Construction and Building Materials 403 (2023) 133180

Contents lists available at ScienceDirect

Construction and Building Materials

Biogalvanic cathodic protection of reinforced concrete structures in

A R T I C L E I N F O A B S T R A C T

Keywords: Cathodic protection is an efficient solution that is implemented on reinforced concrete structures to address
Steel reinforced concrete corrosion issues and prevent further deterioration. However, it has significant drawbacks in terms of sustain­
Bioanode ability. This paper explores a new biological CP technology for reinforced concrete structures in marine envi­
Modelling studies
ronments, where a bioanode is buried in marine sediments, and electrons are supplied through the oxidation of
Numerical Simulation
various substrates by electro-active micro-organisms. The issue was addressed both numerically and experi­
Cathodic Protection
Corrosion mentally, with a first-of-a-kind experimental verification on a laboratory specimen. This new solution offers three
fundamental advantages: autonomous operation, green solution, and free current supply.

1. Introduction ageing of the RC built heritage is not addressed in an effective way, often
leading to irreversible structural deterioration, mainly due to corrosion
The world’s built heritage is aging, especially reinforced concrete [7]. In France, it was estimated in 2020, that around 41.1% of bridges
structures, which are increasingly affected by steel corrosion. The required maintenance, with 4.2% showing deterioration likely to lead to
maintenance of such structures is costly. The overall costs associated a reduction in load-bearing capacity in the short term [8]. Subsequently,
with corrosion average 4% of the GDP of industrialized countries each old structures often have to be demolished and rebuilt, leading to
year [1], adding up to billions of US$ for civil engineering structures additional environmental issues such as waste management. The pres­
alone. Recent catastrophic events linked to corrosion - such as the ervation of the built heritage has therefore become a relevant response
collapse of the Morandi bridge in Genoa (Italy) [2–4] - have reminded to the global environmental challenges, as rehabilitation can lengthen
the world of the need to apply strict maintenance strategies to preserve the service life of a structure and save up to several years’ worth of CO2
infrastructures. Beyond obvious safety and economic benefits, devel­ emissions and resource consumption by reducing the yearly needs in
oping innovative, more efficient, and sustainable anticorrosion ap­ cement and raw materials.
proaches is first and foremost an environmental challenge. In fact, the The deterioration of RC structures is mainly due to the corrosion of
continuous building of reinforced concrete (RC) structures has a major the embedded reinforcing steel. One of the most effective techniques
environmental impact. Cement production alone accounts for 5% of currently available to preserve and lengthen their service life is Cathodic
global CO2 emissions and 2% of the world’s energetic consumption [5], Protection (CP) [9], although the best maintenance solution for a
to which must be added the production and transport of reinforcing steel structure must be assessed on a case-by-case basis, as CP is not always
bars, the consumption and transport of exhaustible resources (sand, suitable [10]. Despite its acknowledged effectiveness to address corro­
aggregates and water), and the destruction of coastal environments and sion issues, CP requires improvements, particularly in terms of envi­
local ecosystems due to sand and aggregates extraction [6]. ronmental impact and sustainability, as the processes involved in
The construction of new structures and the resulting harmful envi­ electron generation are not eco-friendly [11–13] or consume electrical
ronmental impacts could be significantly reduced by using more effec­ energy due to a continuous connection to a DC power supply [14].
tive maintenance strategies for existing RC infrastructures. Today, the Developing an effective and ecological way of providing electrons to

* Corresponding author at: Université de Toulouse, UPS, INSA, LMDC 135 avenue de Rangueil, 31077, Toulouse Cedex 4, France.
E-mail address: [email protected] (J. Dubuit).

https://doi.org/10.1016/j.conbuildmat.2023.133180
Received 31 January 2023; Received in revised form 22 August 2023; Accepted 28 August 2023
Available online 1 September 2023
0950-0618/© 2023 Elsevier Ltd. All rights reserved.
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

protect corroding structures has therefore become of primary interest. potential of active steel becomes more electronegative than that of
In this context, this research work focuses on the CP of steel in RC passive steel, generating a galvanic couple between the active area and
structures in marine environments, as they deteriorate more rapidly and the rest of the passive steel surface. This galvanic coupling leads to the
severely than RC structures on land [15]. In particular, an innovative formation of a growing corrosion spot which acts as the anode of the
and eco-friendly CP technology is being investigated, based on the electrochemical cell, supplying electrons via iron dissolution: Fe → Fe2+
concept of Benthic Microbial Fuel Cells (BMFC) which has been inten­ + 2 e- (Fig. 1 a - Left, b). These electrons are consumed by oxygen
sively developed over the last two decades [16–19]. This emerging CP reduction occurring on the surrounding passive steel surface, which acts
technology is referred to as BioGalvanic Cathodic Protection (BGCP) in as the cathode [23,24]: O2 + 2 H2O + 4 e- → 4 OH–.
the following sections. In marine environments, RC structures often face multiple exposure
conditions: part of the structure is permanently immersed in seawater,
1.1. Corrosion of steel in reinforced concrete another part is subject to tide variations and the rest is constantly above
water (Fig. 1 - a). Under these conditions, chloride ion attack on the
Today, steel corrosion is recognised worldwide as the main cause of reinforcement is the main cause of corrosion initiation (steel depassi­
accelerated deterioration of RC structures under the action of an vation). The alternating cycles of wetting and drying, due to tides,
aggressive environment (seawater, sewers, CO2, etc.). All categories of lapping waves or swell, around the seawater/air interface favour the
RC structures are affected, from simple buildings to strategic in­ diffusion of oxygen and chloride ions into the concrete, towards the
frastructures such as bridges, nuclear power plants, harbours, etc. steel, which intensifies the corrosion kinetics. In fact, oxygen is the drive
When cast in fresh concrete, steel is subject to passivation due to the of the cathodic reaction and chloride ions are a direct cause of corrosion
high alkalinity of the concrete pore solution. Passivation is the rapid initiation [25,26]. The tidal zone is therefore the most critical area
formation of a dense and thin oxide film on the surface of the steel [20]. regarding the risk of corrosion propagation and structural deterioration.
Under normal conditions, when the concrete is sound, steel is in a uni­ The severity of the attack on the reinforcement strongly depends on
form passive state, as expressed in the Pourbaix diagram [21]. the climatic and exposure conditions of the concrete, since corrosion is
The breakdown of the passive oxide film in ageing concrete exposed generally more severe in humid environments, since a higher saturation
to aggressive environments – caused either by carbonation or chloride degree of the concrete decreases its resistivity [27]. In fact, corrosion
attack [22] - leads to the formation of an active steel area. Once kinetics depend on several factors, including:
depassivation has occurred on the steel surface, the electrochemical

