Maruthamuthu, S. et al. - Teesside's Research Repository

This full text version, available on TeesRep, is the post-print (final version prior to publication) of:
Maruthamuthu, S. et al. (2009) 'Impact of ammonia producing Bacillus sp on
corrosion of cupronickel alloy 90:10', Metals and Materials International, 15 (3),
pp.409-419.
For details regarding the final published version please click on the following DOI link:
http://dx.doi.org/10.1007/s12540-009-0409-9
When citing this source, please use the final published version as above.
This document was downloaded from http://tees.openrepository.com/tees/handle/10149/92489
Please do not use this version for citation purposes.
All items in TeesRep are protected by copyright, with all rights reserved, unless otherwise indicated.
TeesRep: Teesside University's Research Repository http://tees.openrepository.com/tees/

Abstract
Impact of ammonia producing Bacillus sp. on corrosion of
cupronickel alloy 90:10
S. MARUTHAMUTHU a* , P. DHANDAPANI a , S. PONMARIAPPAN b ,
JEONG-HYO BAE c , N. PALANISWAMY a & PATTANATHU K.S.M.
RAHMAN d
a Microbial corrosion, Garrett block, Corrosion Protection Division,
Central Electrochemical Research Institute, Karaikudi – 630 006.
b Division of Biotechnology, Defence R & D Establishment Ministry of Defence,
Gwalior - 474 002, M.P .
c Electrokinetic Research group, Electrotechnology Research Institute,
Changwon city, Gyengsangnam-do 641-120, Korea
and
d Biotechnology Research Cluster, School of Science and Technology,
University of Teesside, Middlesbrough –TS13BA, Tees Valley, U.K.
The objective of the present investigation is characterization of ammonia producing
bacteria (Bacillus sp.) and its impact on biodeterioration of cupronickel alloy 90:10. It is
well known that iron sulphate and molybdenum are good inhibitors used in cooling water
system. The role of interaction between inhibitor and ammonia producing bacteria on
corrosion of cupronickel 90:10 was studied. Cupronickel coupons were immersed in
Chavara rare earth containing backwater for a period of six months. The predominated
ammonia producing bacteria were isolated from the six months old biofilm. A total of
four ammonia producing Bacillus sp. (AG1-EU202683; AG2-EU202684; AG3-
EU202685 and AG4-EU202686) were isolated from the biofilm and identified by 16S
rRNA gene sequencing. The corrosion rate of cupronickel in water (System I) was in the
range of 0.046mm/year and 0.052mm/year. In the control system II (Chavara water and
nitrogen free medium without bacteria), the corrosion rate was 0.008 mm/year whereas in
the presence of Bacillus sp. AG1 and AG3, the corrosion rate was in the range of 0.023
and 0.030mm/year. These bacteria fixed atmospheric nitrogen for their cell protein
synthesis and converted into ammonia. Ammonia enhanced pH and ammonical solution
was formed in the presence of Bacillus sp. that acted as an etchant. The presence of some
anodic spots in the presence of bacteria was affected by ammonia and underwent pitting
1

corrosion. The present study reveals that Bacillus sp. encourage intergranular attack
without any stress to the cooling water system.
Keywords: Microbial corrosion, ammonia producing bacteria, Bacillus sp., copper alloys
intergranular attack, inhibitors, cooling water system
*Correspondence: Dr Maruthamuthu Sundram, Corrosion Protection Division,
Central Electro Chemical Research Institute (CECRI), Karaikudi – 630 006, India.
Email: biocorrcecri@gmail.com; Present address: Electrokinetic Research group, Electro
technology Research Institute, Changwon city, Gyengsangnam-do 641-120, Korea
Introduction
The biofilms are important in a wide spectrum of industrially relevant situations and lead
to microfouling and microbially influenced corrosion (Lapin-Scott and Costerton, 1989;
Bott et al., 1993). Copper and its alloys are most commonly used in the fabrication of
heat exchangers in cooling water system. The alloys depend on their natural oxide for
corrosion resistance. Corrosion of copper occurs with the outward movement of the
cuprous ion rather than the inward movement of oxygen (Rao and Nair, 1998). The slime
layer forms a sticky surface, which allows silt and other suspended particles to adhere to
the condenser tubes, there by enhancing the aggregation of deposits on the material
surface that increased fluid frictional resistance and heat transfer resistance (Bott et al.,
1995) in cooling water system. Pope et al., (1984) have documented MIC of cupronickel
(90:10), admirably brass and aluminum brass in cooling system of freshwater and
backwater. It was also reported that copper alloy condenser tubes had under-deposit
corrosion due to the formation of slime deposits and ammonia (Little and Wagner, 1996),
which led to stress corrosion cracking (Little et al., 1991). Iron sulphate and molybdate
(Oung et al., 1998) are added in many cooling water system as corrosion inhibitors
(Uhlig, 1994). It is also well known that iron and molybdate (Oelze et al., 2000; Smith et
al., 1999) act as co-factors for bacterial nitrogenase enzyme. Earlier metallurgical
analyses of the failed tubes revealed that the tubes were damaged due to stress corrosion
cracking (Agarwal et al., 2003; Venugopal and Rawat, 1991). However, no systematic
study has been carried out on the interaction between microbial biofilms and corrosion
2

