See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/261133505

Corrosion Of Carbon Steel In Waters Of Varying Purity And Velocity

Conference Paper in NACE - International Corrosion Conference Series · April 2012

CITATIONS READS
8 19,696

4 authors, including:

Narjes Fredj T. D. Burleigh

United States Navy New Mexico Institute of Mining and Technology
15 PUBLICATIONS 590 CITATIONS 48 PUBLICATIONS 1,249 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Narjes Fredj on 27 March 2014.

The user has requested enhancement of the downloaded file.

C2012-0001461

Corrosion Of Carbon Steel In Waters Of Varying Purity And Velocity

N. Fredj and T.D. Burleigh

Materials & Metallurgical Engineering Department
New Mexico Tech
Socorro, NM 87801

K.L. Heidersbach and B.R. Crowder

Chevronʼs Energy Technology Company
Houston, TX 77002

ABSTRACT

There have been several instances where water purified from seawater by reverse osmosis has caused
accelerated corrosion of carbon steel pipes. Therefore research was undertaken to determine the
corrosion rate of low carbon steel in water with different contaminants, with and without aeration, and at
different velocities.

The results show that in high purity deionized (DI) water with oxygen, the corrosion rates were initially
very high but dropped rapidly within a few hours to <100µm/yr. In jet impingement of DI water with
oxygen, the corrosion rates dropped even lower, approaching zero, because the steel surface
passivated.

Water analysis from the reverse osmosis units from the offshore oil drilling platforms showed that
reverse osmosis units would typically leave behind many ppm of NaCl and Na2SO4. Therefore
laboratory tests were conducted and it was found that adding NaCl and Na2SO4 to the DI water,
increased the corrosion rate, proportionally to the concentration. In the jet impingement tests with
contaminated water containing oxygen, the corrosion rate of the steel was ~5 mm/yr, beneath a thick,
gelatinous, ferric hydroxide film. When the same jet impingement tests were conducted with high purity
DI water saturated with oxygen, the corrosion rate was zero, due to the steel surface passivating.

Key words: purified water, velocity, jet impingement, steel, corrosion rate.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 INTRODUCTION

Steel or cast iron flowlines or piping have reasonable corrosion resistance in water, but their corrosion
rate is very dependent on the water composition,1 the aeration,2 and the velocity of the water.3,4 Even a
slight contamination of salt can make benign water corrosive. For example, the output water from a
reverse osmosis (RO) system on an offshore oil platform corroded a carbon steel pipe at the rate of 2.8
mm/yr. This study was therefore conducted to understand the corrosion resistance of steel in high
purity water and in slightly contaminated water.

There are very few reports on the corrosion rate of steel in almost pure water.5,6 Rabald et al.,5
reported on the corrosion rates of steel and cast iron in flowing tap water and in distilled water. In
flowing tap water, they found that steel and cast iron corroded at 0.25 and 0.35 mm/yr respectively.
However in distilled water, it corroded at 73 mm/yr (about 300 times faster). When the temperature was
increased from 17°C to 65°C, the corrosion rate of the steel increased from 73 to 291 mm/yr.

Corrosion rates can also be affected by velocity.2-4, 7-9 Heyn et al.,7 reported that steel had a lower
corrosion rate in flowing water as compared to quiescent water. He also showed that if iron was
corroded by water, the corrosion rate increased rapidly as the velocity of the water increased. He also
found a maximum corrosion rate at about 1.2 mm/s, and that subsequent increases in water velocity
caused a decrease of corrosion. Other authors like Speller et al.,8 found a constant increase of
corrosion with velocity with no decrease in corrosion over the entire velocity range investigated.
Russell et al., 9 reported that the reduction of corrosion rate was only seen on cleaned or polished steel.
He found that if rusted steel were tested, then corrosion rates increased continually up to his maximum
velocity of 1.2 m/s. The mechanism by which the water flow rate decreased the corrosion rate of steel
was described by Fontana et al.10 Fontana reported that the corrosion rate increased with velocity, until
at a critical velocity, the steel passivated and the corrosion rate decreased by several orders of
magnitude. Passivation refers to the phenomena by which the rapid oxygen transport to the surface,
increased the oxidizing potential of the interface, and caused the formation of a thin iron oxide barrier
film.