Fig. 1. Principle of steel corrosion and biogalvanic cathodic protection illustrated on a reinforced concrete pier partially immersed in seawater. a)-Left Reinforced
concrete pier freely corroding in the tidal zone (no cathodic protection). a)-Right Reinforced concrete pier with a corroding area in the tidal zone under biogalvanic
cathodic protection (BGCP). b) Details of the free corrosion system in the tidal zone, highlighting corrosion current streamlines c) Details of the same corrosion
system under BGCP, highlighting the mitigation of corrosion current streamlines (in red) induced by the cathodic protection current streamlines (in blue) emitted by
the biological anode system, obtained from Comsol Multiphysics® calculations d) Schematic view of the biological anodic process occurring in sediments and
supplying electrons. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

- the overall redox reactions involved in the electrochemical process, their surface (Fig. 1 – d). The chemical species oxidised by the electro­
- the driving force between anodic and cathodic sites, active bacteria on the anodes are reduced inorganic or organic com­
- the local oxygen availability in the vicinity of the reinforcing steel, pounds resulting from the biogeochemical cycles of sulphur and carbon
- the electrical resistivity of the concrete [28] – and therefore the [47]. These compounds are fuel for the BMFC, since their bio-
saturation degree of the concrete pores and chloride content [29], electrochemical decomposition [48] leads to the supply of a flow of
- the geometry of the electrochemical cell [30,31], electrons collected by the anodes, generating an electric current. The
- the temperature [32]. electric currents harvested by BMFC are sustainable since the necessary
fuel is continuously regenerated by natural processes in the sediments,
1.2. Cathodic protection of steel in reinforced concrete from the death and biodegradation of macro- and micro-organisms, the
decomposition of plants, algae, etc. Several BMFC have produced elec­
To slow down the deterioration of RC structures affected by corro­ tricity continuously for more than a year [49–52].
sion, the use of CP has become increasingly common these past decades The application of BMFC’s principles to cathodic protection was re­
[33–36]. CP is implemented to mitigate or even stop the corrosion ki­ ported very recently [53,54], on metallic elements such as pipes.
netics by supplying electrons to the reinforcing steel, preventing iron However, the application to RC structures in marine environment has
dissolution (Fig. 1) [37,38]. A CP system is an electrochemical cell not been mentioned and no experimental evidence has been found in the
consisting of the metallic structure to be protected, here steel, and an scientific literature regarding the actual performance of BioGalvanic
additional anode system. In the case of CP of RC structures, the anode Cathodic Protection (BGCP). BGCP is based on the general principle of
system is either external (applied to the surface of concrete), or internal BMFCs, except that the reinforcing steel to be protected becomes the
(inserted into the concrete). The anode system must be in electrolytic cathode of the electrochemical cell. A metallic wire connection allows
and electronic contact with the protected structure to allow ionic and for electrons to be continuously supplied from the bioanode - located in
electronic current flow, respectively. Electrons are generated by the the sediment layer - to the steel to be protected, similarly to the prin­
anode system and transported to the reinforcing steel bars through an ciples of GCCP and ICCP systems described above (Fig. 1 – a – Right).
electronic connection (metallic wire). Two main CP technologies are BGCP is very promising as a cathodic prevention technique since the
currently available: electrochemical potential of passive steel in RC (-350 to + 50 mV/SCE)
[55] is quite high with regards to that of anodes buried in marine sed­
- Galvanic Current Cathodic Protection (GCCP), also referred to as iments (-535 mV/SCE [56] or − 781 mV/SCE [57] are examples of
Sacrificial Anodes Cathodic Protection (SACP). bioanode open circuit potentials reported in literature). Therefore, a
- Impressed Current Cathodic Protection (ICCP). protective electric current can flow freely, as in a classic galvanic couple,
from the BMFC anode to the passive steel. In the case of active corrosion
GCCP involves a sacrificial metallic material, acting as the anode in the tidal zone, the electron flow supplied by the bioanode is likely to
system. The sacrificial material must be much more electronegative than mitigate the kinetics of iron dissolution (Fig. 1 – c).
the metal to be protected for a galvanic coupling to occur between the
sacrificial anode and the protected metal. Electrons are generated by the 2. Preliminary numerical simulation of BGCP performance
corrosion of the sacrificial anode system - usually zinc in the field of CP
applied to RC structures. The lifespan of GCCP is limited and requires The ISO 12696 standard [10] states that the current range required
regular maintenance as the anode is consumed. Moreover, the con­ for the protection of steel in RC is usually between 2 and 20 mA/m2 of
sumption of the sacrificial material leads to the release of metal ions into steel. To assess the theoretical ability of bioanodes to deliver a sufficient
the environment [11] (Zn, Mg, Al, Cu, Fe, Mn, Si, In, Ga, Hg), which level of protective current, a preliminary numerical case study was
results in the contamination of the environment, the occupying fauna, explored. The preliminary numerical simulation was performed using
and eventually the entire associated food chain [11–13]. the Finite Element (FE) software COMSOL Multiphysics® in steady-state
In the case of ICCP, depending on the pH of the electrolyte, electrons condition. The model was built using a multiphysics approach, meaning
are provided by oxidation of water or hydroxide ions at the surface of an that several physical phenomena were coupled to best represent the
inert anode [39]. As it is not a spontaneous reaction, a DC power supply complexity of cathodic protection of steel in reinforced concrete.
must be used to force this reaction. The main risk for the protected steel
is an over-polarization of the steel due to an excessive current supply
which can lead to the release of hydrogen and cause steel embrittlement 2.1. Modelling approach
followed by brittle failure in prestressed concrete structures [10]. Also,
ICCP requires a complex implementation set-up since a permanent The geometrical model involved a 12 m-high cylindrical RC pier (1
connection to an electrical distribution network throughout the service m-diameter), partially immersed in seawater (5 m) and founded in a 1
life of the protection system is required to impress a flow of electrons. m-thick sediment layer. The steel reinforcement was composed of 12
However, CP represents significant financial costs mainly due to the longitudinal steel rebars with a 2.5 cm-diameter and 40 circular trans­
complexity of in-situ implementation. Thus, worldwide, only a few versal steel rebars with a 1 cm-diameter spaced by 30 cm. The rein­
thousand reinforced concrete structures are equipped with CP, many of forcing steel density in the pier was around 0.5 m2 of steel per m2 of
them in marine environments [40]. Under these circumstances, inves­ concrete surface, which is commonly observed in reinforced concrete
tigating the concept of BMFC as a current supply for the CP of RC structures. In accordance with the exposure conditions of the structure
structures in marine environments appears as highly relevant, as it is to marine chlorides, the concrete cover thickness was set to 6 cm [58]. A
fully autonomous, little costly, and environmentally-friendly [41]. set of 4 bioanodes was embedded in the sediment layer (Fig. 2 – a). The
steel was considered in a uniformly passive state, i.e., the chlorides have
1.3. From BMFC to BGCP not yet reached the steel at this stage of the structure’s life.
The numerical simulation reproduces the 3D galvanic coupling be­
The concept of BMFC was highlighted in the early 2000s in coastal tween the 4 bioanodes and the reinforcing steel bars in concrete. The
marine environments [42,43]. BMFC consist of a microbial anode buried numerical modelling approach relies on several coupled physical phe­
in sediments (anaerobic environment) and a cathode located in the nomena considering the:
overlying seawater (rather aerobic area). Marine sediments naturally
contain electroactive bacteria [44,45] that spontaneously colonise an­ - Electrical transfers in the various electrolytes involved (concrete,
odes to form biofilms [46] and maintain electrooxidation processes on seawater, and sediments): local Ohm’s law and charge conservation.

