Presented at NACE Corrosion 2003, Paper 03563,
      San Diego, CA, March 2003


                              T. Mathiesen, E. Rislund, T.S. Nielsen and J.E. Frantsen
                                               FORCE Technology
                                                   Park Allé 345
                                           DK-2605 Brøndby, Denmark

                             U. Tørnæs, H.G. Pedersen(1), P. Nielsen and M.B. Petersen
                                                  Krüger A/S(2)
                                                Gladsaxevej 363
                                           DK-2860 Søborg, Denmark

             Corrosion of stainless steel pipes in sewage treatment systems has been studied as part of a
Danish research programme. The pipes were attacked by localized corrosion in locations near the final
stage of the treatment process, where the water is practically free of organic substances. Since the
temperature and chloride content is quite low, it has been difficult to explain the reason for corrosion.
An extensive series of field-tests and inspections showed that the failures in most cases could be
explained by bacterial deposition of manganese dioxide, which results in potential ennoblement due to
its effective cathode properties. Preventive measures for this form of corrosion were evaluated with
special focus on cathodic protection. Finally, revised design diagrams were established to support the
selection of resistant stainless steel grades for low chloride environments affected by highly oxidizing
conditions caused by bacteria.

Keywords: sewage treatment, AISI 304, AISI 316L, EN 1.4435, MIC, potential ennoblement,
manganese bacteria, field tests, cathodic protection, CPT

      Now affiliated to EnviDan Øst A/S, Denmark
      A Vivendi Water Systems Company

             Denmark has about 1400 municipal sewage plants serving the population of about 5 mill.
people. Figure 1 illustrates the process of a typical Danish sewage treatment plant. The increasing
demands for environmental protection have necessitated extensive modernization of the sewage plants
over the last 20 years. This development has involved wide use of stainless steel piping for additional
installations such as sand filters, sludge concentrators or systems for internal reuse of the treated waste
water. An example of a modern sand filter is shown in Figure 2. In parallel with this development, the
number of reported failures due to corrosion of stainless steel has steadily increased to an amount that
required special attention.

The waste water is usually neutral pH (≈pH 7) with a temperature around 20°C and it rarely contains
more than 100-200 ppm chloride. Under such conditions, corrosion would normally not be expected,
even for low-alloyed stainless steel types like AISI 304. Our experience has, however, shown that
corrosion may under certain conditions occur with this steel type as well as improperly treated AISI
316L. The corrosion attacks were typically observed at welds or joints (with crevices) that surprisingly
were located near the final stage of the treatment process where the water is relatively clean. Some of the
failures could obviously be explained by improper welding procedures. However, a number of cases
where corrosion occurred in the base material remained unsolved, although they apparently were
associated with a high corrosion potential of the stainless steel as well as deposits rich in iron and

It is believed that the corrosion observed in sewage plants is associated with the mechanism covered in
several papers by e.g. Dickinson1-3, Lewandowski4, Lutey5, Olesen6 and Linhardt7. The reported
mechanism involves biological oxidation of manganese (and iron) to manganese dioxide, which acts as a
strongly oxidizing agent when precipitated on the steel surface. Failures caused by such deposits have
been reported for applications like cooling water systems, waterworks and power plants. Another related
mechanism has also been considered and involves oxidation of manganese due to the presence of strong
oxidizers such as chlorine or hypochlorite (Tverberg8 and Avery9) or peroxide (Kovach10). However,
presence of such oxidizers in a sewage treatment plant is not common, although certain bacteria generate
peroxide as part of their metabolism (Ferón11).

Only a limited number papers have been published on corrosion of stainless steel in sewage plants as a
specific issue. In our first article12 a few cases were introduced, but since then the total number has
almost been doubled. Other authors have reported related work (Tuthill13, Iversen14 and Kikuchi15), but
it is not clear whether the manganese depositing mechanism was involved in the concerned sewage

In order to understand the particular corrosion mechanism in the sewage plants a Danish research
programme has been carried out. In addition to the inspection of the corrosion failures the study has
involved field tests in different sewage treatment plants and bacterial evaluation. Furthermore, two
approaches to prevent corrosion have been pursued, i.e. the application of cathodic protection in repaired
plants and the establishment of safer guidelines for material selection in new plants.


           The inspection programme was based on a questionnaire survey that included 62 Danish
sewage treatment plants that provided their experience on stainless steel. A surprising result of this

survey was the fact that 60% had experienced corrosion failures with stainless steel. Moreover, most of
the failures (80%) were preferentially located in installations near the final stage, i.e. secondary
clarification, sand filter or piping for discharging and recycling treated water.