As seen above, the published literature on the corrosion of steel in potable water is sparse and not self-
consistent, and often does not include the critical parameters such as the velocity, the purity of the
water, and the presence of dissolved oxygen. In this present work, these parameters were studied
using low carbon steel in deionized (DI) water, as well as in an electrolyte that mimicked the reverse
osmosis (RO) water from an offshore oil platform. This offshore oil platform, and similar ones, have
used RO water in the utility system, and they have experienced rapid corrosion of the mild steel piping.

EXPERIMENTAL PROCEDURE

Materials

The steel test coupons were mild steel (UNS G10100) sheet with composition in weight percent of
99.49Fe-0.09C-0.33Mn-0.01Cu-0.01Ni-0.02Cr-0.05Al. The coupons were 0.8 mm in thickness, with
one surface ground, and the opposite surface pickled (acid etched to remove mill scale).

The coupons were cleaned with acetone and methanol prior to the tests, weighed, immersed for a set
period of time, then removed and scrubbed with a plastic bristle brush to remove any rust, then dried
and reweighed. (An inhibited acid was tested for cleaning the corroded samples, but the acid dissolved
too much steel, so its use was discontinued.)

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Weight Loss Tests

The corrosion rates of the 1010 steel coupons were determined by weight loss measurements using
three different corrosion tests. The weight loss of the steel coupons was converted to corrosion rate
(CR in µm/yr) by using equation (1) that had been calculated from first principles:

(1)

In the equation (1), m is the mass loss in mg, A is the surface area in cm2 and t is the time of immersion
in hrs. Similar equations may be found in other corrosion books, e.g. Denny Jones.12 For steel, density
of iron was used, 7.874 g/cm3, and the atomic weight of iron, 55.845 g/mole.13

For the weight loss measurements, the following tests were used: In the first set of tests, the steel
samples (25x75x0.8 mm) were immersed in 200 ml of quiescent water in 250 ml borosilicate
Erlenmeyer flasks. The steel coupons stood on edge so that both sides were exposed to the solution.
These tests were conducted from 1 hour to 14 days at ambient temperature. The water was either
aerated or deaerated by bubbling either oxygen or ultra high purity (UHP) nitrogen gas (99.999%)
respectively through the solution for 20 minutes, then the flask was sealed with a polychloroprene
stopper for the duration of the test. Polytetrafluoroethylene and polypropylene flasks were also tested,
but showed no significant difference in the corrosion rate compared to that of the borosilicate flasks.
(These tests were conducted because dissolved borosilicate has been reported to have an inhibiting
effect on steel at 90°C).13

The second set of immersion tests also used 200 ml of water in a 250 ml borosilicate Erlenmeyer flask
with the same size steel coupons at ambient temperature, but the water was stirred with a magnetic stir
bar and the flask was sealed with a polychloroprene stopper for the duration of the test. The average
linear velocity was estimated based on the revolutions per minute times the average radial diameter of
the stirrer, as shown in equation (2).

(2)

In this equation (2), V is the velocity of water (m/s), D is the diameter of the magnetic stir bar (24 mm)
and ω is the revolutions per minute of the stir bar (rpm). This equation gives only an approximate
velocity since the diameter of the conical Erlenmeyer flask decreased with height, and the coupon was
slightly inclined in the flask.

In the third test, the jet impingement test, the steel sample (25x25x0.8 mm) faced a high velocity jet of
water, shown schematically in Figure 1a and in a photograph in Figure 1b. The vinyl tubes are stained
orange from the rusting steel. The front (pickled surface) of the steel was held in position by a NdFeB
magnet placed on the outside of the tank. The back (ground surface) of the steel was taped with
Scotch 3M tape to prevent crevice corrosion of the backside. A diaphragm pump powered the water
jet, and a valve controlled the velocity of the jet. The total volume of recycled solution was about 4.5
liters. The circular jet nozzle was 4 mm in diameter and the distance from the nozzle to the sample was
8 cm (the jet slightly diverged over this distance.) The velocity of the water jet was determined by
measuring the volume per time and converting to m/s. The jet impingement test was conducted for 24

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 hours for each type of water, and the water was either aerated or deaerated. Oxygen was bubbled to
aerate the water, while an ultra high purity nitrogen (99.999%) gas was bubbled through the system to
deaerate the water (strip out the dissolved oxygen). Rapid gas bubbling was conducted for 20 minutes
prior to each jet impingement test, and then a slow bubbling was conducted continuously during the 24-
hour test at ambient temperature. The aerated solutions were assumed to be saturated with oxygen,
about 8 ppm O2.

