Paper No.

CORROSION2004
04381
LOCALIZED CORROSION IN MULTIPHASE PIPELINES

Cheolho Kang, Madan Gopal and William Paul Jepson

CC Technologies
6869 Stapoint Court, Suite 111
Winter Park, FL 32792

ABSTRACT

Measurements of the localized flow patterns, flow characteristics, and corrosion rates using
high-precision, non-intrusive flow and corrosion/erosion ultrasound system have been made at various
locations along the field pipeline.
The field pipeline is schedule 40, a 20 cm diameter, multiphase carbon steel flowline which
transports oil, water and some gas. The total elevation change is about 19 m.
The flow pattern and its characteristics (e.g. height of the liquid film) were monitored and
corresponding high levels of corrosion were measured there.
The flow regime changed dramatically along the pipeline due to the nature of the topography of
field pipeline. It is noticed that there are slugs present in the pipe even though the pipe is downwardly
declined. .
Slug flow was noticed in the initial sections of the pipe and these were formed due to the dip in
the pipe there. The corrosion rate here was between 0.5 and 0.8 mm/year.
This paper describes the localized flow and corrosion in hilly terrain field pipelines.

Keywords: corrosion rates, flow pattern, localized flow, flow characteristics, ultrasound, hilly terrain,

Copyright
2004 by NACE International. Requests for permission to publish this manuscript in any form, in part or in whole must be in writing to NACE
International, Publications Division, 1440 South Creek Drive, Houston, Texas 77084-4906. The material presented and the views expressed in this paper are
solely those of the author(s) and not necessarily endorsed by the Association. Printed in U.S.A.

1
Publication Right
Government work published by NACE International with permission of the author(s). Requests for permission to publish this manuscript in any form, in part
or in whole must be made in writing to NACE International, Publications Division, 1440 South Creek Drive, Houston, Texas 77084-4906. The material
 INTRODUCTION

Co-current two and three-phase flows are encountered frequently in the oil and gas industry.
The existence of multiphase flow and its importance to industrial units has prompted extensive
research in this field. This type of flow is seen in pipelines, oil and gas producing wells and associated
flow lines, separators, dehydration units, evaporators and other processing equipment. The nature of
multiphase flow is extremely complicated due to the existence of various flow patterns and different
mechanisms governing them.
The multiphase mixture is transported through a single pipeline to a central gathering station.
During this transport, several flow regimes occur depending on the gas and liquid flow rates. The
distances the multiphase mixture must be transported are often long and the deviations from horizontal
flow are always present. These changes in inclination cause changes in the flow regime transitions and
flow characteristics, which have a definite effect on the corrosion rate experienced by these pipelines.
Therefore, it is important to understand the nature and behavior of flow in multiphase systems since
they are crucial to almost all aspects of multiphase corrosion.
Limited flow regime data exists for inclined pipes. Gould, et. al. (1974)1 introduced +45o and
+90 flow regime maps. Govier, et. al. (1972)2 presented a commonly used method of establishing flow
o

regimes for inclined pipes. Barnea, et. al. (1980, 1982)3~4 presented flow regime maps for two phase
flow in ± 0 ~ 10o and downward (0 ~ 90o) small diameter pipes. He also proposed a model for
predicting transitions in downward pipes. Stanislav, et. al. (1986)5 reported inclined flow pattern as
compared with a modified Taitel and Dukler model6 in small diameter pipes. Kokal and Stanislav
(1989)7 have characterized extensively the upward and downward flow patterns. However, all of these
studies have been carried out in small diameter pipes. Flow regime maps have been determined in ±2o,
±15o and ±30o inclinations in large diameter pipes by Kang and Jepson (2002)8.
Corrosion can be broadly categorized into uniform corrosion, localized corrosion and erosion
corrosion. Uniform or general corrosion causes overall metal loss and general thinning of metal.
Localized corrosion has the appearance of pits or grooves. Erosion corrosion is described as the
deterioration of the metal surface due to relative motion between the metal wall and the corrosion fluid
(Jones, 1991)9.
The corrosion rate had been measured using traditional intrusive technology such as coupons,
electrical resistance probes (ER), linear polarization resistance probes (LPR), electrochemical
impedance spectroscopy (EIS), and electrochemical noise techniques (ECN). These technologies
involve the insertion of the probes through drilled taps and these are usually located at sites where the
pipe is readily accessible. Consequently, since most of the corrosion/erosion is localized, there may be
little corrosion taking place at the probe location and little useful information is gathered. The
provision of the taps in all of the locations to be examined is intractable. Therefore, it is very important
that these techniques should be used at critical points.
Corrosion inhibitors are used in the industry to reduce corrosion in the pipeline. Corrosion
inhibitors are substances containing organics that adsorb to the metal surface of the pipeline and form a
protective film to prevent corrosion. The effectiveness of the inhibitor depends on the composition of
the pipeline material, the inhibitor composition, the type of flow, and operating conditions. Some of
the inhibitors are not effective because of the flow conditions. Hence, while selecting an inhibitor,
knowledge of the metal surface conditions, operating conditions, fluid properties, and the flow
conditions are essential.