On this basis 18 sewage plants were selected for closer inspection. The major objective of this effort was
to characterize the form of corrosion and relate it to the concerned material and environment. The
inspections did also include plants without corrosion failures as references.

After just a few inspections it became clear that most failures occurring in the treated waste water were
associated with extensive deposits of manganese dioxide on the internal surface of the stainless parts. An
obvious example of this is the sand filter hatch in Figure 3. This part had been installed for 5 years
before it was penetrated by crevice corrosion along the sealing ring. Similar attacks were found in flange
couplings in at least 6 plants.

In contrast to this, pitting in freely exposed base material was only observed once. The concerned outlet
flow meter station shown in Figure 4 was made from AISI 304 plate. Interior inspection revealed that
the surface was fully covered by manganese deposits, but the possibility that manufacturing defects also
affected corrosion could not be excluded.

The third and most frequent form of attacks appeared in welds. Several failures were obviously related
to poor welding practices, but the inspection did not always allow a thorough evaluation that could
determine whether all requirements had been fulfilled as concerns weld geometry, penetration and
avoidance of heat-tints.

The overall results of the inspections are summarized in Table 1. In general the failures were located at
welds or in crevices, while only one case shows pitting in the base material. This agrees with fact that
welds and crevices are more vulnerable to corrosion than base material. At the same time it appears that
even properly treated AISI 304 steel involves a large risk of localized corrosion in treated waste water,
regardless of the fact it may take up to 8 years to develop penetrations in 1-3 mm material.

Only few corrosion failures were observed in installations of EN 1.4435 material (i.e. AISI 316L with
min. 2.5% Mo). Those cases subjected to a closer examination showed that corrosion was associated
with weld defects such as heat tints or inadequate penetration. However, in a few cases inspection was
not undertaken but assumedly weld defects were involved here as well. The overall picture is, however,
that EN 1.4435 is a safe material in most installations for treated waste water. This is partly confirmed
by the fact that the material resists water containing a relative high amount of chloride (1000 mg/l) in a
sewage plant (G) where manganese deposits were observed.

                                 EXPERIMENTAL PROCEDURES

             The field tests were performed by use of 5 identical systems pulling water from the
clarification stage in different sewage treatment plants. An example of the set-up is shown in Figure 5.
This system consists of a cell made of PVC, which contains 16 stainless coupons. All coupons measure
2x50x100 mm and include the three types of stainless steel listed in Table 2. For each steel type one
plain, two welded and two coupons with crevices were exposed. During exposure the corrosion potential
was measured for each coupon by use of two reference electrodes and the remote operated datalogger
system. The conductivity and temperature of the water in the cell are also measured. The flow through
the cell was approx. 5 litres/min, which corresponds to a flow rate of 0.005 m/s over the coupon surface.

Polarisation experiments were carried out in the cell by use of a portable CorrOcean MKII potentiostat
with a mesh of MMO coated titanium as counter electrode. The polarisation was made in cathodic
direction from the corrosion potential at a scan rate of 1 mV/s.

Furthermore, the field tests were supplemented by collection of water samples taken at the end of the
sewage plant. The water samples were analyzed by use of ICP-OES or potentiometric methods to
determine pH and conductivity as well as the content of chloride, manganese and sulfate. Solid surface
deposits from the different treatment plants were analyzed by use of ESEM/EDS.

Water samples and biofilm on selected coupons were subjected to microbiological analysis after
exposure for at least 100 days. The analysis included a wide variety of evaluation methods that involved
incubation of SRB, nitrate, manganese and iron related bacteria. W. Sand at Hamburg University
performed all microbiological tests.

The critical pitting temperature (CPT) of stainless steel coupons was obtained by polarization at fixed
temperature in the laboratory. To avoid crevice corrosion, a flushed port cell was used, similar to the
ASTM G150 practice16. Polarization was performed from the corrosion potential in anodic direction at a
scan rate of 6 mV/min. The test solution was prepared from demineralised water added different
amounts of NaCl (100, 200, 500 and 1000 mg/l Cl-). The pH was neutral (7-8) as the solution was
unbuffered. The critical pitting potentials were read at 10 µA/cm2 and converted to approximate iso-
potential curves of the critical pitting temperature (CPT) (Arnvig17).

The effect of cathodic protection was studied using a new full-scale system in a sand filter. Protection
was applied by impressed current supplied by titanium anodes. Each anode covered an area of 2.2 m2 in
the pipes (Ø250 mm). The corrosion potential was monitored at a distance of 1.4 m from the anode. In
parallel, the applied current of each system was monitored.