3
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

Fig. 2. Preliminary numerical simulation. a) Numerical model involving a reinforced concrete pile partially immersed in seawater and bioanodes embedded in the
sediment layer b) Throwing of the cathodic protection current from the bioanodes to the steel reinforcements. The protection current appears to be focused on the
steel located in the tidal zone (highest corrosion risk), while little CP current is collected by steel in areas with low oxygen availability.

- Electrochemical behaviour of passive steel in concrete and bioanodes - Electrochemical properties of passive steel in concrete (Butler-
in the sediment layer: polarization curves modelled by Butler-Volmer Volmer parameters): corrosion of steel in reinforced concrete is an
equations. electrochemical process which can be described according to the
( ( )
- Oxygen transport by diffusion in seawater and concrete, following Butler-Volmer equation: i = icorr × exp ln(10)(E− Ecorr )
− exp
Aachib et al.’s modelling approach [59].
βa
( ))
− ln(10)((E− Ecorr )
βc [24], where icorr is the corrosion current density
To perform the coupling of these physical phenomena, the method­ [A/m2], Ecorr the free corrosion potential of passive steel [V vs ref], βa
ology introduced by Garcia et al was implemented [60], the only dif­ the anodic Tafel slope [V/dec], βc the cathodic Tafel slope [V/dec]:
ference being that the corrosion module of COMSOL Multiphysics® was Ecorr = -0.145 V/SCE, icorr = 10-4 A/m2, βa = 1 V/dec, βc = 0.2 V/dec.
used instead of the electric current module. - Electrochemical properties of the bioanode: due to a very stiff po­
The following parameters were used as typical input data for the larization curve compared to that of passive steel, the bioanode can
numerical simulation: be considered as a non-polarizable electrode. Therefore, it is
modelled by a constant potential: Ecorr = − 0.54 V/SCE [67].
- 5 S/m was chosen as seawater conductivity. - In COMSOL Multiphysics®, oxygen availability is considered using
- The sediments electrical conductivity was chosen according to the physics Transport of Diluted Species (tds). The concentration of
Archie’s Law [61] which states that ρsed = ρw × p− m [ohm.m], where oxygen in air is applied to the surfaces exposed to air (cg) while the
ρsed is the sediments resistivity, ρw the resistivity of the fluid satu­ concentration of oxygen in water (caq) is applied to the surfaces
rating the sediments pore (i.e. seawater), p is the porosity of the under water. Fick’s law governs oxygen diffusion in water by
sediments (80% [62]), m is an empirical parameter that depends on applying the effective diffusion coefficient of oxygen in water. Ox­
the nature of the sediments - a typical value of 1.8 can be taken for ygen consumption and the current kinetics are coupled via a modi­
marine sediments with high concentration of biodebris [63]. The fied Butler-Volmer equation taking mass transport of oxygen through
( ( )
electrical conductivity is related to the resistivity according to the
concrete into account: i = icorr × exp ln(10)(E− Ecorr )
− cgc exp
following equation σ = 1ρ, thus σ sed was equal to 3.35 S/m. ( ))
βa
ln(10)(E− Ecorr )
- Concrete porosity: 12 %, recommended porosity of reinforced con­ β . “c” represents the variable associated with oxygen
c

crete structures for marine applications [64,65]. concentration in the physics “tds”. In addition, a “flux” condition is
- Water saturation degree of concrete varying in the pier’s height ac­ applied to passive steel and takes into consideration the current ki­
cording to a sigmoid function: from 71 % in the upper emerged zone netics from the Butler-Volmer equation: Jn = − i4F
c
, F being Faraday’s
to 100% under the tidal zone. constant. ic which is the corrosion current must be replaced by the
- Concrete resistivity (ρ) depending on the water saturation degree associated name in the COMSOL physics that is being used (for
according to the following equation: ρ = 50227 × (p × Sr× example cd.itot when using the physics Secondary Current Distri­
100)− 2.53 [Ωm], where p is the porosity of the concrete and Sr the bution (cd), the flux equation becomes Jn = cd.itot
− 4F ). Please refer to
saturation degree. This empirical relationship is based on experi­ Garcia et al’s (2019) article where the coupling of the two physics is
mental results from the French research project SENSO [66], carried described thoroughly [60].
out on a wide range of concrete mixtures.

4
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

2.2. Numerical results 3. Experimental verification of BGCP performance

Fig. 2 – b shows the simulated CP current streamlines distribution in To validate the numerical results discussed above, a 10 m-length
the overall electrochemical system. This 3D representation highlights a reinforced concrete specimen was designed to reproduce the exposure
preferential CP current exchange between the bioanodes and the steel conditions of a RC pier partially immersed in seawater. The specimen
bars located in the tidal zone which, as mentioned above, is the most was cast horizontally for practical purposes (Fig. 3 – a, Fig. 3 - d).
critical region of the structure regarding the risk of corrosion damage.
In Fig. 2 – c, the CP current density locally collected by the rein­
forcing steel bars is plotted as a function of the structure’s height. A peak 3.1. Materials and methods
in the CP current distribution is clearly observed in the tidal zone, which
is a logical consequence of the CP current streamlines pattern discussed 3.1.1. Preparation of the reinforced concrete specimen
above. Around the tidal zone, the CP current density supplied by the
bioanodes is higher than 2 mA/m2 of steel, i.e. within the range of 3.1.1.1. Concrete mix design. The concrete mix design for 1 m3 of con­
cathodic protection current densities mentioned in the ISO 12696 crete was as follows:
standard. The simulated throwing power of the CP current supplied by
the bioanodes in this critical zone of the structure was considered as a - 425 kg of binder (composed of 297 kg of CEM I 52.5 R Portland
very encouraging result, requiring further experimental verification, cement and 128 kg of fly ashes),
presented in the following section. Elsewhere, most of the steel areas - 1112 kg of gravel from the Sablières Malet’s production site of
collect CP current densities in the range of cathodic prevention (between Portet-sur-Garonne, France (siliceous alluvial gravels with a d/D of
0.2 and 2 mA/m2). 4/10 mm),
Deeper passive steel in the immersed part of the RC structure receives - 741 kg of sand from the Sablières Malet’s production site of Portet-
little protective current as oxygen availability is low. In stagnant sur-Garonne, France (a siliceous alluvial sand with a d/D of 0/4),
seawater, oxygen concentration is mainly governed by diffusive transfer - 191.25 kg of water.
from the surface, which is a slow phenomenon coupled to several bio­
logical processes (photosynthesis, respiration and nitrification [68,69]). 3.1.1.2. Casting of the steel reinforcement blocks. To study the distribu­
tion of the CP current alongside the length of the specimen, the

Fig. 3. Experimental setup implemented to assess the protection current distribution in a reinforced concrete element simulating partial immersion in sea water. a)
Schematic general view of the reinforced concrete element embedding discontinuous steel bars, the sea water PVC tank involved to maintain a part of the concrete
element in saturated condition, the bioanode system at the end of the simulated submerged zone and embedding sea mud and carbon electrodes. b) Precast concrete
elements embedding steel rods with welded connection wires. c) View of the boundary between the emerged and immersed zones. d) General view of the experi­
mental specimen. d) View of the bioanode system and its connection to the sea water tank.