(a) (b)

Figure 1: Jet impingement test: (a) Schematic, (b) Photograph with an open top.

Several compositions of waters were tested and the results are shown in the next section. The
composition of the deionized (DI) water is shown in Table 1. Table 1 also gives the composition, pH
and conductivity values for the RO platform water, and the synthetic RO platform water (351 mg NaCl
and 25.6 mg Na2SO4 added to DI water to make 1 liter of solution.) The final column is the synthetic
RO platform water without sulfate.

Table 1
Concentration (ppm), pH and conductivity of the Deionized (DI) water, the Reverse Osmosis (RO)
platform water, the synthetic RO platform water and synthetic RO platform water without sulfate.

Deionized (DI) Reported RO Synthetic RO Synthetic RO

water platform water platform water platform water
without sulfate
Na+ <1 165 148 140
Cl- <0.1 213 215 215
SO42- <1 17.3 18 0
HCO3- <10 - 0 0
pH 5.55 8.5 5.65 5.80
Conductivity (µS) 215 600 766 722

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 The final micrograph was taken with a Hirox(1) Digital Microscope by tiling photographs around the
center of the jet impact.

RESULTS AND DISCUSSION

Corrosion Rate in Quiescent Water

Immersion in Quiescent DI Water Versus Time.

The steel coupons were immersed in the Erlenmeyer flasks for set times. It was observed that the
pickled steel surface rusted prior to the ground surface regardless of aeration or deaeration. Figure 2
shows the drop in the corrosion rate for steel immersed in aerated and deaerated DI water with different
immersion times. Each data point is a separate steel coupon. For DI water aerated with oxygen, the
corrosion rate was initially very high, ~900 µm/yr, but within half a day, it dropped to 50 µm/yr. In DI
water deaerated with nitrogen, the corrosion rate remained at about 50 µm/yr regardless of the length of
the test. The oxygen had a major role only during the first day, although some steel coupons in DI
water with oxygen passivated, and showed no weight loss. The decrease in the corrosion rate during
the first day could either be due to a change in the water composition or a change on the steel surface.
Tests were conducted to identify the critical factor, but the tests were inconclusive due to the large
scatter in the data. In the deaerated solutions, the corrosion was limited by the slow cathodic reactions,
either that of hydrogen evolution or the reduction of the residual oxygen.

Figure 2: Corrosion rate of steel versus

time in aerated and deaerated DI water.

The corroded steel coupons in the saturated oxygen DI water were uniformly covered by a gelatinous
brown reddish layer. The Pourbaix diagram indicates that at pH=5.5, the precipitate should be a iron
(III) hydroxide, Fe(OH)3.14 This product was also reported by Cox.2 He showed that above 5.5 cm3 of
oxygen per liter (8 ppm), the corrosion product was primarily a red gelatinous ferric hydroxide which
was attributed to the rapid rate of oxidation of Fe+2 to Fe+3 which prevented the precipitation of Fe(OH)2.

(1)
Hirox-USA, Inc.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Effect of Dissolved Anions in Quiescent DI Water.

Figure 3 shows the corrosion rate of two steel samples immersed for seven days in oxygenated DI
water containing different concentrations of NaCl, Na2SO4 or Na2CO3. The corrosion rates increased
with an increase of the concentration of ionic impurities, with the chloride anions being more aggressive
than the sulfate, and carbonate ions being the least aggressive. The 1% NaCl showed a maximum of
~130 µm/yr while the 1% Na2SO4 shows a maximum of ~110 µm/yr. The Na2CO3 had a maximum
corrosion rate of ~60 µm/yr at 100 ppm, and then dropped to zero at 1% Na2CO3 when the carbonate
had an inhibiting effect.

Figure 3: Effect of dissolved anions

in aerated DI water for 7 days.