In-situ Potential Measurements

            The field tests were performed at the eight different sewage treatment plants listed in Table
3. The locations were selected to cover a wide range of environments as concerns cleanliness of the
treated water and content of chloride, manganese and solid substances. In all cases the cell was located
near the final stage in the treatment process, which involves either final clarification tank or a sand
polishing filter. In Plant B measurements were made both before and after the sand filter. It should also
be noted that five of the plants had encountered corrosion of stainless steel pipes according to the
inspection results summarized in Table 1.

The most important result of the tests has been the measurement of corrosion potentials. It was here
observed that the different stainless steel types show a very similar behavior at each location, while the
general behavior between each location may vary to a great extent.

Figure 6 compares the potential curves of AISI 304 steel from the different locations. It appears that the
potential level is quite low and unsteady in plant C where the measurement is made in the water from a
clarification tank. Here an explanation for the low potential level may be the unusual water composition
that involves a rather high residual content of inorganic salts in the treated water.

The potential level is generally much higher in the other plants. In three cases (Plant B-C, B-F & E) the
potential increases to a surprisingly high level during the first 20-30 days of exposure. After this period
the potential remains almost unchanged at a potential of about 470 mV SCE. The other plants show a
less steady behavior but occasionally reach a high potential level in the range of 300 - 400 mV SCE. An
intermediate low potential is usually related to flow problems that caused oxygen-depleted conditions
and thereby affected all coupons in the exposure cell.

When comparing the maximum potentials of clarified and filtered water, a slight difference occurs. The
highest potentials in the range of 480 - 500 mV SCE appear in filtered water (except Plant D), whereas
clarified water results in slightly lower potentials from 420 to 460 mV SCE.

Although corrosion had occurred in the pipe system in five of the included plants, no signs of persistent
critical corrosion was observed on the exposed steel coupons. It should, however, be mentioned that
short-lived events of activation were observed as a sudden drop in potential for some of the coupons.
This phenomenon is shown in Figure 7, which includes the potential curves of all 16 coupons in Plant A.
The activation events have been observed for all three stainless steel types but show no obvious pattern
as regards frequency or magnitude that allows further discrimination of the coupons. It should again be
noted that the stable potential level of the coupons is roughly the same or at least within a band of 20

In-situ Polarization Measurements

            In conjunction with the potential measurements a series of polarization experiments was
carried out in-situ to study the cathodic behavior of the exposed steel coupons. In general the different
coupons at the same location showed a very similar behavior regardless of steel quality. This result
agrees well with potential measurements, which also showed a very similar behavior of all steel

Figure 8 shows the obtained cathode curves at different locations together with a freshly exposed
reference (pickled) without any surface deposits or biofilm. It is noticed that the cathodic response is
quite strong for the coupons (B, F & G) all showing a high corrosion potential of about 470 mV SCE. In
contrast to this, coupons A, D & H show a behavior at an intermediate potential close to the one
observed for the freshly exposed reference. It is also noticed that nearly all curves reach the same
limiting current level of about 10 µA/cm² at –600 to –700 mV SCE. This current is supposed to be
associated with the limiting transport of oxygen to the surface. The current peak for curve G in this
region may indicate a greater abundance of manganese dioxide deposit, which requires a longer overall
polarization time to be fully consumed. Once consumed, the cathodic rate decreases to the oxygen
limiting current.

Composition of Surface Deposits and Microbiologically Analysis

             After about 100 days of exposure some of the coupons were removed from the exposure
cells in order to make an analysis of the surface deposits. None of the coupons showed any signs of
critical corrosion, which was confirmed by weight-loss measurements. The coupons appeared slimy due
to the formation of biofilm on the steel surface. The only exception is the coupons from Plant C, which
were covered by light gray surface deposits. The major compounds in the deposits are listed in Table 3
and were obtained by SEM/EDS analysis.

It appears that most of the coupons show a relatively high amount of manganese and iron. An exception
is Plant C, which showed a grey calcareous deposit that contains significant amounts of chloride and
sodium. This tendency for inorganic deposits may possibly have prevented the formation of biofilm. The
exposed coupon from Plant D was also covered by calcareous deposit. This is in disagreement with the
analysis of a corroded pipe section from the same plant, which in addition contained manganese
deposits. The exposed coupons from the other plants appeared more or less covered by a dark brown
deposit, i.e. manganese dioxide. An example of this is shown in Figure 9.

The deposits from the plants where potential ennoblement is observed were also studied visually in the
scanning electron microscope. In order to preserve bacteria, the coupons were examined in wet
condition at low vacuum. However, this technique revealed no clear indications of the bacteria that
presumably are causing the manganese and iron rich deposits. Generally, the deposits can be
characterized as powdery formations as shown in Figure 10.

The coupons subjected to microbiological analysis had all formed a well-developed biofilm that
contained cells and bacteria products, such as extra cellular polymers and decomposition products of the
bacteria. The bacteria types in the biofilm were mainly denitrifying, manganese reducing and manganese
oxidizing, while sulfur oxidizing bacteria were only present in the water samples.