5
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

reinforcement consisted of a 1 cm-diameter steel rebar, discretized into was embedded into the concrete from the immersed zone to the tidal
30 rods of varied lengths. The various rods were distributed along the zone (Fig. 3 - c).
specimen to have a large number of current measurements in the critical
zones, namely the tidal and splash zones, and therefore to later better 3.1.1.6. Oxygen insulation of the lateral surfaces of the specimen. Since
visualize the current distribution during data analysis. The steel seg­ the water was only applied to the upper surface of the specimen, all
ments were of 4 different lengths: 20 cm for the immersed part of the lateral surfaces were coated with epoxy resin after unmoulding of the
specimen, 5 cm for the transition part between the immersed and tidal concrete element to prevent or at least limit lateral oxygen penetration.
zone, 2 cm for the tidal zone and 10 cm for the emerged part.
BGCP being developed for marine environments, this experimental 3.1.2. Preparation of the biogalvanic cathodic protection system
pilot was designed in accordance with the European standards applied to Near the end of the pier, where anoxic conditions are targeted, a
RC structures in marine environments [58]. Since the specimen was set- bioanode was placed in a plastic tank filled with natural seawater and
up horizontally, marine exposure conditions were only applied to the marine sediments collected from the Mediterranean Sea (Fig. 3 - e).
upper surface of the specimen. The concrete cover thickness from the
steel to the upper surface exposed to seawater was thus set to 6 cm. The 3.1.2.1. Bioanode activation. The bioanodes were designed prior to the
bottom surface did not need to comply with a cover thickness of 6 cm as cathodic protection experiments using the constant polarization method
it was not exposed to seawater. The width of the pier being 10 cm and which increases the growth kinetics of electrogenic biofilms. The bio­
the diameter of the steel being 1 cm, the concrete cover thickness from anodes were developed at room temperature in a plastic tank containing
the bottom surface of the pier was equal to 3 cm. The steel segments 30L of seawater supplemented with acetate, to a final concentration of 8
were precast in concrete blocks, enabling the required bottom cover mM, and inoculated with 5% (V/V) marine sediment samples. A con­
thicknesses to be achieved. To do so, steel rebars were held at the desired ventional three-electrode system was implemented and controlled using
height using 3 cm-high plastic clip spacers. The blocks were cast in a multi-potentiostat (VMP2 Bio-Logic SA).
7x7x28 cm3 laboratory concrete moulds. To ensure the stability of the Acetate was chosen to feed the bacteria since it is one of the major
steel during concrete pouring, the clip spacers were fixed to the moulds components found in their natural habitat – i.e. sediments. The oxida­
using silicone glue. After curing, the concrete blocks were demoulded tion of acetate leads to the release of electrons according to the following
and cut to length using a diamond blade water saw. It was important to equation [70,71]:
use a water saw to prevent the concrete from overheating. As the blocks
contain steel, the water limited the temperature rise caused by contact CH3COO– + 2 H2O → 2 CO2 + 7 H+ + 8 e-.
between the blade and the steel and prevented any damage to the steel/
This reaction is associated with a standard redox potential
concrete interface.
ofE0′CO2 /Acetate = -0.24 V/SHE (-0.484 V/ SCE) at pH 7 [72].
The plastic tank was hermetically sealed, with no gas flow. Three
3.1.1.3. Welding of electric wires to the steel segments for current mon­ carbon felt electrodes (200*500 mm) were embedded vertically in the
itoring. After the concrete blocks had been cut, the cross-section of the bottom of the tank and electrical contact was established with titanium
steel segments was exposed on both sides of the blocks. Before the wires. The auxiliary electrode was a set of 3 graphite rods. The potential
specimen was cast, electric wires were tin-welded to the steel sections to of the carbon felt electrodes was fixed at − 100 mV/SCE for a period of
then be connected to a data logger to track the current received by each two weeks to promote its colonisation by electroactive microbial species
steel segment. organised in the form of a biofilm [48,73]. After the two weeks of
constant polarization, a stable average current of 80 mA was reached for
3.1.1.4. Building of a PVC tank to reproduce immersion conditions. To several days. Chronoamperometry was stopped and a Linear Sweep
reproduce immersion conditions, a bottomless PVC tank was incorpo­ Voltammetry was performed in situ on the bioanodes to obtain their
rated into the freshly poured concrete. The 7-meter-long tank was polarization curve, describing their electrochemical behaviour. The
constructed using PVC strips bonded with PVC glue. Predrilled vertical potential was increased from open-circuit potential (-0.409 mV/SCE) to
PVC struts – for water drainage - were placed every meter, contributing + 300 mV/SCE at a scan rate of 1 mV/s.
to stiffening the structure. The strut located 6 m from the base of the tank
was plugged to ensure waterproofness and create two distinct com­ 3.1.2.2. Connection of the bioanode to the pier. The anodic container was
partments. As the two compartments are independent, one can be connected to the PVC tank through a flexible tube, and the water level
emptied while the other remains full simulating constant immersion on was adjusted to obtain a hydraulic connexion (electrolytic connection
one side and tidal range on the other. The end of the tank was also drilled between the bioanode and the RC specimen). The bioanode was then
to be connected to the anodic container via a flexible pipe (Fig. 3 - e). electrically connected to the steel segments in the RC specimen by
metallic wires (electronic connection via the data acquisition system).
3.1.1.5. Casting of the large-scale concrete specimen. Before casting the
large-scale specimen, the precast concrete blocks containing the steel 3.1.3. Current monitoring system
segments had to be held in place to ensure their stability while pouring The electrical connections between the bioanode and the steel seg­
the concrete. They were fixed with silicone glue to the bottom of the ments (cathode) involved a current monitoring system. The steel seg­
mould (Fig. 3 - b). Their spacing was drawn beforehand on the bottom of ments were connected to the bioanode in a parallel electrical circuit to
the mould, to respect the discretization along the pier according to the evaluate the CP current distribution in the RC pier. Each branch of the
exposure conditions. parallel circuit was fitted with a shunt resistor and connected to the data
Once the silicone holding the blocks had dried and the PVC tank had logger. This electrical arrangement made it possible to measure the
been assembled, the same concrete mix as for the blocks (see 3.1.1.1) current collected by each steel segment and, subsequently, to assess the
was poured in the pier’s mould, covering the blocks. After casting of the throwing power of the CP current within the electrochemical cell. Zn
pier the steel segments are electrolytically coupled through the pore potential sensors were also installed on the concrete surface - one sensor
solution of the concrete (blocks and pier). facing each steel segment - to monitor the polarization and depolariza­
Immediately after placement, while the concrete was still fresh, the tion responses of the steel under BGCP. The acquisition unit was con­
PVC tank was delicately hammered 1 cm into the concrete. This ensured nected to a computer, enabling the current received by each steel
a watertight sealing between the concrete and the PVC tank, to prevent segment to be displayed for a given time.
any water leaking out after filling of the tank with seawater. The tank

