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

net/publication/272360273

On the bacterial communities associated with the corrosion product layer

during the early stages of marine corrosion of carbon steel

Article  in  International Biodeterioration & Biodegradation · April 2015

DOI: 10.1016/j.ibiod.2015.01.003

CITATIONS READS

47 408

6 authors, including:

Isabelle Lanneluc René Sabot

La Rochelle Université La Rochelle Université
60 PUBLICATIONS   1,345 CITATIONS    85 PUBLICATIONS   2,070 CITATIONS   

SEE PROFILE SEE PROFILE

Marc Jeannin Philippe Refait

La Rochelle Université La Rochelle Université
124 PUBLICATIONS   2,330 CITATIONS    192 PUBLICATIONS   6,776 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Plastic deformation effects on electrochemical behavior of polycristallin nickel View project

TALINE View project

All content following this page was uploaded by René Sabot on 30 June 2016.

The user has requested enhancement of the downloaded file.

 International Biodeterioration & Biodegradation 99 (2015) 55e65

Contents lists available at ScienceDirect

International Biodeterioration & Biodegradation

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

On the bacterial communities associated with the corrosion product

layer during the early stages of marine corrosion of carbon steel

Sabot b, c, Marc Jeannin b, c,
Isabelle Lanneluc a, c, Mikael Langumier a, b, c, Rene

a, c, *
Philippe Refait b, c, Sophie Sable
a
Littoral, Environnement et Soci
et

es, UMR 7266 CNRS-Universit
^t. Marie Curie, Av. Michel Cr

e de La Rochelle, Ba epeau, F-17042 La Rochelle Cedex 01, France
b
Laboratoire des Sciences de l'Ing

enieur pour l'Environnement, UMR 7356 CNRS-Universit
e de La Rochelle, Ba^t. Marie Curie, Av. Michel Cr

epeau,
F-17042 La Rochelle Cedex 01, France
c


Federation de Recherche en Environnement pour le D
eveloppement Durable, FR 3097 CNRS, France

a r t i c l e i n f o a b s t r a c t

Article history: Carbon steel coupons were immersed for 1e8 weeks at ~1 m deep in a French harbor of the Atlantic
Received 10 October 2014 coast. The resulting corrosion product layers were characterized by m-Raman spectroscopy. They con-
Received in revised form sisted of an inner stratum of sulfated green rust covered by an outer stratum of lepidocrocite (pre-
9 January 2015
dominantly), goethite and magnetite. Mackinawite FeS was detected after one month of immersion, but
Accepted 9 January 2015
only locally. The bacterial characterization of these layers was coupled to this analysis. The diversity of
Available online 28 January 2015
the culturable bacteria grown in liquid media directly from biofilms or isolated on agar plates was
evaluated by 16S rRNA gene sequencing. The microbial diversity was also estimated, without cultivation
Keywords:
Microbiologically influenced corrosion
and after Temporal Temperature Gradient Electrophoresis (TTGE) separation. A moderate biodiversity
Steel was found in all cases and different bacteria associated with the redox cycles of Fe and S were identified.
Seawater Roseobacter, Erythrobacter and Bacillus were dominant among culturable bacteria whereas Sulfurimonas
Total and culturable bacteria autotrophica, a sulfur-oxidizing bacterium, was detected among the total bacterial community at all
TTGE immersion times. After one month of immersion, sulfide-producing bacteria were detected, e.g. Desul-
Raman spectroscopy fovibrio profundus in liquid mixed cultures and Sulfurospirillum arcachonense in the total bacterial com-
munity, in agreement with the local identification of FeS by m-Raman spectroscopy.
© 2015 Elsevier Ltd. All rights reserved.

Introduction
According to the phenomenological model proposed by
Melchers and Jeffrey (2005) and Melchers and Wells (2006), the
corrosion process of carbon steel in seawater involves two main
periods. During the first period, called the aerobic period, the
corrosion rate is controlled by the reduction of dissolved O2
(Melchers and Jeffrey, 2005). The thickness of the corrosion product
layer increases and during the last phase of the aerobic period, the
corrosion rate is controlled by the diffusion of O2 through this layer.
The consumption of O2 by the aerobic micro-organisms leads to the
* Corresponding author. Littoral, Environnement et Socie
te

s, UMR 7266 CNRS-
Universite
de La Rochelle, Ba
^t. Marie Curie, Av. Michel Cre

peau, F-17042 La Rochelle
Cedex 01, France. Tel.: þ33 5 46 45 82 46; fax: þ33 5 46 45 82 65.
E-mail addresses: isabelle.lanneluc@univ-lr.fr (I. Lanneluc), m.langumier@
hotmail.fr (M. Langumier), rene.sabot@univ-lr.fr (R. Sabot), marc.jeannin@univ-lr.
fr (M. Jeannin), philippe.refait@univ-lr.fr (P. Refait), sophie.sable@univ-lr.fr

).
(S. Sable
second period, called the anaerobic period. The steel surface and
the inner part of the corrosion product layer are then in anoxic
conditions. The beginning of the anaerobic period is associated
with an increase of the corrosion rate attributed to micro-
organisms, notably sulfate-reducing bacteria (SRB). During the
anaerobic period, the corrosion process is then basically a micro-
biologically influenced process and it was proposed that the
corrosion rate could be controlled by the transport of nutrients
(Melchers and Wells, 2006; Melchers and Jeffrey, 2012). For
instance, during the last phase of the anaerobic period of the
phenomenological model, the corrosion rate could be controlled by
the transport of nutrients through the corrosion product layer.
Numerous studies were then devoted to the understanding of
the mechanisms associated with the increase of the corrosion rate
in the anaerobic period. The most recent studies demonstrated that
electroactive SRB, able to use Fe
as the only source of electrons for
sulfate reduction, could accelerate significantly the corrosion pro-
cess in very specific conditions (Enning et al., 2012; Venzlaff et al.,