The analyzed coupons showed various amounts of deposited manganese. However, the presence of the
manganese oxidizing bacteria was not detectable in all cases as it appears from Table 3. This result is
surprising, especially for the case of Plant F where distinctive manganese deposition was found.

In addition to this the concentration of dissolved manganese in the waste water was measured in at least
five samples from each plant. As it appears from Table 3 the concentration varies from the detection
limit of 6 µg/l in Plant B to 210 µg/l in Plant G. There is no clear correlation between this concentration
and the presence of manganese oxidizers. Even the lowest manganese concentration provides enough for
deposition of manganese dioxide.

Critical Pitting Temperature (CPT) Diagrams

            In order to evaluate the risk of pitting of welded material at conditions relevant in sewage
plants, a series of systematic CPT tests were performed. This effort involved measurements for steel
types AISI 304 and EN 1.4435 (≈316L) at fairly low chloride content (100-1000 mg/l), moderate
temperatures and highly oxidizing conditions (450-500 mV SCE). Only few data in the literature are
available for the pitting resistance in this range.

The iso-potential curves of the CPT as function of chloride content were obtained by measuring the
pitting potential various temperatures and chloride concentration. The pitting potential was read at the
point where the corrosion current exceeded 10 µA/cm2. For some combinations of chloride and
temperature the steel cannot exhibit pitting regardless of the applied potential. In such case the
temperature is below the Potential Independent Critical Pitting Temperature (PICPT) of the steel type.
From the obtained matrix of data the PICPT and iso-potential CPT curves for 300, 450, 600 mV SCE
were estimated.

The AISI 304 steel grade was tested in two welded qualities, i.e. slightly heat tinted (light brown) and
pickled condition. The heat-tinted quality showed a severe risk of pitting at potentials below 250 mV
SCE, which is inadequate when related to the high corrosion potentials of 450 - 500 mV SCE in most
sewage plants.

The pickled condition1 shows a higher tolerance against corrosion as shown in Figure 11. It appears that
this steel grade have acceptable resistance at 450 mV SCE at low chloride concentration (<500 mg/l).
However, it should be noted that the presence of small defects or crevices lowers the critical temperature
about 20° in comparison to pickled condition. Therefore, if the AISI 304 grade is to be used in treated
waste water, very strict and costly manufacturing practices are required to obtain ideal weld and surface
quality. On this basis, we generally consider the AISI 304 grade as unsuitable for this purpose.

The EN 1.4435 steel grade was tested in the same welded qualities as mentioned above. For this material
the detrimental effect of heat tints can be related to the pickled condition within the relevant range in
Figures 12 and 13. Whereas the PICPT is unaffected by heat tints, the potential dependent CPT is clearly
affected as it appears from the iso-CPT curves. However, the heat tinted condition still shows a useable
region within the most common combinations of chloride, temperature and potential in treated waste

Cathodic Protection

            Measurements were performed on a new installation for cathodic protection of stainless steel
piping connecting a sand filter. The potential and the impressed current were logged at three different
locations in the outlet pipe of the sand filter from start-up.

The cathodic protecting was applied by use of a commercial design of titanium anodes and a constant
current power supply. Four anodes and three reference electrodes were mounted alternately and
equidistantly 5½ pipe diameters apart in one pipe. Figure 14 shows a schematic of the electrode

In order to evaluate the effective range of each anode, current was supplied to the anodes in stages over
the first 24 days, see Figure 15. It appears that the protection of the first anode (No.1) only reach the
nearest reference electrode (Ref.1) positioned 5½ pipe diameters away. Accordingly, the next anode
(No.4) connected after 16 days mainly affected its adjacent reference electrode (Ref.3).

At 24 days all anodes are connected, each delivering 8 mA (0.4 µA/cm2). From this point the potential
stays at a low level for the next 20 days. The protection potentials then slowly increases and after 70
days reaches 125-175 mV SCE which is stable for the rest of the observation period (400 days). This
potential is still lower than the initial corrosion potential of the unprotected system. When the impressed
current subsequently is increased from 8 mA to 30 mA (and later 60 mA) the resulting effect is limited,
i.e. the potential decreases to 75 mV SCE. It is further noticed that the current demand for CP is in fair
agreement with the polarization data of the ennobled coupons in Figure 8.

Experiments in other sewage plants demonstrated a similar behavior and confirmed the above results.