A solution with 215 ppm Cl-, 18 ppm SO42- and 148 ppm Na+ was made to determine the corrosion rate
of steel in the RO platform water (see Table 1). This synthetic RO platform water had a pH=5.65 and
conductivity 766 µS. Figure 4 shows the corrosion rate in aerated and deaerated synthetic RO platform
water after 1 and 7 days of immersion. After 1 day, the steel coupons with oxygen were corroding at
200 µm/yr, while in the same solution deaerated with ultra high purity nitrogen (99.999%), the corrosion
rate was only 5 µm/yr (40x slower). When the test duration was increased to 7 days, the corrosion rate
of the steel in aerated water and the corrosion rate in the deaerated solution approached a similar value
of ~50 µm/yr. This rate is similar to the corrosion rates measured after one week in DI water in Figure
2, and are within a factor of two of the rates in ppm NaCl and Na2SO4 in Figure 3.

The above experiments were in quiescent water where only the composition of water was varied. In the
next section, the corrosion rates were measured in water where both the velocity and the composition
varied.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Figure 4: Corrosion rate in aerated and
deaerated synthetic RO platform water
containing 215 ppm Cl-, 18 ppm SO42-
and 148 ppm Na+ .

Corrosion Rates in Flowing Solutions

Effect of Velocity.

Figure 5 shows the effect of velocity on the corrosion rate of the steel in aerated DI water for 1 and 7
days. After 1 day of immersion, the corrosion rate was very close to zero, regardless of the velocity,
because the steel was passivated. After 7 days of immersion, from 0 to 100 rpm (~0.1 m/s), the metal
became active and the corrosion rates increased with velocity up to a maximum rate of 250 µm/yr.
However, at velocities greater than 200 rpm (~0.2 m/s), the steel switched to the passive state and the
corrosion rate dropped sharply to <10 µm/yr. The active-passive transition of steel with velocity has
previously been described by many authors including Roetheli,3 Heyn,8 Russel,9 Fontana,10 Friend,15
and Cohen.16

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Figure 5: Effect of velocity on the corrosion rate
of mild steel immersed in aerated DI water.

Figure 6 shows the effect of velocity on the corrosion rate of the mild steel in aerated synthetic RO
platform water for 1 and 7 days. The corrosion rate of the steel in synthetic platform water is strongly
dependent on the velocity of the water. Contrary to the DI water in Figure 5 where the corrosion rate
remained stable and near zero, the corrosion rate after 1 day in the synthetic platform water increased
linearly with the velocity and no passive behavior was observed. As the velocity increased, the steel
corrosion rate increased from 200 to 800 µm/yr in the synthetic platform water. After 7 days of
immersion, the corrosion rates also increased with velocity, but at a slower rate. Above 400 rpm (~0.5
m/s) the corrosion rate began to level out with velocity. These results are in agreement with Figure 4
and show the same tendency as observed by Cox et al.,2 and Speller et al.,.7 The higher velocity
increased the transport of oxygen to the surface, and accelerated the oxygen reduction reaction.

Figure 6: Effect of velocity on the corrosion rate

of mild steel immersed in aerated synthetic
platform water for 1 and 7 days.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 The comparison between Figures 5 and 6 demonstrates that regardless of the velocity of the water, or
the time of immersion, the synthetic platform RO water was more aggressive than DI water, because of
the presence of chlorides (and sulfates). The chemical nature of the water is a critical parameter, which
increases the corrosion rates with increased velocity.

After the corrosion tests, the steel coupons that corroded were covered with a gelatinous, brown
reddish film of ferric hydroxide. These coupons were cleaned and photographed. Figure 7a shows a
cleaned steel coupon after immersion in aerated DI water that was stirred at 700 rpm for 7 days. This
sample shows uniform staining, no recognizable pitting and the superficial attack appears uniform. In
comparison, the Figure 7b shows three steel samples immersed for 1 day in the stirred synthetic
platform water with oxygen at different velocities (100, 200 and 700 rpm). The number of pits on the
surface and the depths of the pits increased with the velocity, which agrees with the higher corrosion
rates seen in Figure 6. The increasing velocity, increases the transport of oxygen to the surface, which
increases the cathodic reaction, oxygen reduction. The corrosion reaction, iron dissolution, which
releases electrons, can only proceed as fast as the cathodic reaction, which consumes electrons.