6
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

3.2. Experimental results m2 of steel. Although closer to the bioanode, they collect less CP current
than the segment of the tidal zone, due to lower oxygen availability (fuel
The steel segments were connected to the bioanode which started of the cathodic reaction).
delivering CP current that increased until it reached a peak value five As predicted by the FE simulations, this experiment confirms that the
days later (blue scatterplot, Fig. 4 – b). The current and potential CP current is mainly focused in the tidal zone and flows directly to this
monitoring of the BGCP system was carried out for over two years. Fig. 4 area which entraps large amounts of oxygen and where concrete re­
- b presents the spatial distribution of the CP current supplied by the sistivity remains quite low due to constant moist conditions.
bioanode at different times from the moment the peak value was Regarding BGCP’s sustainability, a progressive decrease of the CP
observed. current collected by each steel segment is observed over time, high­
Regarding the CP current, a clear peak was observed in the tidal zone lighted by the decrease in the CP peak value in the tidal zone (see data
of the experimental specimen, as predicted by the preliminary simula­ relative to 10 weeks – No bacteria feeding and 41 weeks – No bacteria
tion (Fig. 2 – b). Moreover, the experimental peak value (23 mA/m2 of feeding with respect to the initial CP distribution, red and black scat­
steel after 5 days) was found significantly higher than the preliminary terplots respectively). In this experiment, the bioanode’s reservoir is a
numerical result (6 mA/m2 of steel), meaning that the experimental small closed system, not allowing for natural renewal of the chemical
BGCP system was more efficient than calculated by the preliminary species oxidized by electroactive bacteria. The addition of artificial
theoretical study, which was based on semi-arbitrary parameters. organic matter then made it possible to further increase the CP current
It must be noted that no quantitative correlation was expected here delivered (see data relative to green scatter plot, 41 weeks – After bac­
since the numerical predictions were based on a different structural teria feeding). Today, more than two years after starting the BGCP
geometry and semi-arbitrary simulation parameters. Nevertheless, the system, by regularly adding 10 mM of sodium acetate as fuel for the
attention was focused on qualitative correlation to achieve the first electroactive bacteria, the bioanode is still delivering significant pro­
experimental proof of concept of BGCP. tective current densities to the steel segments (see data relative to 115
Fig. 4 – b displays the prescriptive limit between cathodic prevention weeks – After bacteria feeding). However, the latter experiment un­
(CP current density from 0.2 to 2 mA/m2 of steel) and cathodic pro­ derlines the dependence of the system on the regeneration of nutrients in
tection (>2 mA/m2 of steel), showing that all steel segments up to 7.5 m the sediments. This issue is of primary importance as it governs the
in height are in the cathodic protection domain. The upper steel seg­ operational relevance and the sustainability of BGCP systems.
ments are however in the cathodic prevention domain, meaning that a According to current standards applied to cathodic protection of steel
low level of protection is still provided by the bioanode, even 10 m away in RC structures, achieving a specific current density level is not suffi­
from the bioanode. The deeper steel elements are also in the cathodic cient to prove the performance of a CP system. The efficiency of the
protection domain, with collected current densities higher than 5 mA/ experimental BGCP system was therefore assessed according to the

Fig. 4. Experimental results. a) Schematic view of the reinforced concrete element embedding discontinuous steel bars and the CP current monitoring system. b)
Experimental distribution of the cathodic protection current density along the height of the experimental specimen: CP current evolution over 9 months without
feeding the bioanode bacteria; long term performance of the CP system after over two years in operation. c) Experimental distribution of the steel depolarization
collected 24 h after bioanode disconnection (cathodic protection cut-off).