An important result of the experiments was that the polarization due to the cathodic protection extends
to a distance of several pipe diameters from the anodes. The potential drops to a very low level
immediately after applying cathodic protection. After about 3 weeks the potential rises again, possibly
due to changes in the biofilm, and at this stage a further increase in current have only limited effect.
However, the polarization due to the cathodic protection was still significant after 500 days, which was
proven by cutting off the current for a short period. Moreover, the effectiveness of cathodic protection

    Pickled in 2%HF, 20%HNO3 for 30 min at ambient temperature.

for damaged stainless piping is confirmed by the experience from the system in Plant K as mentioned
previously in Table 1.


            It is well known that several groups of bacteria enable oxidation and precipitation of
manganese and iron oxides. These bacteria are generally characterized as ferro/manganese oxidizers and
include Galionella as the best-known example in relation to corrosion. From the literature it is known
that other bacteria, such as Siderocapsa, Leptothrix, Sphaerotilus, Crenothrix and Clonothrix produce the
same by-products. All these bacteria are quite common and may be expected to be present in sewage
water. The exact conditions required for colonization of these specific bacteria are, although, difficult to
predict due to the competition with other bacteria or microbiological reactions.

The results of the chemical analysis correlate well with the above hypothesis, since the biofilm of the
coupons in all cases were enriched by iron and manganese. However, the results of the microbiological
examinations were less obvious due to the absence of manganese oxidizing bacteria in some of the
incubation tests.

There are several facts that points out the ferro/manganese oxidizing bacteria to be important for the
corrosion of stainless steel in sewage plants. The manganese dioxide observed on the exposed parts is
known as an effective cathode material, which explains it’s common use in batteries. This fact also
explains the strong potential ennoblement combined with the enhanced cathodic response observed in
two of sewage plants. The overall mechanism is illustrated in Figure 16.

Another interesting observation is the amount of precipitated manganese dioxide when considering the
low content of dissolved manganese in the water. Although this content in several cases was less than 6
µg/l, deposits of manganese still occurred as one of the major elements in the biofilm. This fact indicates
a highly selective deposition mechanism of manganese, which in the absence of strong oxidizing agents,
can only be explained by bacteria. In addition to this the results of the eight different locations indicate
that these bacteria become dominating when the water is relatively clean. Moreover, a relatively long
period of 20-30 days is observed before the maximum potential is reached. This effect is usually related
to a time-dependent build-up of the biofilm on the exposed steel surface.

It was not possible to reproduce corrosion of the exposed coupons to the same extent as observed in the
damaged steel pipes. Several reasons may be given for this somehow disappointing result. First of all we
know that there is a considerable initiation time for this type of corrosion, since some of damaged pipe
systems did not fail until after 3-5 years of installation. This effect is possibly related to a slow build-up
of the required amount of manganese dioxide to develop corrosion in the water with a low content of
chloride and dissolved manganese. Another related explanation is the possibility that the area of coupons
has been too small to provide the required cathode area to stimulate a persistent active attack.

At the time where the stainless materials were selected, it was difficult to predict the actual conditions in
the concerned sewage plants. For instance, the common potential criterion used for determination of the
critical pitting temperature (CPT) in fresh waters was 300 mV SCE. This value was believed to give a
considerable safe margin for the AISI 304 steel in the present environments as concerns chloride content
and temperature. The results of this study have, however, clearly shown the need for reconsidering this
design criterion when selecting stainless steel grades for fresh water systems where iron and manganese
oxidizing bacteria can be expected. For this purpose the CPT diagrams were established for the relevant

range of chloride content and potentials. In general, the obtained diagrams correlate well with the
practical observations made during the inspections. For instance, both data sets have clearly shown that
304 type steel have inadequate resistance under influence of manganese bacteria, unless strict
precautions are taken to obtain optimum corrosion resistance of this material. The EN 1.4435 (≈316L)
grade possesses much higher chloride tolerance and may in some cases even be applied in slightly heat
tinted condition.

Another approach in preventing corrosion has been to evaluate solutions for existing plants that already
suffer from corrosion failures. The only feasible solution in this regard is cathodic protection, since
coating or environmental control rarely are possible options in complex pipe systems in a sewage plant.
The monitoring of cathodic protection in plants has contributed with essential information as concerns
geometric design and the required current to obtain proper protection. Based on this experience, cathodic
protection is now considered as an appropriate method that in some cases can minimize the need for
extensive replacements.

Finally, the above study has demonstrated that traditional water analysis often is an insufficient tool to
predict the corrosion behavior and risk of MIC with stainless steel in certain media. From our
experience, the most rewarding approach is field potential measurements combined with proper CPT


            The presented results have related the mechanism of manganese depositing bacteria to
several corrosion failures of AISI 304 stainless steel in treated waste water. In some cases the higher
alloyed 316L grade was attacked too, but such failures were associated with poor welds.