inhibitors in a cooling water system. It was expected that the ammonia producing bacteria
can proliferate as biofilm in nitrogen free environment where it fixes the nitrogen from
the atmosphere and convert to ammonia. Hence the work was undertaken to find the
activity of ammonia producers on the corrosion of Cupronickel 90:10 in a nitrogen-free
water system in the presence of corrosion inhibitors like iron sulphate and molybdate.
Materials and methods
Sample collection
The cupronickel coupons 6cm x 4cm were immersed in samples obtained from the Chavararare
earth environment that extend over 22Km from Neendakara to Kayankulam in Kollam
district, Kerala, India for a period of 6 months. The six months old biofilm samples were
washed with sterilized water for removing pelagic bacteria in biofilm and scrapped from the
cupronickel (90:10) metal surface with the help of a sterile surgical knife. The biofilm sample
was collected in a sterile conical flask and the collected biofilm sample was stored in a ice
box and transported for microbiological analysis at CECRI- microbiological lab. The water
sample was collected from the site in a sterilized 10 liters polythene container and transported
to CECERI to carry out corrosion study in the laboratory. The chemical characteristics of
Chavara water were analysed by the standard method (Grasshoff, 1999). The location of the
study site in India is presented in Figure 1.
Bacterial Isolation and Identification
The samples were serially diluted using 9ml of sterile distilled water. Total viable bacterial
counts were enumerated by standard pour plate method using the nitrogen-free medium
(Cappucino and Sherman, 2005). The composition of nitrogen free medium used is as follows
(g/l): K2HPO4, 1.0; MgSO4.7H2O, 0.2; FeSO4.7H2O, 0.050; CaCl2.H2O, 0.1; Na2MoO4.2H2O,
0.0010 and Glucose, 10. The samples were mixed thoroughly and allowed to solidify. Then the
inoculated plates were incubated at 37°C. Triplicates were also maintained. Total bacterial
count was enumerated after 48-72 hours of incubation. The bacterial population was expressed
as colony forming units per cm 2 (CFU/cm 2 ).
Morphologically dissimilar colonies were selected and isolated from nitrogen-free
agar plates. The isolated colonies were purified using appropriate medium by streak plate
3

method. The pure cultures were stored in respective slants at 4°C to maintain the microbial
viability. Biochemical characterization of the isolates was carried out by API biochemical test
kit (Bio-Merieux, SA, France) and by standard tests.
16S rRNA gene sequencing and phylogenetic analysis
The genomic DNA was isolated from four isolates according to the procedure described by
Murmur et al. (1961) and the small subunit rRNA gene was amplified using the two primers
16S1 (5’-GAGTTTGATCCTGGCTCA-3’) & 16S2 (5’- CGGCTACCTTGTTACGACTT-
3’). The purified PCR product, approximately 1.5Kb in length was sequenced using five
forward and one reverse primer as described earlier (Reddy et al., 2000). The deducted
sequence was subjected to blast search for closest match in the database. The 16S rRNA gene
sequence of ammonia producing Bacillus sp. AG1-AG4 was submitted in the Gene bank. The
pair wise evolutionary distances were computed using the DNA DIST program with the
Kimura 2 parameter model (Kimura et al., 1980). The phylogenetic trees were constructed by
using four tree making algorithms the UPGMA, KITSCH, FITCH and DNAPARS of the
PHYLIP package (Felsentein et al., 1993). The stability among the clads of a phylogenetic
tree was assessed by taking 1000 replicates of the dataset and was analyzed using the
programs SEQBOOT, DNADIST, UPGMA and CONSENSE of the PHYLIP package.
Ammonia spot test and estimation
Bacterial isolates were tested for the production of ammonia in nitrogen-free broth. 10ml
nitrogen free broth in tubes were inoculated with freshly grown culture and incubated for 48-
72 hours at 28°C. Nessler ’ s reagent (0.5ml) was added in each tube. Development of white to
yellow colour indicated ammonia production (Cappucino and Sherman, 2005). The culture
samples were withdrawn at different times, centrifuged and filtered (through cellulose acetate
membranes; pore size, 0.45μm). An appropriate amount of supernatant or filtrate was tested
for the presence of ammonia by the indophenols method (Bali et al., 1992). The mixture was
incubated for 30 minutes at room temperature. The absorption value was measured at 625nm
by UV-spectrophotometers and the ammonia concentrations were estimated.
4

Observation by epi-fluorescence microscope
A loopful of culture was taken for the fluorescence studies. In a clean glass slide the smear
was prepared. Two drops of 0.01% aqueous solution of acridine orange was added to the
smear, incubated for 10 minutes, the excess stain was washed with sterile distilled water.
Morphology of the cells was studied under the oil immersion (100X) epi-fluorescene
microscope.
Corrosion studies
(a) Weight loss measurement
Cupronickel coupons of 5 x 1cm size with a hole on the top were used for weight loss
experiments. The coupons were machine polished to mirror finish, degreased with
trichloroethylene and rinsed with deionized water. Four systems were designed by using
Chavara water. System I consisted water alone. System II consisted of chavara water with
nitrogen-free medium without bacteria. System III had chavara water and nitrogen-free
medium along with bacteria AG1, whereas system IV consisted of chavara water and
nitrogen-free medium with bacteria AG3. The pre–weighed coupons were immersed in
Chavara water and nitrogen-free broth inoculated with and without bacteria. Three coupons
were exposed in a conical flask and duplicate cells were made for each system. Totally six
coupons were used for the weight loss for each system. The average weight loss and standard
deviations (SD) were calculated. The broth was continuously replaced once in 3 days to
maintain the biofilm alive. Nitrogen-free medium without bacteria acted as the control
system. The weight loss of the metal obtained in milligrams was converted to corrosion rate in
millimeter per year (mm/y) using the formula (Treseder et al., 1991).
437 X Weight loss (mg)
Corrosion rate (mm/y) = ----------------------------------
Area (cm 2 ) X Time (h)
Electrochemical studies
(a) Potential measurements
5

Open circuit measurements (OPC) were made for cupronickel (90:10) in nitrogen-free
medium with and without bacteria inoculated system. Three commercial saturated calomel
electrodes (SCE) were intercalibrated with the other at least once in a week and used.
Reference electrode was only used when their potential variation was within 5mV of the
others. OCP values were measured with time using a digital multimeter of high resistance.
(a) Polarization studies
Cupronickel coupons of 1 x 1cm dimension with an extended stem of 15cm length were used
for polarization studies. Specimens were polished to mirror finish by using emery paper 1/0
down to 4/5 and finally degreased with trichloroethylene followed by deionized water. Three
specimens were immersed in separate 250ml conical flasks containing nitrogen free broth and
were sterilized. AG1 and AG3 were inoculated in nitrogen-free medium and used as
experimental system and uninoculted specimen was used as a control system.
Polarization measurements were carried out potentiodynamically using modern PGP
201, potentiostat with voltammeter – 1 software employing a large platinum electrode,
saturated calomel electrode (SCE) as reference electrode and the cupronickel metal as
working electrode. The system was allowed to attain a steady potential value for 10 minutes.
The study state polarization was carried out from OCP to -200mV SCE and +200mV SCE
from the OCP separately using separate electrodes at a scan rate of 1800mV/hr. Polarization
study was done on the 20 th day of immersion period.
(b) Impedance studies
Electrodes of the same specification employed in the polarization studies were used for
impedance studies. Specimens were polished and degreased as described in the polarization
studies. Impedance studies were carried out using computer control EG & G system: M6310
with software M398. After attainment of steady state, an AC signal of 10mV amplitude was
applied and impedance values were measured for frequencies ranging from 0.1 Hz to 100
KHz. The values of Rt were obtained from Bode plots. Impedance measurements were taken
on the 20 th day of immersion.
Surface studies
(a) Fourier Transform infrared spectroscopy ( FTIR)
6