2
 There are many equations and models that give predictions of corrosion rates. These include
the DeWaard et al.10, Nyborg et al.11, Dayalan et al.12, Jepson et al.13, Pots14, and Gunaltun15. It is not
possible to use the same model or correlation for the prediction of the corrosion for both gas lines and
oil lines or for wells and flow lines due to the fact that corrosive mediums and flow regimes are quite
different.16 For each case, parameters to be used need to be identified and calculated separately for the
specified conditions.
This paper describes the measurements of flow characteristics in multiphase field pipelines
such as flow pattern, height of the liquid film and corrosion rates using high precision, non-intrusive
flow and corrosion/erosion ultrasound system. The localized flow and corrosion are identified.

HIGH-PRECISION, NON-INTRUSIVE MULTIPHASE FLOW AND CORROSION/EROSION

ULTRASOUND TECHNIQUES

The method used for this study is based on ultrasound and a simplified description is supplied
here. This utilizes an array of ultrasound transducers that are non-intrusively placed on the outside of
the pipe wall. For low liquid flowrates 4-6 transducers are needed but for the current work, which
involves high liquid volumes, only 2 are required. This is outlined in Figure 1.
The upper or lower transducers detect the presence of liquid and from this, the flow regime is
determined. Then based on a cross-correlation technique, the velocity of the liquid flow can be
calculated. In a cross-correlation technique (Figure 1), tow pairs of transducers are placed on the pipe,
separated by a known distance L and the media is interrogated continuously. The cross-correlation
function16 is defined as,

1 ∆t
R=
∆t ∫
0
( x t − τ y t )dt (1)

where,
∆t = time interval in which interrogation is carried out (s)
τ = time delay that is progressively varied in steps (s)
xt-τ = upstream signal at time t-τ
yt = downstream signal at time t

The cross-correlation function R reaches a maximum when the time delay τ has a value τ *
which is the time taken for the tag to move from upstream to downstream. The velocity of the fluid Vf
(m/s) is then given by

L
Vf = (2)
τ*

3
 A single transducer at the bottom of the pipe measures the thickness of the liquid using a
“pulse-echo” technique as shown in Figure 2. This is a well known phenomenon where the ultrasonic
pulse is passed through the pipe wall and into the liquid. It then travels to the gas/liquid interface and is
reflected back towards the wall. The signal is collected at the same transducer. From these
measurements, the flowrate at any location along the pipeline can be calculated.
In a similar way, the corrosion rate can be monitored as shown in Figure 3. By a suitable choice
of a different transducer, the thickness of the pipe wall can be measured accurately (measurement
precision: ±0.00125 mm). Depending on the level of corrosion, the measurement can again be taken at
a later time and from the difference in thickness and the time period, the corrosion rate is calculated.
This is illustrated in Figure 4.
Fracture of materials (e.g. vessels and tanks) and sludge can be also detected by this ultrasound
system.

RESULTS AND DISCUSSIONS

Topography of the Flowline

A simplified, two-dimensional graph of the topography is shown in Figure 5 and it indicates the
changes in vertical position with distance along the line. The field pipeline is schedule 40, a 20 cm
diameter, unlined, multiphase carbon steel flowline which transports oil, water and some gas. The
liquid phase consists of about 90% water and 10% oil. The gas flowrate was not known but a ballpark
estimate of 1 SCF/barrel of liquid was commonly accepted as being reasonable. The total elevation
change is about 19 m and includes an initial decrease from about 230 m to 211 m. The average
pressure and temperature in the line are approximately 0.3 MPa and 60 oC, respectively.
At the beginning of the pipeline, before section B (about 1,200 m along the line) as shown in
Figure 5, about 1590 m3/day is flowing but a further 636 – 795 m3/day flow into the line from section
B.
It should be noted from Table 1, that before Section B, the superficial gas velocity is below 0.1
m/s and the superficial liquid velocity is less than 0.6 m/s. As mentioned earlier, this would suggest
that the 20 cm diameter pipe is too large for this flowrate and terrain induced intermittent flows are
likely to occur. After Section B, the superficial liquid velocity is higher at about 0.9 m/s but is still low
enough to allow some localized changes in the ensuing sections.
Due to the terrain changes, localized measurements are needed to identify the flow
characteristics and the corresponding corrosion rates. The measurement locations are also included in
Figure 5.
Since the topography changes continuously along the pipeline, a knowledge of the effect of
inclination on the flow regime is needed. If the pipeline changes in elevation along its length, localized
flows occur that cannot be predicted by flow regime prediction software. Typical examples are given
below.
Figure 6 shows the flow regimes at a change of inclination as the pipe goes from downward to
horizontal, and then to upward inclinations. It indicates that the flow in the upward section is slug flow
even if the downward flow is stratified. However, this flow pattern may only occur for about 10 – 20
m. Here, since gas is present, the pipe cross-section is not completely full of liquid. There is usually