The field tests performed in different sewage treatment plants indicate that the corrosion of stainless
steel is related to microbiological activity that result in a strong ennoblement of the steel surface. This
was verified by the measurement of high corrosion potentials in the range of 400-500 mV SCE in seven
out of eight sewage plants.

The chemical analysis of the deposits from these plants showed a high content of manganese (and iron)
even in the situations where the content of dissolved manganese in the water was less than 6 µg/l. This
observation indicates with some support of the microbiological analysis that manganese and iron
oxidizing bacteria are the major cause for the observed potential ennoblement.

The observed form of corrosion should preferably be prevented by proper selection of material grade.
The obtained CPT data show an enhanced risk of localized attack of the AISI 304 grade steel in treated
waste water at 0-20°C under influence of manganese bacteria. Pickling can improve the performance,
but generally the EN 1.4435 grade is considered as a better choice.

Finally, it was shown that cathodic protection can prevent corrosion in damaged stainless steel pipe
systems and thereby reduce the need for extensive replacements.


            The research programme was carried out by Krüger A/S, Helsinge Kommune and FORCE
Technology with financial support from the Danish Agency of Trade and Industry. Sincere thanks go to
graduate student Dagny Jepsen who performed some of the experimental work as part of her thesis.


1.     Dickinson WH, Caccavo FJr, Lewandowski Z, “The Ennoblement of Stainless Steel by
       Manganese Dioxide Biofouling”, Corrosion Science, Vol.38, No. 8, pp.1407-1422 (1996).
2.     Dickinson WH, Lewandowski Z, Geer RD, ”Evidence for Surface Changes During Ennoblement
       of Type 316L Stainless Steel: Dissolved Oxidant and Capacitance Measurements”, Corrosion,
       Vol.52, No.12, pp. 910-920 (1996).
3.     Dickinson WH, Pick RW, "Manganese-Dependent Corrosion in the Electric Utility Industry",
       Proceedings of NACE Corrosion 2002, Denver, Paper No. 02444, 2002.
4.     Lewandowski Z, Hamilton A, "MIC of Stainless Steels as a Model System to Study Metal-
       Microbe Interactions", Proceedings of NACE Corrosion 2002, Denver, Paper No. 02474, 2002.
5.     Lutey RW, Richardson GM, "Procedure Used for Root Cause Analysis: Pitting in Stainless Steel
       Tubes Associated with MnO2 Deposition", Proceedings of NACE Corrosion 2002, Denver, Paper
       No. 02472, 2002.
6.     Olesen BH, Yurt N, Lewandowski Z, "Effect of Biomineralized Manganese on Pitting Corrosion
       of Type 304L Stainless Steel", Materials and Corrosion, vol. 52, pp. 827-832, 2001.
7.     Linhardt P, “Manganoxiderende Baterien und Lochkorrosion an Turbineteilen aus CrNi-Stahl in
       einem Laufkraftwerk”, Werkst. Korros., Vol.45, pp.79-83 (1994)
8.     Tverberg JC, Pinnow KE, Redmerski LS, “The Role of Manganese Fixing Bacteria on the
       Corrosion of Stainless Steel”, NACE, CORROSION/90, Paper No. 151 Houston, Texas (1990).
9.     Avery RE, Lutey RW, Musick J, Pinnow KE, Tuthill AH, “Corrosion of Stainless Steel Piping in
       High Manganese Fresh Water”, Materials Performance, Vol. 35, No. 9, pp. 59-62 (1996)
10.    Kovach C V, “Types 304 and 316 Stainless Steels Can Experience Permanganate Pitting in
       Water Handling Systems”, Mater. Perform.,Vol.38, No.9, pp.70-75 (1999).
11.    Féron D, Roy M, "Corrosion behaviour of stainless steels in natural waters : focus on
       microbiological and chemical aspects", Proceedings of EuroCorr 2000, London, Sept. 2000.
12.    Mathiesen T, Rislund E, Nielsen TS, Frantsen JE, Tørnæs U, Pedersen HG, Nielsen P,
       "Microbial Corrosion of Stainless Steel Pipes in Sewage Treatment Systems", Proceedings of
       EuroCorr 2000, London, Sept. 2000.
13.    Tuthill AH, Lamb S, "Stainless Steel in Municipal Waste Water Treatment Plants", NiDI
       Technical Series No. 10076, Nickel Development Institute Mar. 1998
14.    Iversen, A, "MIC on Stainless Steels in Wastewater Treatment Plants Field Tests and a Risk
       Assessment", Proceedings of NACE Corrosion 2002, Denver, Paper No. 02451, 2002.
15.    Kikuchi Y, Matsuda F, Tomoto K, Mishimura M, "Microbiologically Influenced Corrosion
       (MIC) of Stainless Steel Weldments on the Pipe Line in the Sewage Treatment Plant",
       Transactions of the JWRI (Japan Welding Research Institute) (Japan), vol. 24, no. 1, pp. 63-67,
16.    ASTM G150-99 Standard, "Standard Test Methods for Electrochemical Critical Pitting
       Temperature Testing of Stainless Steels", ASTM, 1999.
17.    Arnvig PE, Bisgård AD, "Determining the Potential Independent Critical Pitting Temperature
       (CPT) by a Potentiostatic Method Using the Avesta Cell, Proceedings of NACE Corrosion/96,
       Denver, Paper No. 96437, 1996.