FTIR was used for the analysis of the biochemical characteristics of the absorbed
products collected from the metal surface exposed to control and experimental systems.
The spectrum was taken in the mid IR region of 400–4000 cm -1 . The spectrum was
recorded using ATR (attenuated total reflectance) technique. The sample was directly
placed in the zinc selenide crystal and the spectrum was recorded in the transmittance
mode.
(b) X-ray diffraction (XRD) and Scanning electron microscopy (SEM)- Energy dispersive Xray
spectroscopy (EDX) studies
Cupronickel coupons of 5 x 1cm 2 size were machine polished to mirror finish, decreased with
trichloroethylene, washed with deionized water and dried. The coupons were immersed in
nitrogen-free broth with and without AG1 and AG3 for 20days and the coupons were
removed and the nature of oxides formed on the metal surface were estimated by XRD and
SEM-EDX. The specimen prepared as described above was analyzed by X’pert PRO PAN
analyzed X-ray diffractometer with Syn Master 793 software to identify the corrosion
product. The XRD pattern was recorded using computer controlled XRD-system, JEOL, and
Model: JPX-8030 with C α K radiation (Ni filtered = 13418 A o ) at the range of 40kV, 20A.
The ‘peak search’ and ‘search match’ program built in software (syn master 7935) was used
to identify the peak table and ultimately for the identification of XRD peak. The same
specimens were examined at different magnifications [500X, 1000X and 2000X] by scanning
electron microscope [Model – Hitachi – S 3000 H]. The elements were identified by Energy
dispersive X-ray spectroscopy (EDX) model: Naron system SIX (Thermo electron
corporation).
(c ) Atomic force microscope (AFM)
The cupronickel 90:10 specimens of size 2.0cm x 0.6cm × 0.06cm were machine and
cloth polished to mirror finish to give a homogeneous surface then washed with
trichloroethylene, followed by deionized water and dried. The specimens were immersed
in nitrogen-free medium in the presence and absence of bacteria at 30ºC for 5 days. The
specimens were removed on the 6 th day and picked. The corrosion product was removed
and dried at room temperature, and then characterized by atomic force microscopy,
7

model Pico scan 2100 (Molecular Imaging, USA) using gold coated SiN3 cantilevers
(force constant 3 n/W) of 30 nm tip area.
Results and discussion
Chemical characteristics of Chavara water
The chemical characteristics of water are presented in Table I. The water quality data
showed that Chavara water contained chloride and oxygen in the range of 9000 ppm and
4.2-5.0 ppm respectively. pH of the water in the field was 7.2-7.8 and the total hardness
was 3000 ppm.
16S rRNA gene based identification
The bacterial density of six months old cupronickel biofilm was 2.13 x 10 5 CFU/cm 2 .
Ammonia producing Bacillus sp. were identified by biochemical test and presented in the
table -II and confirmed by 16S rRNA gene sequence analysis. The standard biochemical
analysis revealed that the isolates AG1 and AG3 were almost similar except two tests.
The isolate AG1 was catalase negative where as AG3 was catalase positive; moreover
AG3 produced pigment but AG1 did not. The isolate AG2 and AG4 had close similarity
except arabinose and lactose sugar fermentation. The blast results as well as the
phylogenic analysis (Figure 2) revealed that the Bacillus sp. AG1 (EU202683), AG2
(EU202684), AG3 (EU202685) and AG4 (EU202686) had 99% sequence similarity with
Bacillus anthracis (AY138382), Bacillus cereus (AE016877) and Bacillus thuringensis
(AF155955). Since Bacillus anthracis and B. cereus were phenotypically similar except
for the presence of PA gene in the plasmid, PA gene specific PCR was used to
differentiate the four cultures. The results were found to be negative for PA specific
primers. It clearly indicates that none of the four isolates were B. anthracis. So far there
is no report available for the effect of ammonia production by Bacillus sp. and its impact
on the corrosion behavior of cupronickel alloy 90:10. Gaylarde and Johnshon (1980)
8

eported that Desulfovibirio bacteria was capable of attacking freshly cleaned 90/10
alloys. Similar trends were noticed in the present study with Bacillus sp. AG1 and AG3.
These bacteria were resistant to copper toxicity and were capable of attacking mirror
polished copper coupon in the medium by the release of ammonia. Bacillus sp. as
manganese oxidizers was already reported as a dominating genus on copper coupons in
Tuticorin seawater (India) by Palanichamy et al., (2002). They suggested that Bacillus sp.
was resistant to copper toxicity by the formation of spores.
Ammonia spot test and estimation
Appearance of yellow colour in the nitrogen-free medium inoculated with Bacillus sp.
within 72 hours of incubation indicated ammonia production by Bacillus sp. Ammonia
production by the bacteria would have resulted in a sudden increase in pH of the medium.
pH of the medium in the presence and absence of the bacteria were measured and
presented in table-III. The pH of the nitrogen-free medium in the absence of Bacillus sp.
was 7.2. In presence of bacteria, the quantity of ammonia produced was in the range of
2.5 and 8.0ppm and the pH was in the range of 8.2 and 8.5. This showed that ammonia
production has caused an increase in pH of the medium. Ammonia concentration was
estimated by Rao and Nair (1998) in natural biofilms and they noticed that ammonia
generation was caused by nitrate reducing bacteria (NRB). The denitrifying bacteria
observed during the course of this study were Alcaligenes sp., Bacillus sp., Micrococcus
sp., Pseudomonas aeruginosa and Pseudomonas fluorescens. The estimated amount of
ammonia in the natural biofilms was in the range of 0.5 and 5.5 ppm. The present result
supports the observation made by Kanamori et al., (1990) who suggested that Bacillus
produced ammonia. Ahmad et al. (2007) also identified Bacillus sp. from different
rhizospheric soil and plant root nodules in the vicinity of Aligarh and found that 80% of
the isolates were ammonia producers. The present study is the first report on ammonia
production by Bacillus sp. and its impact on corrosion of cupronickel.
9