http://dx.doi.org/10.1016/j.ibiod.2015.01.003
0964-8305/© 2015 Elsevier Ltd. All rights reserved.
56 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
2013; Yu et al., 2013). Besides, peculiar bacterial consortia are
suspected to be responsible for accelerated localized corrosion
phenomena. In particular, it was shown that the association be-
tween SRB and sulfur-oxidizing bacteria (SOB) contributed to the
accelerated low water corrosion (ALWC) of carbon steel (Beech and
Campbell, 2008). Iron reducing bacteria (IRB) could also be
involved, but their influence is controversial. In some cases,
corrosion rates were lowered when IRB were associated with SRB
(Duan et al., 2008) while in other cases these micro-organisms
enhanced corrosion in the absence or the presence of other bac-
teria (Herrera and Videla, 2009). This result however indicates that
bacteria associated with the redox cycle of iron could play a role in
the microbiologically influenced corrosion process. Other micro-
organisms may influence the rate of cathodic and/or anodic re-
actions, for instance via the hydrogenase enzyme that catalyzes
proton reduction (Bryant and Laishley, 1990; Mehanna et al., 2008)
or by direct electron transfer with the steel surface (Mehanna et al.,
2009).
According to the phenomenological model (Melchers and
Jeffrey, 2005; Melchers and Wells, 2006), and in agreement with
Malard et al. (2008), the micro-organisms may have only little in-
fluence on the corrosion mechanisms and kinetics during the aer-
obic period. However, the evolution with time of the bacterial
communities forming a biofilm on the steel surface during this
period may provide information on the initiation of the microbio-
logically influence corrosion processes (MIC). This evolution of the
bacterial communities is necessarily associated with that of the
corrosion product layer that develops rapidly on the steel surface
(the initial corrosion rate can reach 5.7 mm per year according to
Memet, 2000). So, it is likely that the evolution of this mineral layer
influences that of the micro-organisms, and vice-versa. For
instance, the combined study of both corrosion products and bac-
teria, performed at the beginning of the anaerobic period, that is
~12 months of immersion, revealed that SRB were associated with
FeS and sulfated green rust (GR(SO2 II III
4 ) ¼ Fe 4Fe 2(OH)12SO4$8H2O)
uniformly all over the steel surface (Pineau et al., 2008).
So, the main aim of our study was to characterize thoroughly
both bacterial communities and solid phases present in the corro-
sion product layers during the early stages of marine corrosion of
carbon steel. For this purpose, carbon steel coupons were perma-
nently immersed at a constant depth (~1 m) in a French Atlantic
harbor (i.e. in natural seawater) for 1 week to 2 months. At the end
of each immersion period, the layers covering the steel surface
were studied via a combination of physico-chemical investigations
with microbiological and molecular biology techniques. The
corrosion products were characterized by m-Raman spectroscopy
while the associated bacteria were studied simultaneously. A
combination of culture-dependent (traditional culture techniques)
and culture-independent approaches (16S rDNA PCR amplification
and TTGE) was used. The culture-independent approach was
required because of the limitations of culture-based profiling of
microbial communities. The more traditional culture-based
methods were used in the attempt to obtain bacteria representa-
tive of biocorrosion phenomena that can be grown in culture and
thus used for further experiments.
Materials and methods
Sampling of steel coupons immersed in seawater
Carbon steel coupons (70  70  6 mm) were immersed in one
of the harbor of La Rochelle (Atlantic coast, France) for 1 week to
2 months at a constant depth of 1 m. The approximate steel
composition (in weight %) was: 98.2% Fe, 0.122% C, 0.206% Si, 0.641%
Mn, 0.016% P, 0.031% S, 0.118% Cr, 0.02% Mo, 0.105% Ni and 0.451%
Cu. The surface was shot blasted (Sa 2.5, angular shot) to obtain a
roughness value of 50e70 mm, degreased with acetone and dried.
The coupons were disposed in a Teflon holder and set in the im-
mersion site. At the end of the experiment, the corroded coupons
were removed, carried to the laboratory in sealed bags filled with
seawater and immediately processed. For each immersion time,
one coupon was used for microbiological analysis in aerobic con-
ditions, a second was used for coupled m-Raman analysis and
microbiological analysis in anaerobic conditions and two other
coupons were stored at 20
C for DNA extraction. The experi-
mental procedure is summarized in Fig. 1.
Preparation of corrosion product layers for microbiological analysis
The corrosion product layer was scraped from the corroded
coupons with a scalpel, weighed and then resuspended in 2 ml of
synthetic sterile seawater (NaCl 23.4 g l1; KCl 1.5 g l1;
MgSO4$7H2O 1.2 g l1; CaCl2$2H2O 0.15 g l1; CaCl2 0.15 g l1).
Some samples were then subjected to mechanical grinding with
glass beads for 1 min to dissociate adherent bacteria from the
particles of corrosion products. This step allowed us to increase the
number of bacteria counted by microscopy as well as by culture on
solid media (data not shown). These samples were used to
enumerate total and culturable bacteria and to inoculate the
different media (solid and liquid).
In order to determine the amount of water in each sample,
approximately 3e4 g of each sample were dried at 80
C and
weighed. All bacterial concentrations were expressed as the num-
ber of bacteria per gram of dry sample.
Numeration and isolation of culturable bacteria (Fig. 1, “path 1”)
Isolates were obtained from the corrosion product layers by
inoculation of 100 ml of serial dilutions (101 to 106) on solid
culture media (with 1.2% agar). Marine Agar (Difco) was used in
aerobic and anaerobic conditions and Baar modified solid medium
for sulfate reducers (Atlas, 2005) was used in anaerobic conditions.
Baar modified solid medium contained: MgSO4 2 g l1; sodium
citrate 5 g l1; CaSO4$2H2O 1 g l1; NH4Cl 1 g l1; NaCl 25 g l1;
K2HPO4 0.5 g l1; sodium lactate 3.5 g l1; yeast extract 1 g l1; agar
12 g l1 (pH 7.5). After sterilization (115
C, 20 min), Baar modified
medium was supplemented with Fe(NH4)2(SO4)2 1 g l1. After
3 weeks at 30
C in an anaerobic incubator (Don Whithey Scientific
limited MAC500, atmosphere N2/H2/CO2, 80/10/10) or in aerobic
conditions, bacterial growth was evaluated by counting the number
of colony forming units (CFU) (two replicates). The results are
expressed in CFU g1 of dry sample. Purification of the isolates was
achieved by selecting colonies with different morphologies. The
colonies formed on plates were picked up and further purified by
re-streaking on the corresponding medium and by incubation in
the same conditions as described above.
Culture of mixed bacteria in liquid medium (Fig. 1, “path 2”)
For each immersion time, 5 ml of liquid medium (Marine Broth
and Baar modified medium) were inoculated with 2% of the
corrosion product sample and incubated for 7 days at 30
C under
aerobic and anaerobic atmospheres. The cultures were then
transferred to 100 ml of fresh medium and incubated under the
same conditions for 3 additional weeks.
Determination of total bacterial abundance (Fig. 1, “path 3”)
The cells from the corrosion product layers were fixed in
formaldehyde (final concentration 1%). The mixture was then
 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
Fig. 1. Methods to evaluate the bacterial communities and to characterize the mineral phases present in the corrosion product layer. Three paths were followed to study the
bacterial communities.
diluted in 12 mM di-sodium pyrophosphate solution to the
appropriate bacterial concentration for counting and incubated for
20 min at 4
C. The cells were stained with 0.5 mg l1 4-6-
diamidino-2-phenylindole (DAPI) in the dark for 15 min and the
samples were then filtered onto 0.2 mm membrane filters (What-
man Scheicher and Scull, Nuclepore Track-Etch Membrane, 25 mm
diameter). The membranes were mounted on a glass slide in a non-
fluorescent oil drop (Leica type F) for microscopic analysis. Bacterial
cells were counted with an epifluorescence microscope (Leica
DMRB, mercury steam light, magnification 1000, filter UV A Leica)
over an average of 10 fields. All experiments were performed in
triplicate and standard deviations were calculated. The results are
expressed in bacteria g1 of dry sample.
Genomic DNA extraction
The genomic DNA from each bacterial strain isolated on solid
medium (Fig. 1, “path 1”) was extracted with the “Genomic DNA
from Tissue” kit (Macherey Nagel) from 1 to 5 ml of culture under
the corresponding conditions. The “PowerMax Soil” kit (MoBio)