7
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

second performance criterion specified in ISO 12696, which is based on the applied parameters. To determine the geometrical factor G, the
steel depolarization measured 24 h after CP has been switched-off. following formula was used: ρ = G×ΔV I , with ρ being the electrical re­
To assess the standard performance of the experimental BGCP system sistivity (Ωm), ΔV the voltage difference between the two inner points
discussed above, the bioanode was temporarily disconnected from the (V) and I the applied current (A).
steel segments. The 24-hour depolarization of each steel segment was After G was determined with the numerical model, the same formula
measured and plotted in Fig. 4 - c. All segments up to a height of 7.5 m was applied using the experimental data which led to an electrical re­
showed 24-hour depolarization values higher than + 100 mV and were sistivity of the concrete of 420 Ωm in a saturated configuration.
therefore found to be under cathodic protection according to the stan­
dard. These depolarization results are fully consistent with the results of 4.1.2. Porosity measurement
CP current distribution discussed above and demonstrate that this Porosity measurements were performed on the concrete to determine
experimental BGCP system meets the standard performance require­ the voids accessible to water following the French standard NF-P18-459
ment in the tidal zone, which is the most critical part of the structure [74]. For each concrete casting (blocks and pier), the mean porosity was
regarding the risk of corrosion damages for RC structures in marine the same, namely, about 16.5 %.
environments. Conversely, the highest steel segments (above 7.5 m) do
not meet this performance criterion. 4.1.3. Electrochemical properties of steel in the experimental concrete
To determine the Butler-Volmer (BV) characteristics of steel in the
4. Numerical simulation of BGCP according to the laboratory concrete used, cylindrical specimens were cast in molds where a
experimental specimen segment of steel rebar was inserted vertically. About 2 cm of the rebar
was above the upper surface of the cylinder. The concrete was left to
This first-of-a-kind experiment on the feasibility of BGCP was then cure for 28 days and was put in water up to 1 cm bellow its upper sur­
modelled and simulated using COMSOL Multiphysics®. A three- face. The steel segment was connected to a potentiostat (PARSTAT) as
dimensional numerical model of the experimental specimen was built the working electrode. The counter electrode was made using stainless
and used to benchmark numerical predictions and experimental data of steel mesh folded to create a hollow cylinder placed around the concrete
CP current distribution (Fig. 4). The 3D geometry of the experimental RC sample and the reference electrode used was Ag/AgCl (GAMRY, Ag/
specimen was reproduced as accurately as possible, in terms of geometry AgCl Sat KCl) The steel was then polarized in anodic and cathodic di­
and various coupled processes, such as oxygen transport, the electro­ rections to obtain the polarization curve. For purposes of homogeneity
chemical behaviour of steel and of the bioanode, as they directly affect within the paper, the obtained potentials were converted with regards to
current exchanges. the saturated calomel reference electrode (SCE).
The recorded data provided the current corresponding to a specific
4.1. Material and methods polarization/potential. Before polarization, the open circuit potential
was recorded, corresponding to Ecorr in the BV equation. IR-drop
Simulation parameters specific to RC, such as concrete porosity and compensation was applied to the potential values, using the resistance
resistivity, as well as the specific Butler-Volmer parameters of the steel for each potential step (increase or decrease) and Ohm’s Law according
in the employed concrete, were measured beforehand, as were the pa­ to the following formula: E = Emeas − Imeas × R, with E the corrected
rameters specific to the bioanode (such as the electrooxidation or pro­ potential, Emeas the measured potential, Imeas the corresponding current
tective current generation kinetics through the polarization curve). and R the resistance associated with the step. The resistance R was
The model was built using a multiphysics approach to consider the determined using Ohm’s Law reasoning in terms of potential and current
influence of oxygen concentration in the concrete volume on the delta. The theoretical expression of the Butler Volmer equation was
cathodic kinetics, and thus on the current exchanges between the steel obtained using a solver tool to fit the experimental data. icorr , βa and βc
and the bioanode. Using the model, the theoretical current distribution were determined:
to the different steel segments was plotted as a function of concrete
porosity, electrical resistivity, oxygen availability, seawater conductiv­ - Ecorr = -0.095 V/SCE
ity and bioanode electrooxidation kinetics. - icorr = 6e-4 A/m2
The first step of this numerical study was to determine the various - βa = 0.39 V/dec
parameters required for accurate modelling of the experimental - βc = 0.335 V/dec
specimen.
The 3D numerical simulation was performed using COMSOL Multi­
4.1.1. Concrete electrical resistivity measurement physics® (version 6.0) in a stationary state. The model was built using a
Concrete from the blocks and the pier was cast in cylindrical molds to multiphysics approach, meaning that two physics were coupled to best
measure the electrical resistivity. The molds for the blocks’ concrete represent the complexity of steel corrosion in reinforced concrete
were 9.39 cm in diameter and 16.6 cm long while the molds for the following Garcia et al’s (2019) methodology [60]. The Butler-Volmer
pier’s concrete were 12.49 cm in diameter and 23.7 cm long. A home­ parameters of passive steel as well as oxygen diffusion were taken into
made four-point probe instrument was connected to a Source Meter account. The steel was considered passive, as the experimental specimen
(Keithley) to impress the current and to a Multimeter (Fluke) to measure was a freshy cast concrete pier. The geometry of the numerical model
the voltage gradient according to the Wenner method. The four-point was based on the experimental specimen. The dimensions of the con­
probe was positioned so that the outer points were at equal distance crete element, the size of the anodic container and overlying water tank
from each end of the cylinders. were numerically reproduced (Fig. 4 -a). The two physics used were
The Wenner measurement was modelled numerically using COMSOL Secondary Current Distribution (cd) and Transport of Diluted Species
Multiphysics® for both configurations (block and pier), to take the (tds) to model the system’s corrosion behavior and oxygen transport
geometrical factor into account. The positions of the probe’s four points respectively. The numerical model considered all parameters affecting
were reproduced. A chosen current intensity of 0.1 mA was applied to the current kinetics. Concrete porosity was applied to the concrete
one of the outer points, representing the inlet - or application point - of volume. The modelled RC pier was divided into three parts to apply the
the current. A value of − 0.1 mA was applied to the other outer point appropriate boundary conditions to the immersed zone, tidal zone and
representing the outlet. A chosen resistivity of 100 Ωm was applied to emerged zone.
the cylinder and a simple electric study was performed to numerically
obtain the voltage difference between the two inner points according to

8
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

4.1.4. Assumptions The electrical resistivity was then related to the hypothesized satu­
Some parameters could not be measured on the pier, thus a few ration degree according to an empirical model proposed in the SENSO
hypotheses were put forward regarding the saturation degree of the project [66], which states that the electrical resistivity ρ can be linked to
concrete along the pier, as well as the oxygen concentration on the outer the saturation degree (Sr) and porosity (p). The following equation was
surfaces. used: ρ = 3.865 × (p × Sr)− 2.53 so that the hypothesized resistivity fit the
Regarding the saturation degree, the concrete in the immersed zone experimental value of 420 Ωm for a saturation degree of 95%, given that
and in part of the tidal zone (up to 6.7 m) was considered fully saturated, the concrete cylinder on which the resistivity was measured was not
as it was constantly immersed in water. The chosen degree of pore 100% saturated.
saturation was set at Sri = 95 %. The concrete in the emerged zone was Regarding oxygen concentration on the outer surfaces of the pier,
considered very dry. In fact, the concrete was cast in June 2020. Here, despite having applied epoxy resin, the paintbrush did not allow for the
the current distribution was numerically modeled to be compared to the resin to penetrate through some surface holes that most certainly allow
latest experimental results - i.e. 115 weeks – After bacteria feeding – oxygen to get through. It was therefore assumed that oxygen was in fact
after two years in operation. The experimental specimen was located present on the epoxy-coated outer surface, up to 0.22 times the con­
indoors, in an uncontrolled environment and was subject to seasonal centration of oxygen in air. It means that instead of applying an oxygen
variations in humidity and temperature. During these two years, severe concentration of 0 mol/m3 on the epoxy coated surfaces of the pier, a
heat waves were experienced, favouring concrete drying. A saturation concentration of 0.22 × cg mol/m3 was applied, with cg being the
degree of Srmin = 50% was assumed. In the tidal zone, the saturation concentration of oxygen in air.
degree was assumed to evolve linearly and rapidly from Sri to Srmin be­
tween 6.7 m and 7 m.

Fig. 5. Comparison of the experimental and numerical results. Comparative study of the experimental results from 115 weeks - After bacteria feeding and the
numerical results obtained using measured and assumed input data.