Weight loss
Corrosion rate of cupronickel 90/10 in Chavara water (India) in the presence or absence
of Bacillus sp. and nitrogen-free medium are presented in table IV. The corrosion rate of
cupronickel in Chavara water (control I) was in the range of 0.046mm/year and
0.052mm/year. In control system II (Chavara water with nitrogen free medium), the
corrosion rate was 0.008 mm/year whereas in the presence of (AG1 and AG3) Bacillus
sp., the corrosion rate was in the range of 0.023 and 0.030mm/year. Harrison and
Kennedy (1986) reported that corrosion rates exceeding 1.0mm/year were considered as
very high for copper alloys. So it could be concluded that the corrosion rate was lower in
the presence of Bacillus sp.
Potential measurement
Potential measurement versus time in the presence and absence of Bacillus sp. are
presented in figure 3. The initial potential was about -225mv SCE in the presence of
Bacillus sp. and it increased rapidly to –50mv within 5 days. After 11 th day, the potential
slowly shifted to negative side of the range (-190mV). But in the absence of Bacillus sp.,
it slowly shifted towards the positive side and reached about 50mV on the 11 th day. The
potential maintained at the range of -100mV on the 20 th day. The shifting of potential to
negative side in presence of Bacillus indicates that bacteria enhance corrosion when
compared to the control.
Polarization
Polarization data for cupronickel (90/10) in the presence and absence of Bacillus sp. are
presented in figure 4 and table V. In the control system (in the presence of nitrogen-free
medium) icorr (Corrosion current) was 7.69x10 -7 A/cm 2 . In the presence of Bacillus sp. in
the nitrogen-free medium, the icorr value was in the range of 1.21 X 10 -6 and 1.65 X 10 -6
A/cm 2 . The nature of the curve also indicates that Bacillus sp. enhances corrosion by the
enhancent of anodic reaction through ammonia production.
10

Impedance
Impedance spectroscopy data are presented in table VI and figure 5. In the control
system, Rt value was 32.93 kΩ/cm 2 , whereas in bacterial system, the Rt value was in the
range of 16.37 and 19.02 kΩ/cm 2 . It indicates that corrosion is higher in low resistance
systems. The nature of the curve in Bacillus sp. system also indicates that the corrosion is
due to activation control. Electrochemical impedance spectroscopy was used by Hashem
(2002) to study the effect of ammonia residuals on the corrosion of copper in seawater
polluted with ammonia and suggested that ammonia enhances the attack by dissolving the
complex with copper ion.
FTIR
FTIR spectrum for cupronickel with and without bacteria is presented in figure 6.
In the control system, a peak at 2355cm-1 was due to metal hydride bond (M-H). Peaks at
1630 cm -1 and 1117 cm -1 were due to the presence of metal –O-C and C-O (alcoholic)
stretching vibrations. The other peaks at 3288 cm -1 and 627 cm -1 were due to the presence
of hydrogen bond (O-H) and C-X bond stretches. In the presence of AG1, a peak at 3517
cm -1 was due to the N-H stretching vibration. A peak at 2767 cm -1 was correspond to the
OH- for COOH stretch. The peaks at 2354 and 1757 cm -1 were due to the presence of
metal-hydride bond (M-H) and metal- C-O respectively. The peaks at 1604, 1491 and
1265 cm -1 were due to the presence of Metal- O-C / N-H bending, N-H bending and
primary N-H (amine) stretch vibrations respectively. Peaks at 964 and 818 cm -1
corresponded to the -CH- out of bending and adjacent hydrogen atom vibration
respectively. In the presence of AG3, a peak at 3200 cm -1 was due to the N-H stretching
vibration. A peak at 2357 cm -1 corresponded to the metal - H (M-H). The peaks at 1757,
1627 and 1257 cm -1 were due to the presence of Metal- O-C / N-H bending, N-H bending
and primary N-H (amine) stretch vibrations respectively. Peaks at 1127 and 615 cm -1
were due to the C-O (alcoholic) and C-H out of plane/C-X stretch. FTIR study reveals
that the presence of nitrogen based compound was due to the attachment of bacteria or
physiological activity of ammonia producers on cupro-nickel 90:10.
11

Surface analysis by X-ray diffractometer
X-ray diffraction peak for corrosion products from control and Bacillus sp. (sample)
system are presented in figures 7-9. Ni2CuO3, Ni2O3 and CuO were noticed in all the
systems, whereas the maximum scale of intensity was in the range of 800 counts.
Besides, the high intensity peaks in the presence of bacteria (maximum 2500 counts) in
Bacillus sp. system when compared to the uninoculated system, it was also interesting to
note that the adsorption of SO4 complex also occurred in the presence of bacteria in the
nitrogen free medium. Xiao et al. (2007) noticed Cu2O (Cuprite) as a major corrosion
product when they exposed copper to drinking water environments for a period of six
months. They suggested that Cu2O oxidation to CuO increased with alkalinity, and
depended on time and pH. Also in the present study CuO is found to be the major
corrosion product at high pH electrolyte. Cassagne et al. (1990) concluded from their
detailed studies that breakdown of cuprous oxide by micropitting yields to initiation of
transgranular SCC. Rao and Nair (1998) noticed cuproammonium complex and copper
oxide as minor peaks on admiralty brass. But in the present study, cupro ammonium
complex could not be noticed while Cu2O(SO4) was noticed on the surface of cupronickel
90:10 in the presence of bacteria. Since cuproammoniun complex is an unstable
compound, it could not be detected in XRD.
Scanning Electron Microscopy
SEM results are presented in figures 10a, 10b, 10c and 10d. Uniform corrosion was
noticed in Chavara water (Fig 10 a). In the absence of bacteria in nitrogen-free medium,
an uniform film was noticed on cupronickel (figure 10b). Figure 10c and 10d show
pitting of cupronickel, which was exposed to the nitrogen-free medium along with
Bacillus sp. Moreover, etchings of grain boundaries were noticed in the figure 10d. It
indicates that Bacillus sp. encourages pitting corrosion as well as intergranular attack.
The present study also indicates that the adsorption of nutrients on cupronickel present in
the medium results in the formation of film and is broken by the bacteria.
12