was used to extract the DNA of each mixture of bacteria grown in
100 ml liquid medium (Fig. 1, “path 2”) from the pellet obtained
after a 5 min centrifugation of the culture at 8000 g. For each im-
mersion time, the DNA of the total bacterial community was
extracted directly from the scraped corrosion product layer (0.5 g)
(Fig. 1, “path 3”) with the “Power Soil” kit (MoBio).
16S rRNA gene amplification
Amplification of about 1400 bp of the 16S rRNA gene of bacterial
isolates was carried out using 50 ng of genomic DNA in a total
volume of 50 ml. The reaction mixtures contained 0.2 mM 16SUnivF
(50 AGAGTTTGATCCTGGCTCA30 ) and 16SUnivR (50 GGCTACCTT-
GTTACGACTT30 ) primers, 3 mM MgCl2, 320 mM of each dNTP, and
0.04 Taq polymerase Units (Fermentas), in the corresponding 1
57
buffer. Denaturation at 95
C for 2 min was followed by 30 cycles of
amplification (92
C for 30 s, 54
C for 30 s, 72
C for 1 min 30).
About 300 ng of each amplified DNA were sent to GenoScreen (Lille,
France) for sequencing.
A polymorphic portion of the 16S rRNA gene of about 620 bp,
including the V5, V6, V7 and V8 regions, was amplified from about
4  50 ng (4 tubes with 50 ng for each sample) of the DNAs
extracted from the mixed bacteria grown in liquid medium and
from the corrosion product layer (total bacterial community), in a
total volume of 4  50 ml (4 tubes of 50 ml for each sample). The
reaction mixtures contained 0.2 mM of the primers 774F
(50 GGAGCAAACAGGATTAGATA30 ) and 1392Rgc containing a GC
clamp (50 CGCCCGGGGCGCGCCCCGGGCGGGGCGGGGGCACGGGGG
GTGACGGGCGGTGTGTA30 ), 3 mM MgCl2, 320 mM of each dNTP, and
0.04 Taq polymerase units (Fermentas), in the corresponding 1
buffer. Denaturation at 94
C for 2 min 30 was followed by 30
amplification cycles (92
C for 30 s, 56
C for 30 s, 72
C for 30 s).
About 500 ng of each of these amplified samples were analyzed by
TTGE.
TTGE fingerprinting
The Temporal Temperature Gradient Electrophoresis (TTGE) gels
(16 cm  16 cm, 1 mm thickness) were prepared with 30 ml of
polyacrylamide (8% acrylamide-bisacrylamide (37.5:1) solution,
7 M urea, 1.25 TAE buffer), 188 ml of 10% ammonium persulfate
and 19 ml of N,N,N0 ,N0 -tetramethylethylenediamine (TEMED).
Electrophoresis was performed at 80 V with a temperature gradient
ranging from 65.8
C to 69.7
C (0.2
C/h ramp) using the DCode
Universal Mutation Detection System (Bio-Rad). After electropho-
resis, the gels were stained in Gel-Star (Lonza), photographed un-
der UV light using Gel Doc (Bio-Rad) and analyzed with Image Lab
software (version 4.1, 2012, Bio-Rad). The bands were excised from
the gel, the DNA was eluted in 30 ml H2O and used to perform PCRs
with the primers 774F and 1392R (without the GC clamp), as
58 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
described previously. Amplicons were sent to GenoScreen (Lille,
France) for sequencing. DNAs not sufficiently amplified to be
directly sequenced were cloned: for a 20 ml final volume, 8 ml
amplicon were mixed with 50 ng pGEMT-easy (Promega) and 3
units of T4 ligase in the corresponding buffer (Promega). After 15 h
at 16
C, 3 ml of this mixture was used to transform competent
Escherichia coli TG1 (as described by Sambrook et al., 1989). Bacteria
were grown on solid LB medium containing ampicillin 100 mg ml1,
X-Gal 40 mg ml1 and IPTG 40 mg ml1. PCRs using pGEMT-easy
primers which frame the inserted fragment were performed
directly on the white colonies, to select the bacteria containing a
fragment whose length was about 620 bp. Plasmid DNAs of positive
colonies were extracted with the NucleoSpin Plasmid Kit
(Macherey Nagel) and sent to GenoScreen (Lille, France) for
sequencing.
Sequence and phylogenetic analyses
In an attempt to identify the bacterial isolates and the bacteria
detected in the TTGE gels, the sequences were compared with those
in GenBank using the Blast software (http://blast.ncbi.nlm.nih.gov/
Blast.cgi). A 97% sequence similarity threshold was used to assign
16S rRNA gene sequences to operational taxonomic units (OTUs)
(Stackebrandt and Goebel, 1994). Multiple alignments were per-
formed with the sequences of the cultured bacteria and related
GenBank sequences using Multalin (http://multalin.toulouse.inra.
fr/multalin/multalin.html) (Corpet, 1988). Based on these results,
phylogenetic and molecular evolutionary analyses were conducted
using MEGA version 4 (Tamura et al., 2007) and a neighbor joining
phylogenetic tree was built.
Nucleotide sequence accession numbers
The 16S rRNA gene sequences determined in this study were
submitted to the GenBank database. Accession numbers: KJ814510
to KJ814616.
m-Raman spectroscopy analysis
Micro-Raman analysis was performed on a Jobin-Yvon High
Resolution Raman spectrometer (LabRAM-HR) equipped with a
microscope (Olympus BX 41) and a Peltier-based cooled Charge
Coupled Device (CCD). The analyzed zone had a diameter of ~6 mm.
Spectra were recorded with the LabSpec acquisition software at
room temperature with a resolution of ~0.1 cm1. Excitation was
provided by a HeeNe laser (632.8 nm). The power was varied be-
tween 1.94 and 0.07 mW in order to prevent excessive heating that
could induce a transformation of the analyzed sample into hema-
tite a-Fe2O3. The acquisition time was variable and was chosen for
each analysis so as to optimize the signal to noise ratio.
Results
Change over time of the bacterial communities
Total and cultured bacterial counts
Fig. 2 shows the distribution of the bacterial populations on the
carbon steel coupons according to immersion time. In corrosion
product layers, the average density of DAPI-stained cells (total
bacteria) ranged from 1.8  108 to 7.2  108 bacteria g1. Plate count
analysis revealed variations in the number of bacteria that could be
grown from the carbon steel coupons in different conditions of
oxygenation and different media. In aerobic conditions, the number
of bacteria grown on the rich MA medium generally increased with
time (from 1.2  105 to 2.6  106 CFU g1 in 1 month), as would be
Fig. 2. Average number of total and culturable bacteria in the corrosion product layer
(per gram of dry weight) determined by epifluorescence microscopy and plate counts,
according to immersion time, the conditions of oxygenation and the culture medium.
MA: marine agar, Baar: Baar modified solid medium, CFU: colony forming units.
expected during the growth phase of the bacterial biofilm. On the
other hand, in anaerobic conditions, the cell concentration
decreased slightly with time in all culture media (from 1.8  106 to
1.3  105 CFU g1 on MA and from 4.1  105 to 1.2  105 CFU g1 on
Baar modified medium). In any case, the percentage of cultured
bacteria, given by the cell density ratio of cultured bacteria/total
bacteria, was low (from 0.01% to 1.76%).
Diversity of isolated bacteria
A total of 79 morphologically different colonies were isolated
from the solid culture media (Marine Agar and Baar modified me-
dium). The isolates were identified by sequencing the PCR ampli-
fied 16S rRNA gene and comparing them to GenBank database
sequences (Fig. 1, “path 1”). All sequences matched with at least one
identified strain (with 97% similarity). From these data, a phylo-
genetic analysis was performed (Fig. 3). 59.5% of the isolates were
Gram negative and shared an affiliation with members of Alphap-
roteobacteria (30.4%), Gammaproteobacteria (13.9%) and Flavobac-
teria (15.2%) (Fig. 3). The Gram positive isolated strains were related
to Bacilli (38%) and Actinobacteria (2.5%) phyla. Overall, the isolated
strains were distributed in 19 different genera, 8 families, 7 orders
and 5 major bacterial taxa (phylum or class).
The results presented in Fig. 4 show that the proportion of each
taxonomic group varied with immersion time. Bacilli (mainly the
Bacillus genus) was the most commonly represented phylum at 1
week and 2 months of immersion, whereas Alphaproteobacteria
(mainly the Erythrobacter and Roseobacter genera) were dominant
among the 2 weeks and 1 month isolates. No Deltaproteobacteria,
Epsilonproteobacteria or Clostridia were isolated from the carbon
steel coupons.
Diversity of mixed bacterial cultures and total bacterial community
In order to detect bacteria that would be able to grow in con-
sortium only, and could not then been detected after isolation on
solid medium, mixed bacteria from each sample were cultured in
liquid medium (Fig. 1, “path 2”). The samples were concomitantly
analyzed directly without cultivation in order to study the diversity
of the total bacterial community (Fig. 1, “path 3”). A variable region
of the 16S rRNA gene was then amplified by PCR using the total DNA
extracted from the mixed bacteria grown in liquid medium and
from the corrosion product layer (total bacterial community). The
 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65 59