9
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

4.2. Comparison between experimental and numerical current density a fully efficient and mature technology and improve its outcome, as
distributions there still are limitations to overcome.
So far, cathodic protection has been achieved on passive steel,
The numerical results obtained were in accordance with the exper­ highlighting the ability of the technology to preserve new in­
imental observations. In fact, Fig. 5 shows the general distribution of the frastructures. It is now necessary to address the protection of actively
current density along the pier. In both configurations, the current dis­ corroding steel, by increasing the electromotive force between the bio­
tribution followed the same pattern, the peak being located about 7 m anodes and the cathode (reinforcing steel). Undergoing research is being
from the bioanode – corresponding to the tidal zone – and being carried out to explore chemical or biological ways of making the natural
concentrated over a length of 2 m. Also, the current received by each potential of the bioanodes more electronegative, in order to increase the
steel segment was of the same order of magnitude. electrical power of the BGCP system and reach effective cathodic pro­
Some differences were observed in terms of current distribution be­ tection on active corrosion systems.
tween the model and the experimental data. In fact, despite the many Furthermore, although existing BMFC systems have shown long
tests carried out to obtain the parameters mentioned above, a few pa­ lasting performances, the autonomy of BCGP in the field still needs
rameters remained uncertain, such as the variation in the saturation validation and is currently undergoing testing. In-situ testing is enabling
degree of concrete along the pier. They were all the more difficult to to assess if laboratory experiments can be replicated in real life condi­
predict as they varied over time due to variations in environmental tions and vice-versa. BGCP is also being tested on a large-scale vertical
temperature and relative humidity. Depending on the ambient temper­ RC element in laboratory conditions, to evaluate the influence of the use
ature or humidity, the saturation conditions and oxygen concentration of several anodes, varying steel surface area to be protected, as well as
in concrete vary. When designing the numerical model, it was decided to the influence of tides and sea spray on the current distribution within the
carry out a stationary state study that considered the saturation condi­ structure and on the behaviour of the bioanodes.
tion and oxygen concentration for a given time.
Despite these slight differences, the experimental and numerical re­ CRediT authorship contribution statement
sults show a rather good correlation, highlighting the robustness of the
modelling and simulation approach developed to predict the behaviour Julie Dubuit: Writing – original draft, Visualization, Validation,
of BGCP. Software, Investigation, Formal analysis, Conceptualization. Stéphane
Laurens: Writing – review & editing, Writing – original draft, Valida­
5. Conclusion tion, Supervision, Investigation, Formal analysis, Conceptualization.
Alexandra Bertron: Writing – review & editing, Supervision, Re­
The new concept of BioGalvanic Cathodic Protection (BGCP) consists sources, Project administration, Funding acquisition, Formal analysis,
of a benthic microbial fuel cell, in which the cathode is the steel struc­ Conceptualization. David Garcia: Writing – review & editing, Visuali­
ture to be protected. The electrons needed to protect the steel are sup­ zation, Supervision, Software, Resources, Formal analysis, Conceptual­
plied by a microbial bioanode buried in the marine sediments. The ization. Luc Etcheverry: Writing – review & editing, Resources,
bioanode continuously produces an electric current by converting the Investigation, Conceptualization. Fabrice Deby: Writing – review &
chemical energy in the sediments through the oxidation activity of the editing. Benjamin Erable: Writing – review & editing, Supervision,
electroactive micro-organisms. Investigation, Formal analysis, Conceptualization.
The first BGCP proof of concept has been demonstrated through a
numerical case study, representative of a reinforced concrete pier Declaration of Competing Interest
partially immersed in seawater and founded in a sediment layer. Next,
an experimental pilot was designed and built to carry out laboratory The authors declare the following financial interests/personal re­
tests and assess the relevance of the numerical predictions. These lationships which may be considered as potential competing interests:
experimental results were then replicated through the development of a Julie Dubuit has patent #FR3123662 issued to Assignee. Alexandra
numerical model. Bertron has patent #FR3123662 issued to Assignee. David Garcia has
These numerical and experimental studies have led to consistent and patent #FR3123662 issued to Assignee. Benjamin Erable has patent
correlated results. They have shown that a bioanode located in a sedi­ #FR3123662 issued to Assignee. Stephane Laurens has patent
ment layer can supply a significant CP current density to a large part of #FR3123662 issued to Assignee. Luc Etcheverry has patent
the steel structure. More specifically, most of the supplied CP current is #FR3123662 issued to Assignee. Julie Dubuit has patent #EP4101944
concentrated in the tidal zone of the RC structure, which is the most issued to Assignee. Alexandra Bertron has patent #EP4101944 issued to
critical area regarding the risk of corrosion damage in marine environ­ Assignee. David Garcia has patent #EP4101944 issued to Assignee.
ments. The experiment has also demonstrated the ability of BGCP to Benjamin Erable has patent #EP4101944 issued to Assignee. Stephane
meet the standard performance criteria established in ISO 12696: 2022. Laurens has patent #EP4101944 issued to Assignee. Luc Etcheverry has
This validation of the innovative concept of BGCP of steel rein­ patent #EP4101944 issued to Assignee.
forcement in concrete has proven to be extremely promising as a new
anti-corrosion technique, both autonomous and eco-friendly, to increase Data availability
the durability of infrastructures in aggressive marine environments. The
sustainability of bioanodes from benthic or sediment MFCs is one of Data will be made available on request.
their major assets, as a number of field studies have already established
that they can produce anodic currents continuously for months or even Acknowledgements
years [49–52]. BGCP has several ecological advantages, the most
important being that it has no deleterious impact on the aquatic envi­ The authors wish to thank the French National Agency for Technical
ronment. Bioanodes even generate positive effects created by the Research (ANRT) for its financial support (Grant number: 2020/1510).
oxidizing activity in the sediments, which enables sediments contami­ The authors are grateful to the Laboratory of Construction Materials and
nated by organic pollutants, persistent pollutants, heavy metals and oil, Durability (LMDC) of the National Institute of Applied Sciences of
etc. to be cleaned up [75]. Toulouse (INSA Toulouse) and Paul Sabatier University (UPS), to the
However, despite very encouraging results that have enabled the Laboratory of Chemical Engineering (LGC) at National Polytechnic
concept to be validated experimentally and numerically, this paper, so Institute (INP-ENSIACET Labège) and CORROHM for their financial,
far, only presents a feasibility study. Further work is required to achieve administrative, and technical supports.