EDX
Figures 11 and 12 show the spectrum received from EDX for cupronickel in the presence
and absence of Bacillus sp. in nitrogen free medium. The percentage weight of copper,
nickel, iron and oxygen were analyzed on the cupronickel and presented in table VII. In
the control system 87.68% of copper and 10.40% of nickel were noticed along with 2%
iron, where iron acts as passivator. In the experimental system, the center of pitted
surface had 90.53% of copper, 8.85% of nickel and 0.62% of iron. The iron content was
lower on pitted surface indicating the initiation of pitting at lower iron adsorbed area. At
point 2 (edge of the pit) the iron and copper contents were 0.19 and 79.96% respectively
whereas oxygen was 8.96%. It indicates that the edge consists of cuprous oxide on the
metal surface. The reduction of iron on the cupronickel in the presence of Bacillus sp.
may be due to the use of iron as nutrient by Bacillus sp., which is a cofactor for
nitrogenase enzyme (Smith et al., 1999). Besides, it can be inferred that the oxide film
was broken by biogenic ammonia and has resulted in pitting.
Atomic force microscope
Atomic force microscopic observations were done for cupronickel 90:10 in the presence
and absence of bacteria and are presented in the figures 13 a & b. The adsorption of
nutrients present in the nitrogen-free medium can be seen in the control system. The film
was heterogonous in nature (figure. 13a). It indicates the presence of inhibitor like iron
sulphate and molybdenum adsorbed on the metal surface. In the bacterial system pitting
and inter granular crack could be noticed. The depth of the largest pit was 1000nm
(Figure 13b).
According to the literature two reaction schemes are possible for tenorite
formation. The reduction of positive three valent nitrogen in nitrite ion to negative, three
valent in ammonia takes place (Todt et al., 1961; Kirchberg et al., 2001). Apart from the
nitrogenase enzyme, the denitrification process proposed by Tiedje et al. (1988) involves
four stages viz; reduction of nitrate to ammonia via nitrate, hyponitrite and
hydroxylamine. In the present study, it is assumed that nitrogenase enzyme is the most
important factor for nitrogen fixation (Abdel Wahab et al.,1975) from atmospheric di-
13

nitrogen with available iron and molybdenum that act as cofactors (Jacobson et al., 1986;
Smith et al., 1999). When Bacillus sp. produces ammonia, it is volatilized within a short
time with an increase in pH. Ammonical solution formed in the presence of Bacillus sp.
acts as an etchant. Since, nitrogen-free medium contains sulphate, it can be assumed that
the ammonical solution may enhance the production of ammonium persulphate that is a
good etchant for copper alloys (Tiedje et al.,1988). Analysis of ammonia and pH upto 20
days, confirms the continuous formation of etchant solution in the presence of Bacillus
sp. EDX analysis indicates the presence of oxide films on the edges of pit and on the
fracture surface. This might be due to the presence of CuO that was noticed as major
peaks in XRD. Venugopal and Rawat (1991) reported the profile of cracks observed on
the inner side of the admiralty brass condenser tube in a nuclear cooling water system.
Both circumferential and longitudinal cracks were observed on the external surface of the
failed tubes and pits were observed close to the cracked regions. The nature of the crack
indicated that admiralty brass tubes had failed due to SCC. But in the present study intergranular
crack was noticed. It is due to the biogenic ammonical action (Syrett and Coit,
1983) of Bacillus sp. After ammonia formation the following possible reaction takes
place (Todt et al., 1961; Mori et al., 2005).
NH3+H2O NH4 + + OH -
Cu +2OH - Cu (OH) 2 +2e- (oxidation process)
H2O +1/2O2 +2e- 2OH -
Cu (OH) 2 + 2NH3 + 2NH4 + [Cu (NH3)4] 2+ + 2H2O
The enhancement of anodic current in polarization study can be explained that the
corrosion on cupronickel is due to the oxidation of Cu(OH)2. Besides, the inter granular
attack is formed due to over saturation of the [Cu (NH3)4] 2+ ions (Kirchberg et al.,2001;
Mori et al., 2005 and Kuznika and Junik , 2007).
Cu [(NH3)4] 2 + H2O CuO + 2NH3+ 2NH4 +
14

Formation of CuO at the metal surface is the critical step for the occurrence of stress
corrosion cracking (SCC) that was noticed on XRD which supports with the observation
made by Mori et al (2005).
The present study supports the observation made by Pope et al., (1984) who
suggested that many organisms form NH3 from the metabolism of amino acids. These
NH4 ions in solution may play a role in the corrosion of copper alloys. Nitrogenase
enzyme is the major causative factor for the production of ammonia (Smith et al., 1999).
Nitrogenase works at room temperature where nitrogen gas requires reduced ferredoxin
or flavodoxin and ATP as substrates. Reduced ferredoxin or flavodoxin transfers
electrons to the azoferredoxin. At the expense of energy from ATP hydrolysis the
potential of redox groups of the enzyme is lowered further and finally a super-reduced
molybdoferredoxin is formed, which binds N2 and reduces it stepwise to ammonia. The
binding of the dinitrogen molecule occurs probably by insertion into a metal-hydride
bond involving molybdenum (Gottsschalk, 1985). A MO = N-NH2 group functions
probably form as an intermediates. Only the azoferredoxin component of nitrogenase has
ATP-binding sites and ATP hydrolysis is primarily associated with the formation of a
super-reductant. Using cell-free nitrogenase preparations, the reduction of N2 has been
found to be coupled to the hydrolysis of an enormous amount of ATP, in the order of 16
ATP per N2.
N2 + 6H + 16 ATP 2NH3 + 16 ADP +16 Pi
It is also well known that iron sulphate is a good passivator for cupronickel,
which adsorbs on the metal surface and improves passivity. Some anodic spots on the
metal surface will be affected by ammonia and undergo pitting corrosion. AFM study
supports the observation made in SEM and reveals that the formed pit depth was in the
range of 1000 nm within 5 days. Subsequently the ammonical solution enhances the
attack at the grain boundaries and accelerates intergranular corrosion without any stress.
It concludes that Bacillus sp. may encourage intergranular attack without any stress in
cooling water system. The mechanism has been explained in figure 14. When the alloys
stressed in tension is also exposed to a corrosive environment, the ensuring localized
15