700 rpm

(a)

100 rpm

200 rpm

700 rpm

(b)
Figure 7: (a) Steel coupon after 7 days
of immersion at 700 rpm in aerated stirred
DI water (b) Steel coupons after 1 day of
immersion at 100, 200 and 700 rpm in
aerated stirred synthetic RO platform water.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Jet Impingement Test.

This section shows the corrosion rates in the high velocity jet impingement tests for aerated and
deaerated DI water (Figure 8), and for synthetic platform water (Figure 9). Each data point is a
separate 24-hour, or 72-hour test in a new electrolyte. Figure 8 shows that in DI water with oxygen, the
steel becomes passive for all tested velocities (0.6 to 3 m/s), and the corrosion rate is <10 µm/yr. The
high oxidizing power of the oxygen provides the driving force needed to form the passive film. If the
water is deaerated with nitrogen, there is an acceleration of corrosion rates, up to 600 µm/yr, because
the steel is now in the active domain, and does not have the driving force necessary to form a
protective passive film.

Figure 8: Effect of jet velocity on the corrosion

rate of mild steel after 1 day test with aerated
and deaerated DI water.

In Figure 9, the deaerated synthetic platform water, the corrosion rate was very low, from 5 to
350 µm/yr, regardless of the velocity. However, when the water is saturated with oxygen, the corrosion
rates increased to a maximum ~7000 µm/yr (~7 mm/yr) for the lowest velocity (0.6 m/s) and 4500 µm/yr
(4.5 mm/yr) for a velocity about 3.2 m/s. The average corrosion rate is about 5 mm/yr, which is slightly
higher than the in-service corrosion rates for the RO platform water, which was 2.8 mm/yr. This
agreement between the laboratory tests and the service experience is very good since we do not know
the exact flow conditions.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Figure 9: Effect of velocity of the jet on the
corrosion rate in aerated and deaerated
synthetic platform water with and without sulfate.

The high corrosion rates in aerated synthetic platform water indicate a synergistic effect between the
oxygen, chloride, sulfate and the jet velocity. The duplicate tests shown in Figure 9 indicate a slight
drop in the corrosion rate (~4000 µm/yr) for the synthetic platform water without sulfate ions, confirming
that the main effect was due to the high concentration of chlorides. The role of sulfate ions is minor
because the concentration is minor compared with the dominant effect of oxygen and chloride.

Figure 10 shows four steel samples after 24 hrs of jet impingement in two different waters, deaerated
and aerated, at a velocity ~2 m/s. The top row Figure 10a and 10b is jet impingement in DI water. In
deaerated DI water, Figure 10a, the steel surface is slightly darker except at the center where the jet
impacted the surface. In oxygenated DI water, (Figure 10b) the steel appears new. The second row
Figures 10c and 10d show the steel coupons in synthetic RO platform water. In the deaerated platform
water, the surface of the sample is stained dark, however, in the presence of O2, the sample is very
different with flow lines radiating from the darkened center where the jet impact occurred. The steel
coupon in Figure 10d is also the most corroded according to weight loss measurements.

(a) (b)

(c) (d)

Figure 10: The cleaned steel coupons after jet

impingement tests for 24-hrs at 2 m/s.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 Figure 11 shows a sample before (a) and after (b) cleaning with a brush. The Figure 11a shows a
deposit of the gelatinous brown film of Fe(OH)3 as described previously. Figure 11b shows the cleaned
steel coupon with the outward radial lines due to the water flow.

(a) (b)
Figure 11: (a) before cleaning, and (b) after
cleaning steel coupons after the jet impingement
test in aerated synthetic platform water for
72-hrs at 2.9 m/s.

Figure 12 shows the surface of a steel sample after 3 days of jet impingement in aerated synthetic
platform water with a velocity of 3.4 m/s. Figure 12a is a 3D micrograph of the surface at the jet impact,
and Figure 12b is a topographic image of the same location. The rough topographic surface confirms
that the increased corrosion rates were due to increased pitting corrosion at the center of the jet impact,
and not due to erosion corrosion. Erosion corrosion would have removed the gelatinous layer and
would have left a polished surface. The pitting was accelerated because of the presence of chlorides,
and the increased oxygen transport to the region.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 (a)

(b)

Figure 12: (a) 3D micrograph of the surface of

Steel after the 72-hr jet impingement test.
(b) topographic map of the same area.
The horizontal axes are 7.3x9.8 mm, and
the vertical axis has been exaggerated by 5x.

