, maximum corrosion
and minimum corrosion cases were simulated in this work. The simulation validation with field corrosion rate data is shown in Fig. 13.
The corrosion rate simulated by HYSYS is higher than the field corrosion rate data from 0 m to 1500 m along the pipeline, and the
simulated results are in good agreement with the field corrosion rate data after 1500 m along the pipeline. It should be noted that the
corrosion variations trend along the pipeline simulated by HYSYS is consistent with the field corrosion rate data, and this indicates that
the HYSYS is proved an accurate tool for determining the corrosion rate of CO2 for the natural gas gathering pipeline.

5. Conclusion

Extensive simulation study has been performed on the CO2 corrosion rate into wet natural gas gathering system. In situ parametric
study stated to check the effects of different factors on the CO2 corrosion rate and based on the results, the following conclusions can be
drawn:
Presence of CO2 and water as in natural gas that can causes corrosion in the gathering pipelines with high pressure depends on the
operational conditions. Increasing operating pressure leads to increase the pH and decrease the CO2 corrosion rate when the operating
pressure over 10 MPa and the solution contents 0.85kmol/m3 iron carbonate. The CO2 corrosion rate does not always increase with
temperature, and which will reach a maximum value at 45 ◦ C, and subsequently decrease with temperature further increase. A high gas

13
M. Dai Engineering Failure Analysis 122 (2021) 105265

Fig. 12. The simulation results for effects of multiphase flow on corrosion rate and shear stress (Concave, DN100, 45 ◦ C, 17 MPa, Fgas = 5 m/s,
Fcondensate = 0.167/0.168 wt%, Fwater = 0.006 wt%).

Fig. 13. Simulation validation with field corrosion rate data (Maximum corrosion case: 45℃, 16.1 MPa, Fgas = 4.2 m/s; Minimum corrosion case:
30℃, 15.5 MPa, Fgas = 2.8 m/s).

velocity will lead to a high corrosion rate. To avoid the CO2 corrosion, the best way is separating the free water from the wet natural gas
before it gets into the gathering pipelines, and less free water present in pipeline may lead to more serious corrosion than that of a large
amount of free water.

14
M. Dai Engineering Failure Analysis 122 (2021) 105265

The corrosion rate may be inhibited by liquid hydrocarbon and glycol.

The pipe parameters usually have significant effect on the shear stress of pipe at the same operating conditions, and the large
diameter and flat pipe are useful for decreasing CO2 corrosion rate.
The shear stress and corrosion rate will increase sharply when the slug appears, and because the slug forms as pulse, the increase of
shear stress and corrosion should be a region instead of an exact value.

Declaration of Competing Interest

The authors declared that there is no conflict of interest.