electrochemical dissolution of metal, combined with localized plastic deformation, opens
up a crack (Agarwal et al., 2002). Protective films that form at the tip of the crack
rupture, causing fresh anodic material to be exposed to the corrosive medium and SCC is
propagated (Kuźnicka and Junik, 2007).
Conclusions
Cupronickel is a very good engineering alloy used as heat exchanger tubes in seawater
and freshwater cooling systems. It is prone to intergranular corrosion (IGC) by Bacillus
sp. in a nitrogen-free environment. Hence, the addition of iron sulphate and molybdate as
corrosion inhibitors in cooling water systems is questionable. If environment does not
contain nitrogen, Bacillus sp. will fix nitrogen (Mishra and Taquikhan, 1987) from the
atmosphere and enhances the production of ammonia in cooling water system. The
present study creates an awareness on the impact of corrosion inhibitors on
microbiologically influenced corrosion on copper alloys in cooling water systems of
process industries.
Acknowledgments
The authors wish to thank Mr.R.Ravishanker, Mr.A.Rathiskumar and Miss.S.Krithika
for their assistance in utilization of facility SEM, AFM and XRD in Instrumentation
division CECRI, India respectively. We also thank Dr.Thahira Rahman, Newcastle
University, UK for her suggestions and advice. Authors thank to BRNS (DAE) for
sponsoring a project (GAP26/05) entitled “Evaluation of engineering materials in rare
earth environment with special reference to biofouling research”. We also thank
Dr.V.P.Venugopalan and Dr.T.S.Rao, Biofouling and Biofilm Processes Section, Baba
Atomic Research Centre, Kalpakkam, India for their valuable discussions on this work.
This paper was selected as Best paper award in 13th National Corrosion Council of India
(NCCI) held at Holy cross college, Trichy from 30 Nov & 1 Dec.2007.
16

References
Abdel Wahab AM. 1975. Nitrogen fixation by bacillus strains isolated from the
rhizosphere of Ammophila arenaria. Plant and soil. 42:703-708.
Agarwal DC. 2002. Stress corrosion in copper-nickel alloys: influence of ammonia.
Br. Corros. J. 37:267-275.
Agarwal DC. 2003. Effect of cyclic stresses on stress corrosion cracking of Cu-Ni alloy.
Corros. Engi. Sci.Tech. 38: 275-285.
Ahmed F, Ahmad I, Khan MS. 2007. Screening of free living rhizosphric bacteria for
their multiple plant growth promoting activities. Microbiol. Res.( in press).
Bali A, Blanco G, Hill S, Kennedy C. 1992. Excretion of Ammonium by a ninth mutant
Azotobacter vinelandii fixing Nitrogen. Appl. Environ. Microbiol. 58: 1711-1718.
Bott T R. 1995. Fouling of Heat Exchangers. Elsevier. New York.
Bott T R. 1993. Aspects of biofilm formation and destruction. Corros. Rev. 11: 2-24.
Cappucino J, Sherman N. 2005. Microbiology: A laboratory manual. 6 th Edition.
Cassagne TB, Kruger J, Pugh EN. 1990. Oxide formation and transgranular stress
corrosion cracking of copper, 11th ICC, AIM, Milan. 3.241–3.247.
Felsentein J. 1993. PHYLIP (Phylogeny Inference Package) Version 3.5c. 21
Department of Genetics, University of Washington, Seattle, USA.
Gaylarde CC, Johnson JM. 1980. The importance of microbial adhesion anaerobic metal
corrosion. Academic Press. p. 511-513.
Gottsschalk G. 1985. Bacterial metabolism. Pub: Spinger-Verlag. New York Inc.
17

Grasshoff K, Kremling K, Ehrhardt M. 1999. Methods of seawater analysis, Pub. Wiley -
vch, Germany.
Harrison JF, Kennedy KW. 1986. Advances in the control of copper and copper alloy
corrosion in chlorinated cooling waters. American Power Conference. Illinois Institute of
Technology, Illinois. U.S.A. April 14-16.
Hashem Al, Carew J. 2002. The use of electrochemical impedance spectroscopy to study
the effect of chlorine and ammonia residuals on the corrosion of copper-based and nickelbased
alloys in seawater. Desalination. 150:255-262.
Jacobson MR, Premakumar R, Bishop P E. 1986. Transcriptional Regulation of Nitrogen
Fixation by Molybdenum in Azotobacter vinelandii. J.Bacteriol.167:480-486.
Kanamori K, Weiss RL, Roberts J D. 1990. Efficiency factors and ATP/ADP ratios in
nitrogen-fixing Bacillus polymyxa and Bacillus azotofixans. J. Bacteriol. 172: 1962-
1968.
Kimura MA. 1980. Simple method for estimating evolutionary rates of base 12
Substitutions through comparative studies of nucleotide sequences. J. Mol. Evol. 6: 11-
13-20.
Kirchberg K, Kupfer- and Kupferlegierungen. 2001. Korrosion and Korrosionsschutz,
vol. 2: Korrosion der verschiedenen Werkstoffe, E. Kunze (Ed.).Wiley-VCH. Weinheim.
Germany. 1199.
Kuźnicka B, Junik K. 2007. Intergranular stress corrosion cracking of copper - A case
study. Corros.Sci., 49 : 3905-3916.
Lapin-Scott, HM, Costerton JW. 1989. Bacterial biofilms and surface fouling.
Biofouling. 1:323-342.
18