CONCLUSIONS

The corrosion rate of mild steel has been studied by using the weight loss method in aerated and
deaerated DI water and synthetic RO platform water containing 215 ppm of chloride, 18 ppm of sulfate
and 148 ppm of sodium. Tests were conducted in quiescent solutions and in flowing solutions, with both
stirring and with jet impingement. The results are the following:

(1) In quiescent water aerated with oxygen, the corrosion rate was very high initially but within half a
day, it dropped to the similar level as deaerated DI water. The coupons were uniformly covered
by a reddish brown, gelatinous ferric hydroxide film.

(2) In quiescent aerated water with dissolved salts of chloride, sulfate or carbonate showed a linear
increase of the corrosion rate with the concentration, with chloride being the most aggressive.

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.
 (3) The synthetic platform water which mimicked the water from the RO unit was found to be much
more aggressive than DI water. The synergistic effect of the residual NaCl and oxygen, together
with high velocity, accelerated the corrosion rate of the steel.

(4) The stirring tests showed that the corrosion of the steel was very dependent on the velocity and
composition of the water. Tests with DI water verified that the steel was an active-passive
metal, with passivation being triggered by the rapid transport of oxygen to the surface. However,
with the synthetic platform water, the steel was always actively corroding, with the corrosion pits
becoming more numerous and deeper as the water velocity increased.

(5) The jet impingement tests showed that in DI water with oxygen, steel was passive with a
corrosion rate <10 µm/yr, whereas in synthetic platform water containing Cl-, SO42-, and O2, the
corrosion rate was ~5.0 mm/yr (500x faster). In deaerated water, the corrosion rates were low
and similar for both the DI water and the synthetic platform water.

ACKNOWLEDGMENTS
The authors thank Jacob Kolar, Trevor Contri and Kanayo Monye for their diligence in the corrosion
tests.

REFERENCES

1. H.H. Uhlig, Corrosion and Corrosion Control, 2nd Ed., John Wiley & Sons, (1963).
2. G.L. Cox, B.E. Roetheli, Industrial and Engineering Chemistry, 23, 1012 (1931).
3. B.E. Roetheli, R.H. Brown, Industrial and Engineering Chemistry, 44, 1010 (1931).
4. H.R. Copson, Industrial and Engineering Chemistry, 44, 1745 (1952).
5. E. Rabald, Corrosion Guide. 2nd Ed., Elsevier Publishing, p. 882 (1968).
6. P.E. Francis, A.D. Mercer, ASTM Technical Publication, p. 184 (1985).
7. Heyn, Bauer, Mitt. Kgl. Materialprufungsamt, 28, 62 (1910).
8. Speller, Kendall, Industrial and Engineering Chemistry, 15, 134 (1923).
9. R.P. Russell, E.L. Chappell, A. White, Industrial and Engineering Chemistry, 19, 65 (1927).
10. M.G. Fontana, N. D. Greene, Corrosion Engineering, 2nd Edition, p. 327 (1978).
11. D.A . Jones, Principles and Prevention of Corrosion, 2nd Edition, p. 34 (1996).
12. Sargent Welch/VWR Scientific Products, Catalog No. WLS18805-50 (1998).
13. P. E. Francis, A. D. Mercer, “Corrosion of Mild Steel in Distilled Water and Chloride Solutions:
Development of a Test Method,” ASTM Technical Publication, (1985), ASTM STP 866, p. 184-
196.
14. P.R. Roberge, Handbook of Corrosion Engineering, McGraw-Hill Handbooks, p. 20 (2000).
15. J.N. Friend, J. Iron Steel Inst., 11, 128 (1922).
16. M. Cohen, J. Electrochem. Soc., 93, 26 (1948).

©2012 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. The material presented and the views expressed in this paper are solely those of the
author(s) and are not necessarily endorsed by the Association.

View publication stats