References

[1] T. Hea, I.A. Karimia, Ju Yonglin, Review on the design and optimization of natural gas liquefaction processes for onshore and off shore applications, Chem. Eng.
Res. Des. 132 (2018) 89–114.
[2] Y. Hua, X.u. Shusheng, Y. Wang, W. Taleb, J. Sun, L. Zhang, R. Barker, A. Neville, The formation of FeCO3 and Fe3O4 on carbon steel and their protective
capabilities against CO2 corrosion at elevated temperature and pressure, Corros. Sci. 157 (2019) 392–405.
[3] M. Javidi, S. Bekhrad, Failure analysis of a wet gas pipeline due to localised CO2 corrosion, Eng. Fail. Anal. 89 (2018) 46–56.
[4] D.A. López, T. Pérez, S.N. Simison, The influence of microstructure and chemical composition of carbon and low alloy steels in CO2 corrosion– A state of art
appraisal, Material and Design 24 (2003) 561–575.
[5] M. Wang, A. Lawal, P. Stephenson, J. Sidders, C. Ramshaw, Postcombustion CO2 capture with chemical absorption: a state–of–the–art review, Chem. Eng. Res.
Des. 89 (2011) (2011) 1609–1624.
[6] M. Asadian, M. Sabzi, S.H. Mousavi Anijdan, The effect of temperature, CO2, H2S gases and the resultant iron carbonate and iron sulfide compounds on the sour
corrosion behaviour of ASTM A–106 steel for pipeline transportation, International Journal of Pressure Vessels and Piping, 171(2019) 184–193.
[7] F.F. Eliyan, A. Alfantazi, Re–examining the influence of chloride ions on electrochemical CO2 corrosion of pipeline steels–corrosion of the heat affected zones
(HAZs) of API–X100 steel, The Canadian Journal of Chemical Engineering 3 (2015) 1044–1052.
[8] R. Elgaddafi, A. Naidu, R. Ahmed, S. Shah, S. Hassani, S.O. Osisanya, A. Saasen, Modeling and experimental study of CO2 corrosion on carbon steel at elevated
pressure and temperature, J. Nat. Gas Sci. Eng. 27 (2015) 1620–1629.
[9] Srdjan Nešić, Kun–Lin John Lee, Vukan Ruzic, A mechanistic model of iron carbonate film growth and the effect on CO2 corrosion of mild steel. CORROSION,
NACE International, Houston/TX, 2002(Paper No. 237).
[10] G. Schmitt, M. Horstemeier, Fundamental aspects of CO2 metal loss corrosion–part II: influence of different parameters on CO2 corrosion mechanisms,
CORROSION, NACE International, Houston/TX, 2006 (Paper No. 112).
[11] H. Mansoori, D. Young, B. Brown, S. Nesic, M. Singer, Effect of CaCO3-saturated solution on CO2 corrosion of mild steel explored in a system with controlled
water chemistry and well-defined mass transfer conditions, Corros. Sci. 158 (2019), 108078.
[12] S. Nešić, Key issues related to modelling of internal corrosion of oil and gas pipelines–a review, Corros. Sci. 49 (2007) 4308–4338.
[13] C. de Waard, U. Lotz, D. E. Milliams, Predictive Model for CO2 Corrosion Engineering in Wet Natural Gas Pipelines, CORROSION, NACE International, Houston/
TX, (Paper No. 19).
[14] William P. Jepson, Samir Bhongale, Madan Gopal, Model for sweet corrosion in horizontal multiphase slug flow, CORROSION, NACE International, Houston/
TX, 1996 (Paper No. 11).
[15] H. Wang, J.–Y. Cai, W.P. Jepson, CO2 corrosion modelling and prediction of horizontal slug flow, CORROSION, NACE International, Houston/TX, 2002 (Paper
No. 238).
[16] S. Nešić, Effects of multiphase flow on internal CO2 corrosion of mild steel pipelines, Energy Fuels 26 (2012) 4098–4111.
[17] Srdjan Nešić, Jiyong Cai, Kun–Lin John Lee, A multiphase flow and internal corrosion prediction model for mild steel pipelines, Corrosion, NACE International,
Houston/TX, 2005 (Paper No. 556).
[18] Muhammad Hashim Abbasa, Rosemary Normana, Alasdair Charles, Neural network modelling of high pressure CO2 corrosion in pipeline steels, Process Safety
and Environmental Protection, 119 (2018) 36–45.
[19] NORSOK, NORSOK standard M–506, CO2 corrosion prediction model, Strandveien (2005).
[20] Ammar Ali Abd, Samah Zaki Naji, Atheer Saad Hashim, Failure analysis of carbon dioxide corrosion through wet natural gas gathering pipelines, Engineering
Failure Analysis, 105 (2019) 638–646.
[21] T. Tanupabrungsun, B. Brown, S. Nešić, Effect of pH on CO2 corrosion of mild steel at elevated temperatures, CORROSION, NACE International, Houston/TX
(2013). Paper No. 48.
[22] Y. Zhang, X. Pang, Q.u. Shaopeng, X. Lib, K. Gao, Discussion of the CO2corrosionmechanism between low partial pressure and supercritical condition, Corros.
Sci. 59 (2012) 186–197.
[23] E.E. Okoroa, A.M. Kurah, S.E. Sanni, A. Dosunmu, E.B. Ekeinde, Flow line corrosion failure as a function of operating temperature and CO2 partial pressure using
real time field data, Eng. Fail. Anal. 102 (2019) 160–169.
[24] C. Plennevaux, N. Ferrando, J. Kittel, M. Frégonèse, B. Normand, T. Cassagne, F. Ropital, M. Bonis, pH prediction in concentrated aqueous solutions under high
pressure of acid gases and high temperature, Corros. Sci. 73 (2013) 143–149.
[25] P. Sui, J. Sun, et al., Effect of temperature and pressure on corrosion behavior of X65 carbon steel in water–saturated CO2 transport environments mixed with
H2S, Int. J. Greenhouse Gas Control 73 (2018) 60–69.
[26] M. Asadian, M. Sabzi, S.H. Mousavi Anijdan, The effect of temperature, CO2, H2S gases and the resultant iron carbonate and iron sulfide compounds on the sour
corrosion behaviour of ASTM A–106 steel for pipeline transportation, Int. J. Press. Vessels Pip. 171 (2019) 184–193.
[27] Yoon–Seok Choi, Fernando Farelas , Srdjan Nešić, Alvaro Augusto O. Magalhães, Cynthia de Azevedo Andrade, Corrosion behavior of deep water oil production
tubing material undersupercritical CO2 environment: part 1–effect of pressure and temperature, CORROSION, NACE International, Houston/TX, 2013, (Paper
No. 80).
[28] J. Armando Garnica-Rodriguez, Genesca, Juan Mendoza-Flores, Ruben Duran-Romero, Electrochemical evaluation of aminotriazole corrosion inhibitor under
flow conditions, J. Appl. Electrochem. 39 (2009) 1809–1819.
[29] Sobhan Hatami, Alireza Ghaderi-Ardakani, Mohammad Niknejad-Khomami, Farzan Karimi-Malekabadi, Mohammad R. Rasaei, Amir H. Mohammadi, On the
prediction of CO2 corrosion in petroleum industry, The Journal of Supercritical Fluids 117 (2016) 108–112.
[30] Naiyan Zhang, Dezhi Zeng, Zhi Zhang, Wentao Zhao, Guangju Yao, Effect of flow velocity on pipeline steel corrosion behaviour in H2S/CO2 environment with
Sulphur deposition, Corros. Eng., Sci. Technol. 53 (2018) 370–377.

15