4
insufficient momentum in the gas phase to carry all the liquid away from the low spot. Consequently,
the liquid moves away from the base of the pipe as a slug. The slug then dissipates and the liquid “rolls
back” toward the low spot. This thickens the liquid layer there and this allows more slugs to form. This
causes a local “hot spot” for enhanced corrosion. Similar flows would occur if the downward flow was
plug flow.
In addition to the above, in low spots, the liquid can accumulate and form a temporary blockage
where the flowrate decreases or even goes to zero. The incoming fluids build up the pressure behind
the blockage and, eventually, this pressure is high enough to force out all of the blockage. This sends a
long length, surge of liquid and gas mixture along the pipe. This can lead to much higher flowrates of
liquid than usually anticipated, albeit for a short period of time. The local velocity of the liquid phase
(to be used in a corrosion rate prediction) can be much higher than the average value calculated from
the average flowrates.
Similar events can occur when the pipeline has a high spot as shown in Figure 7. Here, at the
initiation of the flow, gas can collect and stay there even though a steady flow can be produced from
the pipeline. Thus, locally, much more gas is present than would be predicted from the average
flowrates.
The formation of localized flow is also enhanced by the use of oversized pipes. The dips act as
separators thus increasing the holdup of liquid there.

Flow and Corrosion Measurements

Due to the nature of the topography of field pipeline and the flowrates described above, it is
necessary to monitor the flow and corrosion at certain “hot spots” along the pipe. At these sites, two
sets of ultrasonic probes were placed on the line. The first set is as shown in Figure 1 and the flow
characteristics are determined. These include the flow pattern, local liquid distribution, i.e. liquid film
velocity and height, and the velocity of the pulses delivered from the pump. The second transducer is
used as a corrosion probe to measure the corrosion rate at this fixed location.
At the first location, site 1, there is low point in the line similar to that described in Figure 6. It
is caused by a downward inclination of about 2 degrees followed by an upward inclined section at
about 2 degrees. Both flow and corrosion were monitored at fifteen minute intervals using the
techniques described earlier.
The corrosion rate here was measured between 0.5 ~ 0.8 mm/year and the results for the flow
characteristics are presented in Table 2. It is first noted that the liquid film height is approximately 10
cm. This was much lower than expected. Since the gas flowrate is low, the pipe should be almost full
of liquid and the liquid film height should be at least 18 cm. It can be explained as follows. The low
point in the line fills with liquid and, consequently, in the downward section preceding the dip, gas
separates and accumulates there, thus decreasing the thickness of the liquid layer. The velocity of the
layer was between 0.9 and 1.0 m/s. This corresponded to an overall flowrate ranging between 1,351
and 1,495 m3/day and is very close to the expected average of 1,590 m3/day.
Even though the pipe is downwardly inclined, it is also noticed that there are slugs present in
the pipe and these have a frequency of 4 per minute. These slugs are formed at, or close to, the
injection point of pipeline and are initiated by the flow fluctuations.