                                  TABLE 1.

Plant    Chloride      Stainless    Inspection result
ID.       mg/l a      steel typeb
A          180           304        Penetrating pits in stainless piping for rinse water taken at the plant outlet.
B          100          304,        No signs of corrosion. Extensive deposits of iron and manganese oxides in
                       S31803       the sand filtration plant.
C          3000       304, 316L     Stainless steel is solely used for rinse water taken at the plant outlet. No
                                    corrosion was observed in spite of the high chloride concentration (2000-
                                    6000 mg/l).
D           70        304, 316L     Extensive pitting in all pipes for the sand filter (both types 304 and 316L).
                                    Attacks were located in welds of poor quality. Crevice corrosion in one
                                    flange coupling. Slight manganese deposits.
E           165          304        Penetrating pits in permeate water piping in the sand filter. Attacks were
                                    located in welds and crevices. Moreover, less severe corrosion in piping for
                                    sludge dewatering.
F           160          304        Several corrosion failures in the stainless steel piping for the sand filter
                                    (mainly type 304). Failures involved both inlet and permeate water pipes.
                                    Attacks were located in crevices and welds.
G          1000         316L        Stainless steel piping is solely used for rinse water taken at the outlet. No
                                    corrosion has been observed.
H           180          304        Penetrating pits in flow meter at plant outlet (Figure 4). Extensive deposits of
                                    manganese oxide. Attacks were located in base material, welds and crevices.
I           170          304        Penetrating pits in the stainless steel piping for a sand filter. Corrosion was
                                    restricted to inspection hatches (Figure 3) and pipes for the permeate. Attacks
                                    were located in welds and crevices. Extensive deposits of manganese oxide.
Jb          60           304        Penetrating pits in piping after sand filter. Attacks were located in welds and
                                    crevices. Extensive deposits of manganese oxide.
K           140         316L        Piping for permeate in sand filter was previously attacked by pitting in
                                    several welds (probably of poor quality) and by crevice corrosion in flange
                                    couplings. Corrosion has successfully been prevented by cathodic protection
                                    by impressed current for 6 years.
L            -           304        Penetrating pits in pump station. Attacks were randomly located in base
                                    material as well as welds.
M            -           304        Penetrating pits in piping for sludge transport due to dosage of iron chloride.
N           420          304        Penetrating pits in welds of piping for rinse water taken at the outlet (clarified
                                    water). Extensive deposits of iron oxide.
O           260       304, 316L     Penetrating pits in poor welds of 316L piping for cooling water taken at the
                                    outlet (clarified water). Extensive deposits of manganese oxide. Similar 304
                                    piping was free of corrosion..
P           280         (304)       Penetrating pits in welds of stainless steel piping and pump systems
                                    subsequent to filtration. Pitting in all stainless steel pipes.
Q            80         (304)       No signs of corrosion.
R           120         (304)       No signs of corrosion.
a) Average values b) "316L" covers both AISI 316L and EN 1.4435 grades c) Waterwork

                                          TABLE 2.

           EN              UNS                                      Typical composition (%wt.)
                                                           C          Cr       Ni        Mo              N
           EN 1.4301       S30400                        ≤0.05       18.3      8.7
           EN 1.4435       S31603 (> 2.5%Mo)             ≤0.05        17       11        2.7
           EN 1.4462       S31803                        ≤0.03        22       5.5        3             0.15