Little B, Wagner P. 1996. An overview of microbiologically influenced corrosion of
metals and alloys used in the storage of nuclear wastes. Can. J. Microbiol. 42: 367-374.
Little B, Wagner P, Ray R, Pope R, Scheetz R. 1991. Biofilms: an ESEM evaluation of
artifacts introduced during SEM preparation. J. Ind. Microbiol. and Biotechnol. 8: 213-
222.
Mishra S, Taquikhan MM. 1987. Correlation of growth kinetics with N2 fixing capacity
of bacterial isolates from Arabian sea. J. Mar. Sci. 16:143-145.
Mori G, Scherer D, Schwentenwein S, Walbichler P. 2005. Intergranular stress corrosion
cracking of copper in nitrate solutions. Corros.Sci. 47: 2099-2124.
Murmur J. 1961. Procedure for the isolation of deoxyribonucleic acid from
microorganisms. J. Mol. Biol. 3: 208-218.
Oelze J. 2000. Respiratory protection of Azotobacter species: is a widely hold hypothesis
unequally supported by experimental evidence? FEMS. Microbiol. Rev. 24:321-333.
Oung JC, Chin SK, Shih HC. 1998. Mitigating steel corrosion in cooling water by
molybdate based inhibitors. Corrosion prevention and control. 45: 156-162.
Palanichamy S, Maruthamuthu S, Manickam ST, Rajendran A. 2002. Microfouling of
manganese-oxidizing bacteria in Tuticorin harbour waters. Current. Sci. 82:865-869.
Pope DH, Duquette DJ, Johannes AH, Wayner PC. 1984. Microbiologically influenced
corrosion of industrial alloys. Mater. Perform. 23: 14-18.
Rao TS, Nair KVK. 1998. Microbiologically influenced stress corrosion cracking failure
of admiralty brass condenser tubes in a nuclear power plant cooled by freshwater. Corros.
Sci. 40:1821-1836.
19

Reddy GSN, Agarwal RK, Mastumoto GI, Shivaji S. 2000. Arthrobacter flavus sp .nov.,
a psychrophilic bacterium isolated from a pond in MnMurdo Dry valley, Antarctica. Int.
J. Syst. Evol. Microbiol. 50:1513-1526.
Smith B E, Durrant M C, Fairhurst S A, Gormal C A, Gronberg K L C, Henderson R A,
Ibrahim S K, Gall T Le, Pickett CJ. 1999. Exploring the reactivity of the isolated ironmolybdenum
cofactor of nitrogenase. Coord. Chem. Rev. 185:669-687.
Syrett BC, Coit RL. 1983. Causes and prevention of power plant condenser tube failures.
Mater. Perform. 22:44-50.
Tiedje JM. 1988. Ecology of denitrification and dissimilatory nitrate reduction to
ammonium. Biology of Anaerobic Microorganisms. ed. AJB. Zehnder. Academic Press.
New York. 179.
Todt, F. 1961. Kofrosion und Korrosionsschutz. Walter de Gruyter. Berlin. 237.
Treseder R S, Aboian R, Munger C G. 1991. NACE. Corrosion Engineer’s Reference
Book, 2na Ed. 168-170.
Uhlig HH. 1994. Corrosion Handbook (4 th Ed) John Wiley. New York.
Venugopal K, Rawat MS. 1991. Report on Failure analyses of condenser tubes from
RAPS-II turbine condenser. BHEL R& D, Hyderabad.
Xiao W, Hong S, Tang Z, Seal S, James S. 2007. Effects of blending on surface
characteristics of copper corrosion products in drinking water distribution systems.
Corros.Sci. 49:449-468.
20

Table I. Physico-chemical characteristics of Chavara pond water
Sl. No.
Physico chemical characteristics
21
observation
1. Colour Colourless
2. Odour Odourless
3. Taste Tasteless
4. Temperature 30.4 o C
5. pH 7.2-7.8
6. Total solids 337 mg/l
7. Total dissolved solids 288 mg/l
8. Total suspended solids 49 mg/l
9. Dissolved oxygen content 4.2-5.00 mg/l
10. Chloride 9000 ppm
11. Total Hardness 3000 ppm
12. Total alkalinity (Bicarbonates) 140 ppm
13. Carbonates Below the measurable range
14. Hydroxides Below the measurable range
15. Sulphate 16 ppm
16. Calcium 50 mg/l
17. Magnesium 78.67 mg/l
18. Sodium 104.63 mg/l

Table II. Biochemical test for ammonia producing Bacillus sp.
S.No
Name of the test performed AG1
(EU202683)
AG2
(EU202684)
AG3
(EU202685)
AG4
(EU202686)
1 Gram staining + + + +
2 Shape Rod Rod Rod Rod
3 Motility test + + + +
4 Catalase - - + +
5 Oxidase + + + +
6 ONPG + + + +
7 Lysine decaroxylase + + + +
8 Urease - - - -
9 Deamination - - - -
10 H2S Production - - - -
11 Citrate utilization + + + +
12 Pigment production - - + -
13 Nitrate reduction + + + +
14 Indole - - - -
15 Methyl red - - - -
16 VP - - - -
17 Malonate + + + +
18 Esculin hydrolysis + + + +
19 McConkey + - + -
20 Ammonia spots + + + +
21 Starch hydrolysis + - + -
Growth on
22 Sucrose + + + +
23 Arabinose - - + +
24 Adonitol - - - -
25 Rhamnose + + + +
26 Cellobiose + + + +
27 Melibiose + + + +
28 Saccharose + + + +
29 Raffinose + + + +
30 Trehalose + + + +
31 Glucose + + + +
32 Lactose + - + -
33 Fructose + + + -
34 Maltose - - - -
35 Galactose + + + +
36 Xylose + + + +
37 Mannitol + + + +
+ Positive, - Negative
22