5
 Following the line to position site 2, there is an incline of about 1.5 degrees. When the flow
was monitored here, it was again found that slugs were present. However, due to the upward
inclination, the film had thickened to between 13 and 15 cm and the frequency had increased from 4 to
6 slugs per minute. The film velocity was lower at 0.5 to 0.8 m/s and the former corresponded to only a
flowrate of 1018 m3/day. This will be explained later by examining the flow in the “gully”. The
corrosion rate here was around 0.8 mm/year.
The flow characteristics were then monitored at location site 3, which is on the uphill section
of the pipe exiting the gully. Measurements of the flow were taken every fifteen minutes and the
results are presented in Table 3. It can clearly be seen that the flowrate changes dramatically with time.
Initially, a flowrate of 1,654 m3/day was measured but later this increased to 2,544 m3/day. It was then
noted that the flowrate became zero before beginning to flow again. This cycle was repeated which
indicates that the gully is being filled with liquid with no flow exiting and, with time, the pressure
builds up behind the blockage. Eventually, the pressure is large enough to force out all the liquid that
has accumulated in the gully, giving the high liquid flowrate. The process then repeated in a cyclic
manner.
At location site 4, the flowrate again fluctuates with values of 1,495 to 2,226 m3/day being
measured as indicated in Table 4. The corrosion rate was measured at about 0.8 mm/year. The flow
pattern is wavy stratified flow and the liquid film is now between 15 and 16.5 cm thick with a velocity
ranging from 0.6 to 1.0 m/s
The most severe corrosion problem was occurring at site 5 as shown in Table 5. This was a
short section of pipe where the inclination was sharply downward. The flow measurements indicated
that a maximum of 2,862 m3/day was flowing here. The liquid film height was between 9.0 and 10 cm
with a film velocity of 3.0 m/s. The corrosion rate was measured in two locations and the values were
2.1 and 3.0 mm/year. At these higher velocities, it is likely that the corrosion product film would be
thinner and less protective. It was not possible to obtain the thickness of actual corrosion product layer.

CONCLUSIONS
Measurements of the localized flow patterns, flow characteristics, and corrosion rates using
high-precision, non-intrusive ultrasound system have been made at various locations along the field
pipeline.
The results show that localized flows similar to that illustrated in Figures 6 and 7 were present.
Slug flow was noticed in the initial sections of the pipe and these were formed due to the dip in the
pipe there. The corrosion rate here was between 0.5 and 0.8 mm/year. This was accounted by the
occurrence of a gully downstream. The gully and the following steep incline caused the flow to
become intermittent with the gully filling and then emptying in a cyclic manner.
After the pipeline distance of 1,150 m, the line proceeds downhill. Due to the intermittent flow
before the pipeline distance of 1,160 m, this downhill section acts as a separator and the liquid layer
accelerates. This was especially noticeable at location site 5. This caused an increase in the corrosion
rate there to 2.1 ~ 3.0 mm/year.
All of the above indicates the importance of the sizing of the pipe diameter. The presence of the
localized flows showed that, for the current flowrates, the 20 cm diameter of the pipeline is too large.

6
 REFERENCES
1. Gould, T. L., Tek, M. and Katz, D. L., “Two Phase Flow Through Vertical, Inclined, or Curved
Pipes,” J. Petrol. Tech., 26, 915-926, 1974.
2. Govier, G. W. and Aziz, K., The Flow of Complex Mixtures in Pipes, Van Norstrand Reinhold,
New York, 1972.
3. Barnea, D., Shoham, O., Taitel, Y., and Dukler, A. E., “Flow Pattern Transitions for Gas-Liquid
Flow in Horizontal and Inclined Pipes: Comparison of Experimental Data with Theory,” Int. J.
Multiphase Flow, 6, 217-225, 1980.
4. Barnea, D., Shoham, O. and Taitel, Y., “Flow Pattern Transition for Downward Inclined Two
Phase Flow; Horizontal to Vertical,” Chemical Engineering Science, 37, 735-740, 1982.
5. Stanislav, J.F., Kokal,S., and Nicholson, M. K., “Intermittent Gas-Liquid Flow in Upward Inclined
Pipes,” Int. J. Multiphase Flow, 12, 325-335, 1986.
6. Taitel, Y. and Dukler, A. E., “A Model for Slug Frequency During Gas-Liquid Flow in Horizontal
and Near Horizontal Pipes,” Int. J. Multiphase Flow, 3, 585-596, 1977.
7. Kokal, S. L. and Stanislav, J. F., “An Experimental Study of Two Phase Flow in Slightly Inclined
Pipes-I. Flow Patterns,” Chemical Engineering Science, 44, 665-679, 1989.
8. Kang, C. and Jepson, W. P., “Flow Regime Transitions in Large Diameter Inclined Multiphase
Pipelines,” NACE International Conference and Exhibition, Denver, CO, Paper No. 02243, April
2002.
9. Jones, D.A., Principles and Prevention of Corrosion, Macmillan Publishing Company, New York,
1991.
10. De Waard, C., Lotz, U., and Dugstad, A., “Influence of Liquid Flow Velocity on CO2 Corrosion: A
Semi-Empirical Model,” NACE International Conference and Exhibition, Orlando, FL, Paper No.
128, 1995.
11. Nyborg, R., Anderson, P., and Nordsveen, M., ”Implementation of CO2 Corrosion models in a
Three-Phase Fluid Flow Model,” NACE International Conference and Exhibition, Orlando, FL,
Paper No. 48, 2000.
12. Daylan, E., Vani, G., Shadley, J. R.,Shirazi, S. A., and Rybicki, E.F., “Modeling CO2 Corrosion of
Carbon Steels,” NACE International Conference and Exhibition, Orlando, FL, Paper No. 118,
1995.
13. Jepson, W. P., Stitzel, S., Kang, C., and Gopal, M., “Model for Sweet Corrosion in Horizontal
Multiphase Slug Flow,” NACE International Conference and Exhibition, New Orleans, LA, Paper
No. 11, 1997.
14. Pots, B. F. M., “Mechanistic Models for the Prediction of CO2 Corrosion Rates Under Multiphase
Flow Conditions,” NACE International Conference and Exhibition, Orlando, FL, Paper No. 137,
1995.
15. Gunaltun, Y. M., “Combine Research and Field Data for Corrosion Rate Prediction,” NACE
International Conference and Exhibition, Denver, CO, Paper No. 27, 1996.
16. Xu, L. A., Yang, H. L., Zhang, T., Li, W., Chen, J., and Ran, Z. M., “A Clamp-on Ultrasound
Cross-Correlation Flow Meter for Liquid-Solid Two Phase Flow Measurement,” Flow
Measurement Instrumentation, pp. 203-208, Vol. 5, No. 3, 1994.