Fig. 3. Neighbor joining phylogenetic tree based on 16S rDNA sequences of the bacteria isolated by culture on solid medium and related GenBank species. Bootstrap values (1000
replicates) above 70% are noted at the nodes. The scale bar represents 0.1 changes. GenBank accession numbers are noted after species names.
60 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
Fig. 4. Proportion of each phylum or class of bacteria isolated by culture on solid
medium, according to the immersion time.
bacterial diversity of the different samples was evaluated by TTGE
analysis of the amplified region of the 16S rRNA gene. Fig. 5 shows
the TTGE fingerprints of the mixed bacterial cultures and the total
bacterial community at each immersion time. The number of bands
obtained in a TTGE pattern is related to the number of bacterial
species in the tested sample. Bands at the same position in different
samples were considered as one operational taxonomic unit (OTU).
From 1 to 28 bands were observed in the samples, reflecting a
variable bacterial diversity according to the immersion time, the
sample type (cultured or not) and the culture conditions. The
Fig. 5. TTGE fingerprints of 16S rRNA gene fragments PCR-amplified from DNA extracts
of mixed bacterial cultures and total bacterial community. Ae: aerobic; An: anaerobic;
M: marine broth; Baar: Baar modified broth. Bands were numbered according to their
migration front. Black numbers: DNA sequenced. *: intense bands present in the
cultured samples and not detected in non-cultured samples.
highest diversity was seen in the corrosion product layer of 1
month (non-cultured) while the lowest was found in 1 week
samples, grown in liquid medium in anaerobic conditions. The
analysis of the TTGE fingerprints showed that the main bacteria in
the total community generally differed from the main bacteria
grown in liquid medium. A number of intense bands present in the
samples of cultured bacteria were not detected in the samples from
non-cultured bacteria (bands with a star in Fig. 5). Conversely, the
most intense band that was present in all samples from non-
cultured bacteria (band 9 in Fig. 5) was not observed in the
cultured samples.
The main bands were then excised from the gel and sequenced.
In order to identify the OTUs, the sequences were submitted to a
Blast search to retrieve the corresponding phylogenetic relatives
(Table 1). Eight OTUs were considered as unknown species because
the similarity with the closest related identified genus was less
than 97%. However, all of them were assigned to a microbial group
according to a phylogenetic tree based on the 16S rDNA sequences
of the TTGE OTUs and the GenBank related species (data not
shown): OTUs 2 and 3 were considered as Epsilonproteobacteria,
OTU 11 as Deltaproteobacteria, OTU 14 as Gammaproteobacteria, and
OTUs 13, 21, 25 and 28 as Clostridia (Table 1). After 1 week of im-
mersion, the Sulfurimonas autotrophica species (OTUs 6, 8 and 9)
was observed in each non-cultured sample (corresponding to total
bacterial community) whereas this species was detected neither in
mixed bacteria grown in liquid medium nor in bacteria isolated on
solid medium. S. autotrophica belongs to the Epsilonproteobacteria
class, as does Sulfurospirillum arcachonense (OTU 1), which was also
identified in the total bacterial community at 2 months of immer-
sion but was not detected in the cultured bacteria. In total, 6 OTUs
affiliated to Epsilonproteobacteria were detected and all of them
were observed in non-cultured samples. Similarly, Phaeobacter
arcticus (OTU 12) and Sulfitobacter mediterraneus (OTU 16), which
are related to the Alphaproteobacteria class, were only detected in
the total bacterial community, at the different times of immersion.
Roseobacter (OTUs 15 and 17) was the only genus which was shared
by the total bacterial community and the cultured bacteria (mixed
cultures and isolated strains). This genus was rather common in the
total bacterial community during the first two weeks.
Conversely, in the mixed bacterial cultures, we found genera
that were not detected in non-cultured samples: Pseudidiomarina
(OTUs 4 and 10), Tenacibaculum (OTU 5), Winogradskyella (OTU 7),
Photobacterium (OTU 20), Pseudomonas (OTU 22), Marinobacter
(OTU 23), Vibrio (OTUs 24 and 26), Virgibacillus dokdonensis (OTU
27) and Clostridium aminobutyricum (OTU 29).
Finally, only one species, Desulfovibrio profundus (OTU 19), was
identified as an SRB. It was detected in mixed bacteria cultured in
Baar medium and taken from the coupon left 1 month in seawater.
Change over time of the corrosion products
The morphology of the corrosion product layer was visually the
same at all immersion times (1 week, 2 weeks, 1 month and 2
months). The outer part of the layer was characterized by an
orange-brown color while the inner part, which was in contact
with the metal, was dark, almost black. Of course, the thickness of
the entire layer increased with time. The components of the dark
inner region and those of the orange-brown outer region were
analyzed separately by m-Raman spectroscopy.
Typical Raman spectra obtained with the coupons left 1 or 2
weeks in seawater are presented in Fig. 6. In each case, the orange-
brown outer layer appeared to be essentially made of lepidocrocite
g-FeOOH (Fig. 6a). The two main Raman peaks of this compound
are located at ~250 and ~380 cm1 (Pineau et al., 2008; Refait et al.,
2009). In contrast, the main component of the dark inner layer was
 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65 61