Table III. Estimation of Ammonia and pH in nitrogen free medium with and
without bacteria.
S. no Bacterial strain
Concentration of
ammonia (ppm)
level
23
pH
1 CONTROL 0.00 7.2
2 AG1 4.6 8.4
3 AG2 2.5 8.2
4 AG3 8.0 8.5
5 AG4 5.0 8.4

Table IV. Corrosion rate of cupronickel (90/10) in different systems
S. no System
1
2
3
4
5
Control (System-I)
500ml of Chavara water
immersion on
cupronickel(90/10)
coupon (5cm x 1cm)
500ml of Chavara water
immersion on
cupronickel(90/10)
coupon (5cm x 1cm)
Control (System -II)
500ml of nitrogen free
medium broth and
cupronickel immersion
without (bacteria)
(System III)
500ml of nitrogen free
medium broth and
cupronickel immersion
and inoculated with
bacteria (AG1)
(System IV)
500ml of nitrogen free
medium broth and
cupronickel immersion
and inoculated with
bacteria (AG3)
Immersion
periods (day)
24
Average
weight
loss (mg)
Corrosion
rate
(mm/year)
10 11.18 ±.2 0.046
18 23.13 ±.3 0.052
20 4.1 ±.3 0.008
20 11.4 ±.3 0.023
20 14.7 ±.2 0.030
Form of
corrosion
Uniform
corrosion
Uniform
corrosion
Slight
uniform
corrosion
Pitting
corrosion
Pitting
corrosion

Table V. Polarization studies for cupronickel in different systems with and without
bacteria
S.no System
Ecorr
(mv)
Ba
(mv/decade)
25
Bc
(mv/decade)
Icorr(Amp/cm 2 )
1 Control -100 70 -123 7.69x10 -7
2 AG1 -180 87 -72 1.21x10 -6
3 AG3 -106 80 -80 1.65x10 -6
Table VI. Impedance spectroscopy studies for cupronickel in different systems
with and without bacteria
S.no System Rs ohms.cm 2
1
2
3
Control
nitrogen free
medium
nitrogen free
medium +
AG1 strain
nitrogen free
medium +
AG3 strain
Rct
k.ohms.cm 2
Rt
k.ohms.cm 2
Cdl F/cm 2
133 33.06 32.93 4.98x10 -4
114 16.48 16.37 7.07x10 -4
102 18.01 19.02 6.08x10 -4

Table VII. Energy Dispersive X-ray Spectroscopy for cupro-nickel in presence /
absence of Bacillus sp. system
S.no Position Copper% Nickel% Iron% Oxygen%
1
Control
Whole area
87.68 10.40 2.00 0.00
2 Point-1 90.53 8.85 0.62 0.00
3 Point-2 76.96 8.60 0.19 8.96
26

Figure 1. Bacterial collection from cupronickel 90:10 at Chavara in India.
27

100
Figure 2. UPGMA phenogram showing the phylogenetic position of Ammonia
producing Bacillus strain AG1-AG4 based on 16s rRNA gene sequence analysis.
Bootstrap values are given at the nods.
32
94
17
100
100
Bacillus subtilis (DQ452512)
62
Bacillus licheniformis (AY971527)
Bacillus amyloliquefaciens (AY651023)
56
Bacillus polyfermenticus (DQ659145)
Bacillus pichinotyi (AF519464)
63 Bacillus carboniphilus (AB021182)
96 Bacillus smithii (Z26935)
92 Bacillus alveayuensis (AY606801)
96
Bacillus eolicus (AJ504797)
Bacillus megaterium (DQ275182)
100
Bacillus flexus (AB021185)
AG1 Bacillus.sp (EU202683)
Bacillus anthracis (AY138382)
100 62 AG4 Bacilus.sp (EU202686)
100
98
AG2 Bacillus.sp (EU202684)
95
AG3 Bacillus.sp (EU202685)
Bacillus cereus (AE016877)
53
Bacillus thuringiensis (AF155955)
Serratia marcescens (AY498856)
68
Serratia rubidaea (AJ233436)
42
Serratia odorifera (AJ233432)
55
Serratia entomophila (AJ233427)
78
Serratia ficaria (AJ233428)
78
K.planticola (X93215)
79
Citrobacter freundii (AJ233408)
Enterobacter cloacae (AJ251469)
86
Klebsiella pneumoniae (Y17656)
Serratia fonticola (AJ233429)
48 Serratia plymuthica (AJ233433)
62 Serratia proteamaculans (AJ233434)
99
Serratia grimesii (AJ233430)
28
Bacillus pumilus (AY876289)

Figure 3. Potential measurement for cupronickel 90:10 in presence of Bacillus sp.
and control system.
29

Figure 4. Polarization studies for cupronickel in different systems of control and
bacterial systems with nitrogen free medium.
30

Figure 5. Impedance spectroscopy studies for cupronickel in presence of
Bacillus sp. and control system with nitrogen free medium.
31

Figure 6. FTIR study for cupronickel in presence and absence of Bacillus sp.
32

Figure 7. XRD spectrum for cupronickel (90:10) in control system (nitrogen free
medium).
33

Figure 8. XRD spectrum for cupronickel (90:10) in presence of Bacillus sp. (AG3)
with nitrogen free medium.
34

Figure 9. XRD spectrum for cupronickel (90:10) in presence of Bacillus sp. AG1
with nitrogen free medium.
35

Figure 10. Cupronickel metal surface analysis by Scanning electron microscope
(SEM) for various systems.
(a) Chavara water system (b) Control (NFM)
(c) Bacterial system (AG1) (d) Bacterial system (AG3)
IG attack
36
Pitting corrosion

Figure 11. Energy Dispersive X-ray Spectroscopy analysis for cupronickel in
presence of Bacillus sp.
37
1
2

Figure 12. Energy Dispersive X-ray Spectroscopy analysis for cupronickel
(90:10) in control system (nitrogen free medium).
38

Figure 13. AFM (Atomic force microscope) analysis for cupronickel in different
systems.
(a) Control (nitrogen free medium)
(b) Sample system (Nitrogen free medium with bacteria)
39

Figure 14. Mechanism of MIC on Cupronickel (90:10) in presence of Bacillus sp.
40
Crack initiation