                                             TABLE 3.
                                  SUMMARY OF FIELD EXPOSURE TESTS

Plant     Water          [Cl-]    [Mn] a     Ecorr b   Deposit c       Mn-bacteria     Appearance of exposed coupons
ID.       type           mg/l      µg/l      mV SCE                    (Leptothrix)
A         Clarified       180       60        440      Mn, P               yes         No signs of corrosion
B-C       Clarified        70       <6        460      Mn, Fe              yes         No signs of corrosion
B-F       Filtered        70       <6         500      Mn, Fe           yes – yes d    Slightly discolored (rainbow)
C         Clarified      3020      290        320      Ca, Cl, Na          n.d.        Micro pits in HAZ of SS 304.
                                                                                       Crevices discoloured.
D         Filtered        70         50       390      Ca, Fe               no         No signs of corrosion
E         Filtered      165          20       480      Mn, Fe, Ca           no         Micro pits in HAZ of SS 304.
F         Filtered      160          40       500      Fe, Mn, P            no         No signs of corrosion
G         Clarified     1000        210       420      Mn, Fe, Ca           yes        Slightly discolored (rainbow)
H         Clarified      180         70       430      Fe, Ca, Si           yes        Slightly discolored (rainbow)
a) average of total manganese, b) corrosion potential, max. value, c) primary inorganic compounds in biofilm, d) result of
two separate samplings.

FIGURE 1 - Schematic of the process in a typical Danish sewage treatment plant. The numbered sections (1-6)
indicate different corrosive classes. Most corrosion failures of stainless steel occurred in the treated waste water
(section 4).

                   FIGURE 2 - Stainless steel piping is extensively used for drainage and
                   flushing systems in sand filters. The sand resides in cells to both sides of the

                  FIGURE 3 - Crevice corrosion on a stainless steel hatch covered by dark
                  MnO2 rich deposits. The hatch is located in the pipe system after a sand filter
                  in sewage treatment plant I.

a.                                                       b.
FIGURE 4 - Pits in SS 304 base material at the outlet flow meter in sewage treatment plant H. a) overview
showing two penetrations, b) extensive attack surrounded by MnO2 deposits on the inside.

a.                                                     b.
FIGURE 5. The set-up used for the field tests. a) Schematic of the system, b) Field measurement in a
sedimentation tank.



 E (mV SCE)

                0                                                                                             Plant ID.
              -200                                                                                                C
              -400                                                                                                F

                     0            20           40            60            80           100             120               140
                                                              Time (days)

FIGURE 6 - Potential-time curves of SS 304 coupons in nine different sewage treatment plants.

              480                                                                                                   16


                                                                                                                          T (°C); Cx10(mS/cm)
 E (mV SCE)

              440        Potentials



                                                            Welded                                                  4
              400                                           SS 304

                                 S31803                                                 EN 1.4435
              380                                                                                                  0
                 55.0     55.5         56.0   56.5   57.0         57.5    58.0   58.5       59.0       59.5    60.0
                                                            Time (days)

FIGURE 7 - Potential-time curves of 16 different stainless steel coupons in plant A. All three steel types show
short-lived activation peaks which last for about 4-5 hours before the initial potential level is restored.



   E (mV SCE)


                -600         F
                  0.000001             0.00001   0.0001                0.001     0.01              0.1
                                                          I (mA/cm²)

FIGURE 8 - Cathode polarization curves obtained on stainless steel coupons in different treatment plants after
about 100 days of exposure. The reference represents a freshly exposed coupon without any surface deposits or
biofilm. Scan rate 1 mV/s.

a. Coupon exposed in Plant B (clarified water)  b. Coupon exposed in Plant F.
FIGURE 9 - Stainless coupons exposed for +200 days showing a) moderate and b) extensive deposition
manganese dioxide.

FIGURE 10 - SEM picture of surface deposit from a stainless steel part in
Plant I where the most extreme deposition of MnO2 was observed.

FIGURE 11 - Critical pitting temperature (CPT) of SS 304 in welded and
pickled condition. Potentials are relative to SCE.

FIGURE 12 - Critical pitting temperature (CPT) of EN 1.4435 (≈SS 316L)
in welded and heat tinted condition. Potentials are relative to SCE.

FIGURE 13 - Critical pitting temperature (CPT) of EN 1.4435 (≈SS 316L)
in welded and pickled condition. Potentials are relative to SCE.

               FIGURE 14 - Schematic of the electrode arrangement for cathodic protection.




 E (mV SCE)





              -400                                                                                 Ref.1
                     0      10        20       30        40         50      60       70      80   90        100
                                                              Time (days)

               25           2&3
               20           4
      I (mA)

                     0      10        20       30        40         50      60       70      80   90        100
                                                              Time (days)

FIGURE 15 - Impressed cathodic current and potential development of the pipe for the set-up in Figure 14.
Anodes were connected in stages during the first 24 days.

     Aerated solution containing chloride                 MnO2
                                         Protective oxide layer
             2+                                 2+
        O2 Mn                             O2 Mn
     Bacteria        Mn                Bacteria       Mn2+
        MnO2                              MnO2

                             Me++           e-

                                       Deaerated pit
                                       low pH, high [Cl-]
     Stainless steel

FIGURE 16 - Schematic of the corrosion mechanism that involves
manganese depositing bacteria.


To top