Table 1
Phylogenetic affiliation of 16S rRNA gene fragments obtained from mixed bacterial cultures and from total bacterial community studied by TTGE.

OTU Closest related genus in Genbank (reference; % similarity) Microbial group affiliation OTU detection
Immersion time

1 Week 2 Weeks 1 Month 2 Months

An An Ae NC Ae NC An An Ae NC An An Ae NC
Baar M M M Baar M M Baar M M

1 Sulfurospirillum arcachonense (NR_026408; 98%) 3-Proteobacteria þ

2 Unknown species/Sulfurovum lithotrophicum (NR_024802; 95%) 3-Proteobacteria* þ þ þ
3 Unknown species/Sulfurimonas autotrophica (NR_074451; 96%) 3-Proteobacteria* þ þ
4 Pseudidiomarina sp. (EU600202; 99%) g-Proteobacteria þ
5 Tenacibaculum sp. (JN791367; 99%) Flavobacteria þ
6 Sulfurimonas autotrophica (NR_028643; 97%) 3-Proteobacteria þ þ þ
7 Winogradskyella (JX854069; 100%) Flavobacteria þ þ þ
8 Sulfurimonas autotrophica (AB088432; 97%) 3-Proteobacteria þ þ þ þ
9 Sulfurimonas autotrophica (AB088432; 98%) 3-Proteobacteria þ þ þ þ
10 Pseudidiomarina sp. (EU600202; 99%) g-Proteobacteria þ þ
11 Unknown species/sulfate-reducing bacterium (EU908726; 98%) d-Proteobacteria* þ þ
12 Phaeobacter arcticus (NR_043888; 99%) a-Proteobacteria þ þ þ
13 Unknown species/iron-reducing bacterium (FJ269100; 90%) Clostridia* þ
14 Unknown species/Steroidobacter denitrificans (NR_044309; 91%) g-Proteobacteria* þ
15 Roseobacter sp. (EU195946; 99%) a-Proteobacteria þ þ þ þ þ þ þ
16 Sulfitobacter mediterraneus (DQ915636; 99%) a-Proteobacteria þ þ þ þ
17 Roseobacter (GU584172; 100%) a-Proteobacteria þ þ
18 Sulfitobacter sp. (AB583772; 99%) a-Proteobacteria þ
19 Desulfovibrio profundus (FR733706; 98%) d-Proteobacteria þ
20 Photobacterium (GQ454940; 99%) g-Proteobacteria þ
21 Unknown species/Desulfotomaculum sp. (DQ386219; 95%) Clostridia* þ
22 Pseudomonas sp. (AM913883; 100%) g-Proteobacteria þ
23 Marinobacter sp. (AB526350; 100%) g-Proteobacteria þ þ þ þ þ
24 Vibrio sp. (GQ455012; 99%) g-Proteobacteria þ
25 Unknown species/Desulfitispora alkaliphila (FJ788525; 90%) Clostridia* þ
26 Vibrio sp. (FR744846; 99%) g-Proteobacteria þ þ þ þ þ
27 Virgibacillus dokdonensis (KF453778; 99%) Bacilli þ
28 Unknown species/iron-reducing bacterium (FJ269085; 93%) Clostridia* þ
29 Clostridium aminobutyricum (X76161; 97%) Clostridia þ

When the similarity with the closest related genus in Genbank was <97%, the species was considered as unknown. *: unknown species were assigned to a microbial group
according to a phylogenetic tree based on the 16S rDNA sequences of the TTGE OTUs and GenBank related species (data not shown). Ae: aerobic; An: anaerobic; M: marine
broth; Baar: Baar modified broth; NC: not cultured samples; þ: presence of the OTU. GenBank accession numbers of OTUs 1 to 29: KJ814588 to KJ814616.