7
 TABLE 1
FLOW RATES AND SUPERFICIAL VELOCITIES AT 0.3 MPa AND 60 oC

Section A Section B
Flowrate, m3/day 1,590 2,226 2,385
Superficial liquid velocity, 0.6 0.8 0.9
m/s
Superficial oil velocity, 0.06 0.08 0.09
m/s
Superficial water velocity, 0.5 0.7 0.8
m/s
Superficial gas velocity, 0.1 0.15 0.15
m/s

TABLE 2
FLOW CHARACTERISTICS AND CORROSION RATE AT SITE 1.

Film Height Velocity Slug Freq. Flow Rate Corrosion rate (mm/yr)
(cm) (m/sec) (1/min.) (m3/day)
Sample #1 10 1.0 4 1,495 0.5 ~ 0.8
Sample #2 10 0.9 4 1,352

TABLE 3
FLOW MEASUREMENTS AT SITE 3

Film Height Velocity Flow Rate

(cm) (m/sec) (m3/day)
Sample # 1 16.5-17.8 0.7 263
Sample # 2 15 1.0 405

TABLE 4
FLOW CHARACTERISTICS AND CORROSION RATE AT SITE 4

Film Height Velocity Flow Rate Corrosion rate (mm/yr)

(cm) (m/sec) (m3/day)
Sample #1 15 – 16.5 0.6 1,526
Sample #2 15 – 16.5 1.0 2,226 0.8
Sample #3 15 – 16.5 0.6 1,495

8
 TABLE 5
FLOW CHARACTERISTICS AND CORROSION RATE AT SITE 5

Film Height Velocity Flow Rate Corrosion rate (mm/yr)

(cm) (m/sec) (m3/day)
Sample #1 9.0 – 10.0 3.0 2,862 2.1 and 3.0

Transducers

FIGURE 1 - Flow regime and velocity FIGURE 2 – Pulse - echo technique

determination. for film thickness

∆t

Day
Amplitude (mV)

Day

Time (ns)
∆t controls the minimum measurable
thickness
FIGURE 4 - Typical responses from
FIGURE 3 - Technique for corrosion transducer.

9
 234
Site 1
231
Site 4
228 Site 2
Site 5
225
Height, m

222

219
216
Section B
213
Site 3
210
0 300 60 900 1200 1500 1800

Distance, m

Site 1, Flow and Corrosion Detection (-2 deg. Decline)

Site 2, Flow and Corrosion Detection (1.5 deg. Incline)
Site 3, Flow Detection (Top of Gully)
Site 4, Flow and Corrosion Detection (-3 deg. Decline)
Site 5, Flow and Corrosion Detection (-20 deg. Decline)

FIGURE 5 - Field Topography and Measurement Locations

10
 Flow Direction

FIGURE 6 - Effect of Change of Inclination

on Flow Regime

Gas Pocket
Low

High Point

Flow Direction

Figure 7 – Flow Over High Spot

0.9 m/s

10 cm

Figure 8 - Film Height and Velocity at Site 1

11