a GR compound, characterized by three Raman bands at 220, 430 Discussion

and 505 cm1 (Fig. 6b). This spectrum is typical of GR(SO2 4 ) (de
Faria et al., 1997; Pineau et al., 2008; Refait et al., 2011), a major
component of the corrosion product layers that form on steel in
seawater (Pineau et al., 2008; Refait et al., 2011). GR(SO2 4 ) and
lepidocrocite were also often observed together, indicating that the
frontier between the inner GR(SO2 4 ) layer and the outer g-FeOOH
layer was not clearly determined.
Typical Raman spectra obtained with the coupons left 1 or 2
months in seawater are presented in Fig. 7. The two strata that
constituted the corrosion product layers were thicker and thus
more clearly defined. The outer orange-brown stratum was mainly
composed of Fe(III)-oxyhydroxides, with lepidocrocite still pre-
dominant, as illustrated by Fig. 7a. However, Fig. 7b shows that
lepidocrocite was accompanied at this stage by goethite a-FeOOH,
characterized by two main Raman peaks at ~300 and ~390 cm1 (de
Faria et al., 1997; Pineau et al., 2008) and magnetite Fe3O4, essen-
tially characterized by a broad band at ~680 cm1 (de Faria et al.,
1997). Similarly, the dark inner stratum did not change much and
proved to be almost entirely composed of GR(SO2 4 ). Numerous
spectra, similar to those in Fig. 7c, showed only the three main
Raman bands of the GR.
However, another compound could be identified very locally on
the coupon left 1 month in seawater, as illustrated by Fig. 7d. Its
spectrum consisted of only two sharp peaks located at 208 and
282 cm1. This is the signature of nanocrystalline mackinawite
(Bourdoiseau et al., 2008), the FeS compound that precipitates from
dissolved Fe(II) and S(-II) species (Ohfuji and Rickard, 2006; Jeong
et al., 2008).
Change in the bacterial communities inside the corrosion product
layer
It has been frequently reported in studies of oligotrophic to
mesotrophic aquatic habitats that direct microscopic counts of
micro-organisms exceed counts of cells grown in culture by several
orders of magnitude (Amann et al., 1995; Lebaron et al., 2001). We
also observed this phenomenon with all the corrosion product
layers formed on carbon steel coupons analyzed in this work.
Moreover, the number of culturable bacteria associated with the
studied corrosion product layers generally increased with immer-
sion time in aerobic conditions but decreased in anaerobic condi-
tions. These tendencies need to be confirmed by studying coupons
immersed for a longer time.
To evaluate bacterial diversity in these corrosion product layers,
16S rRNA gene sequences were analyzed from heterotrophic bac-
teria grown in aerobic and anaerobic conditions and from total
bacteria, after TTGE separation. Both approaches revealed a mod-
erate bacterial diversity in the biofilm samples. Similar results were
previously obtained by Bermont-Bouis et al. (2007) and Boudaud
et al. (2010) on steel coupons immersed in seawater, and by
Neria-Gonzalez et al. (2006) and Jan-Roblero et al. (2004) in oil and
gas pipeline biofilms. This limited bacterial diversity might be
related to the strong selective pressure of this environment,
particularly in relation to the immediate contact with high saline
and metallic ions concentrations, a modified pH and a low con-
centration of oxygen.
62 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
Fig. 6. m-Raman analysis of corrosion product layers formed on steel coupons after 1 or
2 weeks of permanent immersion in seawater. (a) outer stratum and (b) inner stratum.
L: lepidocrocite; GR: GR(SO2
4 ).
In this work, Roseobacter, Erythrobacter and Bacillus were the
dominant genera identified among the isolated strains. They are
also commonly found in the marine environment. In fact, all the
taxonomic groups identified have already been described in the
literature as micro-organisms that are common in marine envi-
ronments (seawater, sediments). Gonz
alez and Moran (1997) pre-
sented Alphaproteobacteria as a dominant and easy-to-culture
group from coastal seawater. Among this group, the genus Rose-
obacter may constitute up to 25% of marine bacterioplankton
(Gonza
lez and Moran, 1997; Suzuki et al., 2001; Pernthaler and
Amann, 2005). Similarly, Bacillus is the dominant culturable che-
moorganotrophic bacteria from estuarine and marine environ-
ments (Frette et al., 2004). Gammaproteobacteria were also detected
in various marine biotopes, in the water column (Cho and
Giovannoni, 2004) and in sediments (Gray and Herwig, 1996;
Ravenschlag et al., 2001), and they were associated with the
dominant microflora on carbon steel surfaces immersed in
seawater (Bermont-Bouis et al., 2007). In the marine environment,
alpha and gamma groups of Proteobacteria may account for 75% of
cultured bacteria (Hugenholtz et al., 1998). The Flavobacteria
phylum is also common in marine sediments (Ravenschlag et al.,
2001) or in coastal water (Yokokawa and Nagata, 2005; Lo
Giudice et al., 2012) and is numerically dominant among the cul-
turable estuarine bacteria in the Baltic Sea (Kisand et al., 2002).
Conversely, neither Deltaproteobacteria, Epsilonproteobacteria nor
Clostridia were isolated from culture on solid medium.
It is well documented that the Deltaproteobacteria class includes
most sulfidogenic bacteria, in particular SRB. Only one Deltapro-
teobacterium, D. profundus, was identified. It was detected among
the mixed bacteria, grown in liquid medium, that were taken from
the coupon immersed for 1 month. It has been shown that
D. profundus can reduce sulfate, sulfite or thiosulfate to produce
sulfide at a pH between 4.5 and 9, at elevated pressures and in strict
anaerobic conditions (Bale et al., 1997). Its detection in the mixed
culture bacterial sample taken from the 1 month immersion
coupon was consistent with the concomitant presence of iron
sulfide in various places at this immersion time. Indeed, in
seawater, sulfide species are more likely to result from the meta-
bolic activity of sulfide-producing bacteria. Our results also show
the difficulty of isolating SRB from carbon steel during the first two
months of immersion, probably because of their low abundance in
the corrosion product layer and the use of culture conditions that
were not suitable (on solid medium). Their detection only in mixed
cultures suggests that a bacterial consortium, present in the liquid
culture medium, may favor SRB growth. In fact, Vibrio was the other
bacterial genus detected in this consortium. Similar observations
were previously made by other authors: Desulfovibrio in association
with members of the Vibrio genus was detected on carbon steel
immersed for 8 months (Bermont-Bouis et al., 2007) and 12 months
(Boudaud et al., 2010) in various French harbors.
Moreover, no SRB were observed in mixed cultures from the
coupon immersed for 2 months. This is in accordance with
Bermont-Bouis et al. (2007), who described a corroding marine
biofilm formed on carbon steel immersed for 1 and 8 months, and
who showed that Desulfovibrio species dominated only after 8
months of immersion. The detection of SRB in the coupon
immersed for 1 month only could be explained by the appearance
of local environmental conditions allowing SRB to colonize a spe-
cific area of this coupon. This specific area may be the first where
anaerobic conditions appear within the corrosion product layer,
which we know to be very heterogenous.
We also found sulfidogenic bacteria related to the
S. arcachonense species in the total bacterial community. These
microaerophilic bacteria of the Epsilonproteobacteria class can
reduce elemental sulfur to sulfides for organotrophic growth
(Finster et al., 1997). They were identified on the coupon immersed
for 2 months, where the layer of corrosion products was thickest.
Another member of Epsilonproteobacteria, S. autotrophica, was
detected among the total bacterial community at all immersion
times. This species is interesting because of its significant contri-
bution to the sulfur cycle as it oxidizes sulfur compounds
(elemental sulfur, sulfide and thiosulfate) to sulfate in aerobic
conditions (Inagaki et al., 2003; Sikorski et al., 2010). The
expanded group of Epsilonproteobacteria, which was only found in
our non-cultured samples, has been recognized as ubiquitous in
marine and terrestrial ecosystems, and would have a significant
role in sulfidic habitats (Campbell et al., 2006). The
S. mediterraneus species, which was detected in the total bacterial
community at all immersion times, is also associated with the
organic sulfur cycle in the ocean because this member of
Alphaproteobacteria is able to oxidize sulfite to sulfate (Pukall et al.,
1999; Ivanova et al., 2004).
Marinobacter, identified in the mixed bacteria culture at 1 week
and 2 months of immersion, has been described as capable of
oxidizing Fe(II) to Fe(III) in neutral media (pH 6.5e7.5) (Edwards
et al., 2004) and these bacteria could be indirectly involved in the
sulfide oxidation of marine sediments via Fe3þ production (Muller
et al., 2010). Moreover, the presence of suspected iron-reducing
bacteria of the Clostridia phylum (OTUs 13 and 28 Table 1), sulfur-
oxidizing bacteria (OTUs 2 and 3 Table 1) and sulfidogenic bacte-
ria (OTUs 11, 21 and 25 Table 1), suggests that various micro-
 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
Fig. 7. m-Raman analysis of corrosion product layers formed on steel coupons after 1 or 2 months of permanent immersion in seawater. (a, b) outer stratum and (c, d) inner stratum.
L: lepidocrocite; G: goethite; M: magnetite; GR: GR(SO24 ); Mk: nanocrystalline mackinawite.
organisms associated with the redox cycles of Fe and S have
developed in the corrosion product layers.
A significant discrepancy between the microbial communities
determined by culture-dependent and independent analyses was
observed in this study. Only one genus, Roseobacter, was detected
by both procedures. These bacteria may also play an important role
in the total sulfur cycle in many coastal and benthic marine envi-
ronments (sulfide-rich habitats) because the ability to oxidize
inorganic forms of sulfur, including elemental sulfur, sulfide, sulfite
and thiosulfate, has been demonstrated for several Roseobacter
strains (Buchan et al., 2005). The significant discrepancy in the
results obtained with these procedures is not surprising since many
elements of the bacterial communities, especially anaerobic bac-
teria, may be extremely difficult to grow in culture. This difficulty
may be due to the low capacity of certain bacteria to grow in culture
and to difficulties in determining suitable culture conditions. As
some bacteria could only grow in consortium and could then not be
detected after isolation by cultivation on solid medium, mixed
bacteria were cultivated in liquid medium. It is clear that the bac-
teria observed after cultivation in these conditions were not an
accurate picture of the initial bacterial community present on the
coupon, because, for example, dominant organisms in the liquid
culture could be strains initially present in small amounts. How-
ever, this experiment allowed us, on the coupon immersed for 1
month, to detect one SRB that we were not able to detect after
cultivation on solid medium. Interactions between different bac-
teria could occur in liquid medium, and the growth of some bac-
teria could be favored by the presence of others. It is therefore
important to continue to improve culture techniques in order to
reduce the discrepancy between the results obtained by cultivation
and molecular techniques. On the other hand, our results indicate
that both methods may offer different perspectives on the
63
biodiversity study of bacteria in biofilms: the culture methods were
able to detect species that were not recovered using 16S rDNA TTGE
and vice-versa. Clearly, the culture conditions used in this study led
us to select heterotrophic aerobes and facultative anaerobes.
Additionally, culture conditions tend to select for fast growers, such
as Bacillus or Roseobacter species rather than SRB (Desulfovibrio)
species.
Evolution with time of the corrosion products and its relationship
with bacterial communities
During the early stages of marine corrosion of steel, the kinetics
of the corrosion process is controlled by the reduction of dissolved
O2 (Melchers and Jeffrey, 2005). The accumulation at the steel/
seawater interface of Fe2þaq and OH ions, produced by the disso-
lution of steel and the reduction of O2, respectively, rapidly leads to
the formation of a solid phase. The only Fe(II)-based compound
observed is GR(SO2 II III
4 ), that is Fe 4Fe 2(OH)12SO4 $ 8H2O, a mixed
valence Fe(IIeIII) compound. It forms the inner stratum of the
corrosion product layer which confirms that it is the first solid that
precipitates. The overall reaction leading to the formation of this
inner stratum is written as:
þ
12Fe þ 2SO2
4 þ 7O2 þ 26H2O þ 4H /
2(FeII4FeIII2(OH)12SO4$8H2O) (1)
GR(SO2
4 ) can be oxidized by dissolved O2 into various Fe(III)-
based compounds, namely lepidocrocite g-FeOOH, goethite a-
FeOOH and magnetite Fe3O4. Lepidocrocite is, however, the main
product of GR(SO2
4 ) oxidation in seawater-like media (Refait et al.,
2003) and is favored by large oxygen flows (Detournay et al., 1974)
and moderate temperatures (T  20
C) (Detournay et al., 1976;
64 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
Olowe et al., 1991). Our analyses confirmed that lepidocrocite was
the main oxidation product in the early stages of the process. Thus,
after 1 or 2 weeks, the corrosion product layer consisted in an outer
layer of lepidocrocite covering a discontinuous inner layer of
GR(SO24 ). At that time the corrosion product layer is thin, loose and
porous and does not constitute a physical barrier. The access of
dissolved O2 to the steel surface is not hindered significantly. This
explains why lepidocrocite can be observed not only in the outer
part of the corrosion product layer, but also in contact with the
metal surface. The reaction leading to the outer stratum of the
corrosion product layer can be written as follows:
4(FeII4FeIII2(OH)12SO4$8H2O) þ 3O2 / 20(g-FeOOH) þ 4Fe2þ
þ 4SO2
4 þ 46H2O
After 1e2 months of immersion, goethite and magnetite were
also detected in the outer stratum of the corrosion product layer.
Goethite can be obtained by oxidation of GR(SO2 4 ) with moderate
oxygen flows (Gilbert et al., 2008) while the formation of magnetite
requires very low oxygen flows (Ruby et al., 2010). The identifica-
tion of these two compounds inside the corrosion product layer
indicates that the access of O2 to the inner stratum becomes more
and more difficult as the thickness of the corrosion product layer
increases. The regions where magnetite was locally observed point
out O2-depleted zones where the growth of SRB and other anaerobe
micro-organisms should be favored.
In agreement with this observation, FeS could be identified
locally as soon as 1 month of immersion. The microbiological
analysis confirmed, as it was expected, that the formation of FeS
was due to the metabolic activity of sulfidogenic bacteria, such as
D. profundus and S. arcachonense identified after 1 or 2 months of
immersion (Table 1). At those early stages, SRB are growing in the
first O2-depleted zones of the corrosion product layer, and their
influence on the corrosion process is only local. This induces
localized corrosion phenomena, as observed in many studies (e.g.
Cragnolino and Tuovinen, 1984; Hamilton, 1994; Jeffrey and
Melchers, 2007; Castaneda and Benetton, 2008).
The corrosion product layer is also a suitable environment for
bacteria associated with the redox cycle of Fe and various iron
oxidizing and iron reducing bacteria were identified, especially
after 1 or 2 months of immersion. IRB, such as the Clostridia iden-
tified after 1 and 2 months of immersion (Table 1), may preferen-
tially develop at the GR(SO2 4 )/FeOOH interface, where anoxic
conditions are met and Fe(III) compounds are present. Iron(III)
oxyhydroxides such as g-FeOOH can be reduced by IRB and this
reduction can lead to GR(SO2 4 ) (Zegeye et al., 2005). However, if
sulfide species are produced nearby by SRB, the reduction of the
FeOOH phases should rather lead to Fe(II)-based sulfides.
It must finally be noted that experiments performed in abiotic
conditions in deaerated seawater-like electrolytes demonstrated
that GR(SO24 ) was also the first solid to form on the steel surface
during the anaerobic period (Refait et al., 2011). Thus GR(SO2
4 ) is in
any case the Fe(II)-based compound resulting from the purely
electrochemical (i.e. abiotic) corrosion process, while FeS is that
connected to the MIC process that involves SRB. It can then be
proposed that the FeS/GR(SO2 4 ) ratio quantifies the influence of
SRB and/or other sulfidogenic bacteria.
Conclusions
During the early stages of marine corrosion of steel (1e8 weeks),
the corrosion products forming are mainly the solid phases that
result from the purely electrochemical (abiotic) corrosion process,
i.e. GR(SO2
4 ) and its oxidation products a-FeOOH, g-FeOOH and
magnetite Fe3O4. The kinetics of the corrosion process may not be
influenced or only slightly by micro-organisms during these early
stages, but the observed evolution of the bacterial community
associated with the corrosion product layer foreshadows the MIC
process that will prevail later on. The identification of FeS in a few
rare areas of the inner stratum of the corrosion product layer
indicated that, as early as 1 or 2 months after immersion, the local
development of sulfide-producing bacteria was sufficient to influ-
ence the nature of the corrosion products. The development of
these micro-organisms was ascertained by the thorough microbi-
ological analysis of the bacterial community associated with the
corrosion products. Besides, SOB were identified for each immer-
sion time. After 1e2 months of immersion various bacteria asso-
ciated with the redox cycle of Fe, i.e. IOB and IRB, were also
(2) identified. The development of bacterial consortia including SRB,
SOB, IOB and IRB, i.e. various species that may play a role in the
increase of the corrosion rate, is then already possible for such short
immersion times.
Finally, it would be interesting now to specify the role of the
bacteria detected by culture and by TTGE in the current work.
Further studies in which the isolated bacteria are used separately
and in consortia, in contact with metallic coupons, should provide
information about the influence of these micro-organisms on the
corrosion process.
Acknowledgments
This work, part of the PhD thesis of M. Langumier, was finan-
cially supported by the Conseil Ge
ne

ral de la Charente Maritime
(17). We thank Vale
rie Sopena for technical assistance.
References
Amann, R.I., Ludwig, W., Schleifer, K.H., 1995. Phylogenetic identification and in situ
detection of individual microbial cells without cultivation. Microbiol. Rev. 59,
143e169.
Atlas, R.M., 2005. Handbook of Media for Environmental Microbiology, second ed.
Taylor and Francis Group, CRC Press, Boca Raton, USA.
Bale, S.J., Goodman, K., Rochelle, P.A., Marchesi, J.R., Fry, J.C., Weightman, A.J.,
Parkes, R.J., 1997. Desulfovibrio profundus sp. nov., a novel barophilic sulfate-
reducing bacterium from deep sediment layers in the Japan sea. Int. J. Syst.
Bacteriol. 47, 515e521.
Beech, I.B., Campbell, S.A., 2008. Accelerated low water corrosion of carbon steel in
the presence of a biofilm harbouring sulphate-reducing and sulphur-oxidising
bacteria recovered from a marine sediment. Electrochim. Acta 54, 14e21.
Bermont-Bouis, D., Janvier, M., Grimont, P.A.D., Dupont, I., Vallaeys, T., 2007. Both
sulfate-reducing bacteria and Enterobacteriaceae take part in marine bio-
corrosion of carbon steel. J. Appl. Microbiol. 102, 161e168.
Boudaud, N., Coton, M., Coton, E., Pineau, S., Travert, J., Amiel, C., 2010. Biodiversity
analysis by polyphasic study of marine bacteria associated with biocorrosion
phenomena. J. Appl. Microbiol. 109, 166e179.
Bourdoiseau, J.A., Jeannin, M., Sabot, R., Re
mazeilles, C., Refait, Ph, 2008. Charac-
terisation of mackinawite by Raman spectroscopy: effects of crystallisation,
drying and oxidation. Corros. Sci. 50, 3247e3255.
Bryant, R.D., Laishley, E.J., 1990. The role of hydrogenase in anaerobic biocorrosion.
Can. J. Microbiol. 36, 259e264.
Buchan, A., Gonzalez, J.M., Moran, M.A., 2005. Overview of the marine Roseobacter
lineage. Appl. Environ. Microbiol. 71, 5665e5677.
Campbell, B.J., Engel, A.S., Porter, M.L., Takai, K., 2006. The versatile epsilon-
proteobacteria: key players in sulphidic habitats. Nat. Rev. Microbiol. 4,
458e468.
Castaneda, H., Benetton, X.D., 2008. SRB-biofilm influence in active corrosion sites
formed at the steel-electrolyte interface when exposed to artificial seawater
conditions. Corros. Sci. 50, 1169e1183.
Cho, J.C., Giovannoni, S.J., 2004. Cultivation and growth characteristics of a diverse
group of oligotrophic marine Gammaproteobacteria. Appl. Environ. Microbiol.
70, 432e440.
Corpet, F., 1988. Multiple sequence alignment with hierarchical clustering. Nucleic
Acids Res. 16, 10881e10890.
Cragnolino, G., Tuovinen, O.H., 1984. The role of sulphate-reducing and sulphur-
oxidizing bacteria in the localized corrosion of iron-base alloys. A review. Int.
Biodeterior. Bull. 20, 9e26.
^ncio Silva, S., de Oliveira, M.T., 1997. Raman microspectroscopy
de Faria, D.L.A., Vena
study of some iron oxides and oxyhydroxides. J. Raman Spectrosc. 28, 873e878.
Detournay, J., Ghodsi, M., Derie, R., 1974. Etude cine
tique de la formation de goethite
par ae
ration de gels d'hydroxyde ferreux. Ind. Chim. Belge. 39, 695e701.
 I. Lanneluc et al. / International Biodeterioration & Biodegradation 99 (2015) 55e65
Detournay, J., De
rie, R., Ghodsie, M., 1976. Study of the oxidation by aeration of
Fe(OH)2 in chloride environment. Z. Anorg. Allg. Chem. 427, 265e273.
Duan, J., Wu, S., Zhang, X., Huang, G., Du, M., Hou, B., 2008. Corrosion of carbon steel
influenced by anaerobic biofilm in natural seawater. Electrochim. Acta 54,
22e28.
Edwards, K.J., Bach, W., McCollom, T.M., Rogers, D.R., 2004. Neutrophilic iron-
oxidizing bacteria in the ocean: their habitats, diversity, and roles in mineral
deposition, rock alteration, and biomass production in the deep-sea. Geo-
microbiol. J. 21, 393e404.
Enning, D., Venzlaff, H., Garrelfs, J., Dinh, H.T., Meyer, V., Mayrhofer, K., Hassel, A.W.,
Stratmann, M., Widdel, F., 2012. Marine sulfate-reducing bacteria cause serious
corrosion of iron under electroconductive biogenic mineral crust. Environ.
Microbiol. 14, 1772e1787.
Finster, K., Liesack, W., Tindall, B.J., 1997. Sulfurospirillum arcachonense sp. nov., a
new-microaerophilic sulfur-reducing bacterium. Int. J. Syst. Bacteriol. 47,
1212e1217.
Frette, L., Johnsen, K., Jørgensen, N.O.G., Nybroe, O., Kroer, N., 2004. Functional
characteristics of culturable bacterioplankton from marine and estuarine en-
vironments. Int. Microbiol. 7, 219e227.
Gilbert, F., Refait, Ph, Le
ve^que, F., Remazeilles, C., Conforto, E., 2008. Synthesis of
goethite from Fe(OH)2 precipitates: influence of Fe(II) concentration and stir-
ring speed. J. Phys. Chem. Solids 69, 2124e2130.
Gonza
lez, J.M., Moran, M.A., 1997. Numerical dominance of a group of marine
bacteria in the alpha-subclass of the class Proteobacteria in coastal seawater.
Appl. Environ. Microbiol. 63, 4237e4242.
Gray, J.P., Herwig, R.P., 1996. Phylogenetic analysis of the bacterial communities in
marine sediments. Appl. Environ. Microbiol. 62, 4049e4059.
Hamilton, W.A., 1994. Biocorrosion: the action of sulphate-reducing bacteria. In:
Ratledge, C. (Ed.), Biochemistry of Microbial Degradation. Kluwer Academic
Publishers, Dordrecht, pp. 555e570.
Herrera, L.K., Videla, H.A., 2009. Role of iron-reducing bacteria in corrosion and
protection of carbon steel. Int. Biodeterior. Biodegrad. 63, 891e895.
Hugenholtz, P., Goebel, B.M., Pace, N.R., 1998. Impact of culture-independent studies
on the emerging phylogenetic view of bacterial diversity. J. Bacteriol. 180,
4765e4774.
Inagaki, F., Takai, K., Kobayashi, H., Nealson, K.H., Horikoshi, K., 2003. Sulfurimonas
autotrophica gen. nov., sp. nov., a novel sulfur-oxidizing epsilon-proteobacte-
rium isolated from hydrothermal sediments in the Mid-Okinawa Trough. Int. J.
Syst. Evol. Microbiol. 53, 1801e1805.
Ivanova, E.P., Gorshkova, N.M., Sawabe, T., Zhukova, N.V., Hayashi, K.,
Kurilenko, V.V., Alexeeva, Y., Buljan, V., Nicolau, D.V., Mikhailov, V.V.,
Christen, R., 2004. Sulfitobacter delicatus sp. nov. and Sulfitobacter dubius sp.
nov., respectively from a starfish (Stellaster equestris) and sea grass (Zostera
marina). Int. J. Syst. Evol. Microbiol. 54, 475e480.
Jan-Roblero, J., Romero, J.M., Amaya, M., Le Borgne, S., 2004. Phylogenetic charac-
terization of a corrosive consortium isolated from a sour gas pipeline. Appl.
Microbiol. Biotechnol. 64, 862e867.
Jeffrey, R., Melchers, R.E., 2007. The changing topography of corroding mild steel
surfaces in seawater. Corros. Sci. 49, 2270e2288.
Jeong, H.Y., Lee, J.H., Hayes, K.F., 2008. Characterization of synthetic nanocrystalline
mackinawite: crystal structure, particle size, and specific surface area. Geochim.
Cosmochim. Acta 72, 493e505.
Kisand, V., Cuadros, R., Wikner, J., 2002. Phylogeny of culturable estuarine bacteria
catabolizing riverine organic matter in the northern Baltic sea. Appl. Environ.
Microbiol. 68, 379e388.
Lebaron, P., Servais, P., Troussellier, M., Courties, C., Muyzer, G., Bernard, L.,
Scha€fer, H., Pukall, R., Stackebrandt, E., Guindulain, T., Vives-Rego, J., 2001. Mi-
crobial community dynamics in Mediterranean nutrient-enriched seawater
mesocosms: changes in abundances, activity and composition. FEMS Microbiol.
Ecol. 34, 255e266.
Lo Giudice, A., Caruso, C., Mangano, S., Bruni, V., Domenico, M., Michaud, L., 2012.
Marine bacterioplankton diversity and community composition in an Antarctic
coastal environment. Microb. Ecol. 63, 210e223.
Malard, E., Kervadec, D., Gil, O., Lefevre, Y., Malard, S., 2008. Interactions between
steels and sulphide-producing bacteria e corrosion of carbon steels and low-
alloy steels in natural seawater. Electrochim. Acta 54, 8e13.
Mehanna, M., Basseguy, R., Delia, M.-L., Girbal, L., Demuez, M., Bergel, A., 2008. New
hypotheses for hydrogenase implication in the corrosion of mild steel. Elec-
trochim. Acta 54, 140e147.
Mehanna, M., Basseguy, R., Delia, M.-L., Bergel, A., 2009. Effect of Geobacter sul-
furreducens on the microbial corrosion of mild steel, ferritic and austenitic
stainless steels. Corros. Sci. 51, 2596e2604.
Melchers, R.E., Jeffrey, R., 2005. Early corrosion of mild steel in seawater. Corros. Sci.
47, 1678e1693.
View publication stats
65
Melchers, R.E., Jeffrey, R., 2012. Corrosion of long vertical steel strips in the marine
tidal zone and implications for ALWC. Corros. Sci. 65, 26e36.
Melchers, R.E., Wells, T., 2006. Models for the anaerobic phases of marine immer-
sion corrosion. Corros. Sci. 48, 1791e1811.
Memet, J.B., 2000. (PhD thesis). University of La Rochelle, France.
Muller, M., Handley, K.M., Lloyd, J., Pancost, R.D., Mills, R.A., 2010. Biogeochemical
controls on microbial diversity in seafloor sulphidic sediments. Geobiology 8,
309e326.
Neria-Gonzalez, I., Wang, E.T., Ramirez, F., Romero, J.M., Hernandez-Rodriguez, C.,
2006. Characterization of bacterial community associated to biofilms of
corroded oil pipelines from the southeast of Mexico. Anaerobe 12, 122e133.
Ohfuji, H., Rickard, D., 2006. High resolution transmission electron microscopic
study of synthetic nanocrystalline mackinawite. Earth Planet. Sci. Lett. 241,
227e233.
Olowe, A.A., Pauron, B., Ge
nin, J.-M.R., 1991. The influence of temperature on the
oxidation of ferrous hydroxide in sulphated aqueous medium: activation en-
ergies of formation of the products and hyperfine structure of magnetite.
Corros. Sci. 32, 985e1001.
Pernthaler, J., Amann, R., 2005. Fate of heterotrophic microbes in pelagic habitats:
focus on populations. Microbiol. Mol. Biol. Rev. 69, 440e461.
Pineau, S., Sabot, R., Quillet, L., Jeannin, M., Caplat, Ch, Dupont-Morral, I., Refait, Ph,
2008. Formation of the Fe(IIeIII) hydroxysulphate green rust during marine
corrosion of steel associated to molecular detection of dissimilatory sulphite-
reductase. Corros. Sci. 50, 1099e1111.
Pukall, R., Buntefuss, D., Frühling, A., Rohde, M., Kroppenstedt, R.M., Burghardt, J.,
Lebaron, P.h., Bernard, L., Stackebrandt, E., 1999. Sulfitobacter mediterraneus sp.
nov., a new sulfite-oxidizing member of the alpha-Proteobacteria. Int. J. Syst.
Bacteriol. 49, 513e519.
Ravenschlag, K., Sahm, K., Amann, R., 2001. Quantitative molecular analysis of the
microbial community in marine arctic sediments (Svalbard). Appl. Environ.
Microbiol. 67, 387e395.
Refait, Ph, Memet, J.B., Bon, C., Sabot, R., Ge
nin, J.-M.R., 2003. Formation of the
Fe(II)eFe(III) hydroxysulphate green rust during marine corrosion of steel.
Corros. Sci. 45, 833e845.
Refait, Ph, Girault, P., Jeannin, M., Rose, J., 2009. Influence of arsenate species on the
formation of Fe(III) oxyhydroxides from Fe(II). Colloids Surfaces Physicochem.
Eng. Aspects 332, 26e35.
Refait, Ph, Nguyen, D.D., Jeannin, M., Sable
, S., Langumier, M., Sabot, R., 2011.
Electrochemical formation of green rusts in deaerated seawater-like solutions.
Electrochim. Acta 56, 6481e6488.
Ruby, C., Abdelmoula, M., Naille, S., Renard, A., Khare, V., Ona-Nguema, G., Morin, G.,
Ge
nin, J.-M.R., 2010. Oxidation modes and thermodynamics of FeIIeIII oxy-
hydroxycarbonate green rust: dissolutioneprecipitation versus in situ depro-
tonation. Geochim. Cosmochim. Acta 74, 953e966.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989 Molecular Cloning: a Laboratory
Manual, second ed., 3-vol. set. Cold Spring Harbor Laboratory Press.
Sikorski, J., Munk, C., Lapidus, A., Djao, O.D.N., Lucas, S., Del Rio, T.G., Nolan, M.,
Tice, H., Han, C., Cheng, J.F., Tapia, R., Goodwin, L., Pitluck, S., Liolios, K.,
Ivanova, N., Mavromatis, K., Mikhailova, N., Pati, A., Sims, D., Meincke, L.,
Brettin, T., Detter, J.C., Chen, A., Palaniappan, K., Land, M., Hauser, L., Chang, Y.J.,
Jeffries, C.D., Rohde, M., Lang, E., Spring, S., Go €ker, M., Woyke, T., Bristow, J.,
Eisen, J.A., Markowitz, V., Hugenholtz, P., Kyrpides, N.C., Klenk, H.P., 2010.
Complete genome sequence of Sulfurimonas autotrophica type strain (OK10(T)).
Stand. Genomic Sci. 3, 194e202.
Stackebrandt, E., Goebel, B.M., 1994. A place for DNAeDNA reassociation and 16S
ribosomal- RNA sequence analysis in the present species definition in bacteri-
ology. Int. J. Syst. Bacteriol. 44, 846e849.
Suzuki, M.T., Preston, C.M., Chavez, F.P., DeLong, E.F., 2001. Quantitative mapping of
bacterioplankton populations in seawater: field tests across an upwelling
plume in Monterey Bay. Aquat. Microb. Ecol. 24, 117e127.
Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: molecular evolutionary
genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol. 24, 1596e1599.
Venzlaff, H., Enning, D., Srinavasan, J., Mayrhofer, K.J.J., Hassel, A.W., Widdel, F.,
Stratmann, M., 2013. Accelerated cathodic reaction in microbial corrosion of iron
due to direct electron uptake by sulfate-reducing bacteria. Corros. Sci. 66, 88e96.
Yokokawa, T., Nagata, T., 2005. Growth and grazing mortality rates of phylogenetic
groups of bacterioplankton in coastal marine environments. Appl. Environ.
Microbiol. 71, 6799e6807.
Yu, L., Duan, J., Du, X., Huang, Y., Hou, B., 2013. Accelerated anaerobic corrosion of
electroactive sulfate-reducing bacteria by electrochemical impedance spec-
troscopy and chronoamperometry. Electrochem. Commun. 26, 101e104.
Zegeye, A., Ona-Nguema, G., Carteret, C., Huguet, L., Abdelmoula, M., Jorand, F.,
2005. Formation of hydroxysulphate Green Rust 2 as a single iron(IIeIII) min-
eral in microbial culture. Geomicrobiol. J. 22, 389e399.