10
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

Appendix A. Supplementary data critical review by RILEM TC 281-CCC, RILEM Tech. Lett. 5 (2020) 85–100, https://
doi.org/10.21809/rilemtechlett.2020.127.
[28] W. López, J.A. González, Influence of the degree of pore saturation on the
Supplementary data to this article can be found online at https://doi. resistivity of concrete and the corrosion rate of steel reinforcement, Cem. Concr.
org/10.1016/j.conbuildmat.2023.133180. Res. 23 (1993) 368–376, https://doi.org/10.1016/0008-8846(93)90102-F.
[29] S. Goñi, C. Andrade, Synthetic concrete pore solution chemistry and rebar
corrosion rate in the presence of chlorides, Cem. Concr. Res. 20 (1990) 525–539,
References https://doi.org/10.1016/0008-8846(90)90097-H.
[30] D. Landolt, Electrochemical and materials science aspects of alloy deposition,
[1] G. Koch, 1 - Cost of corrosion, in: A.M. El-Sherik (Ed.), Trends Oil Gas Corros, Res. Electrochim. Acta 39 (1994) 1075–1090, https://doi.org/10.1016/0013-4686(94)
Technol., Woodhead Publishing, Boston, 2017, pp. 3–30, https://doi.org/10.1016/ E0022-R.
B978-0-08-101105-8.00001-2. [31] R.P. Frankenthal, F. Mansfeld, Corrosion and Corrosion Protection: Proceedings of
[2] S. Invernizzi, F. Montagnoli, A. Carpinteri, Fatigue assessment of the collapsed an International Symposium Honoring Professor H.H. Uhlig on His Seventy-fifth
XXth Century cable-stayed Polcevera Bridge in Genoa, Procedia Struct. Integr. 18 Birthday, Electrochemical Society, 1981.
(2019) 237–244, https://doi.org/10.1016/j.prostr.2019.08.159. [32] M. Pour-Ghaz, O.B. Isgor, P. Ghods, The effect of temperature on the corrosion of
[3] M. Morgese, F. Ansari, M. Domaneschi, G.P. Cimellaro, Post-collapse analysis of steel in concrete. Part 1: simulated polarization resistance tests and model
Morandi’s Polcevera viaduct in Genoa Italy, J. Civ. Struct. Health Monit. 10 (2020) development, Corros. Sci. 51 (2009) 415–425, https://doi.org/10.1016/j.
69–85, https://doi.org/10.1007/s13349-019-00370-7. corsci.2008.10.034.
[4] C. Nuti, B. Briseghella, A. Chen, D. Lavorato, T. Iori, I. Vanzi, Relevant outcomes [33] C.L. Page, G. Sergi, Developments in cathodic protection applied to reinforced
from the history of Polcevera Viaduct in Genova, from design to nowadays failure, concrete, J. Mater. Civ. Eng. 12 (2000) 8–15, https://doi.org/10.1061/(ASCE)
J. Civ. Struct. Health Monit. 10 (2020) 87–107, https://doi.org/10.1007/s13349- 0899-1561(2000)12:1(8).
019-00371-6. [34] B.S. Wyatt, Cathodic protection of steel in concrete, Corros. Sci. 35 (1993)
[5] E. Holban, G. Deák, V. Daescu, E. Diacu, A.I. Daescu, G.S. Tanase, P. Marinescu, C. 1601–1615, https://doi.org/10.1016/0010-938X(93)90390-3.
Sirbu, J. Paceagiu, Ways to reduce CO2 emissions and energetic consumption at [35] K. Wilson, M. Jawed, V. Ngala, The selection and use of cathodic protection
clinker producing under Romanian specific conditions, (n.d.) 11. systems for the repair of reinforced concrete structures, Constr. Build. Mater. 39
[6] Sand, rarer than one thinks, Environ. Dev. 11 (2014) 208–218. https://doi.org/ (2013) 19–25, https://doi.org/10.1016/j.conbuildmat.2012.05.037.
10.1016/j.envdev.2014.04.001. [36] J. Mietz, Corrosion of Reinforcement in Concrete (EFC 25): Monitoring, Prevention
[7] G. Koch, N. Thompson, O. Moghissi, J. Payer, J. Varney, IMPACT (International and Rehabilitation, CRC Press, 2020.
Measures of Prevention, Application, and Economics of Corrosion Technologies) [37] P. Pedeferri, Cathodic protection and cathodic prevention, Constr. Build. Mater. 10
Study, NACE International, Houston, 2016. (1996) 391–402, https://doi.org/10.1016/0950-0618(95)00017-8.
[8] Sécurité des ponts : la dégradation se poursuit malgré les mises en garde du Sénat, [38] S. Laurens, R. Francois, Cathodic protection in reinforced concrete structures
Vie-Publiquefr. (2022). http://www.vie-publique.fr/en-bref/285508-securite-des- affected by macrocell corrosion: a discussion about the significance of the
ponts-une-degradation-qui-se-poursuit (accessed May 25, 2023). protection criteria, RILEM Tech. Lett. 2 (2017) 27–32, https://doi.org/10.21809/
[9] A. Goyal, H.S. Pouya, E. Ganjian, P. Claisse, A review of corrosion and protection of rilemtechlett.2017.38.
steel in concrete, Arab. J. Sci. Eng. 43 (2018) 5035–5055, https://doi.org/ [39] E.Q. Zhang, Z. Abbas, L. Tang, Predicting degradation of the anode–concrete
10.1007/s13369-018-3303-2. interface for impressed current cathodic protection in concrete, Constr. Build.
[10] ISO 12696 - Cathodic protection of steel in concrete, (n.d.). Mater. 185 (2018) 57–68, https://doi.org/10.1016/j.conbuildmat.2018.07.025.
[11] J. Deborde, P. Refait, P. Bustamante, C. Caplat, O. Basuyaux, A.-M. Grolleau, M.- [40] CEREMA, Protection cathodique du béton armé - Retour d’expérience, (2017).
L. Mahaut, C. Brach-Papa, J.-L. Gonzalez, S. Pineau, Impact of galvanic anode [41] M.E. Nielsen, C.E. Reimers, H.K. White, S. Sharma, P.R. Girguis, Sustainable energy
dissolution on metal trace element concentrations in marine waters, Water Air Soil from deep ocean cold seeps, Energ. Environ. Sci. 1 (2008) 584–593, https://doi.
Pollut. 226 (2015) 1–14, https://doi.org/10.1007/s11270-015-2694-x. org/10.1039/B811899J.
[12] C. Rousseau, F. Baraud, L. Leleyter, O. Gil, Cathodic protection by zinc sacrificial [42] L.M. Tender, C.E. Reimers, H.A. Stecher, D.E. Holmes, D.R. Bond, D.A. Lowy,
anodes: Impact on marine sediment metallic contamination, J. Hazard. Mater. 167 K. Pilobello, S.J. Fertig, D.R. Lovley, Harnessing microbially generated power on
(2009) 953–958, https://doi.org/10.1016/j.jhazmat.2009.01.083. the seafloor, Nat. Biotechnol. 20 (8) (2002) 821–825.
[13] L. Leleyter, F. Baraud, T. Reinert, S. Gouali, M. Lemoine, O. Gil, Fate of aluminium [43] C.E. Reimers, L.M. Tender, S.J. Fertig, W. Wang, Harvesting energy from the
released by sacrificial anodes – contamination of marine sediments by marine − water interface, Environ. Scence Technol. 35 (2001) 192–195.
environmentally available compounds, Comptes Rendus Geosci. 350 (2018) [44] B. Erable, N.M. Duţeanu, M.M. Ghangrekar, C. Dumas, K. Scott, Application of
195–201, https://doi.org/10.1016/j.crte.2018.05.003. electro-active biofilms, Biofouling 26 (2010) 57–71, https://doi.org/10.1080/
[14] S. Szabó, I. Bakos, Impressed current cathodic protection, Corros. Rev. 24 (2006) 08927010903161281.
39–62, https://doi.org/10.1515/CORRREV.2006.24.1-2.39. [45] D.R. Lovley, Bug juice: harvesting electricity with microorganisms, Nat. Rev.
[15] A. James, E. Bazarchi, A.A. Chiniforush, P. Panjebashi Aghdam, M.R. Hosseini, Microbiol. 4 (2006) 497–508, https://doi.org/10.1038/nrmicro1442.
A. Akbarnezhad, I. Martek, F. Ghodoosi, Rebar corrosion detection, protection, and [46] H.-C. Flemming, J. Wingender, U. Szewzyk, P. Steinberg, S.A. Rice, S. Kjelleberg,
rehabilitation of reinforced concrete structures in coastal environments: a review, Biofilms: an emergent form of bacterial life, Nat. Rev. Microbiol. 14 (2016)
Constr. Build. Mater. 224 (2019) 1026–1039, https://doi.org/10.1016/j. 563–575, https://doi.org/10.1038/nrmicro.2016.94.
conbuildmat.2019.07.250. [47] D.A. Lowy, L.M. Tender, J.G. Zeikus, D.H. Park, D.R. Lovley, Harvesting energy
[16] M.N. Naseer, A.A. Zaidi, H. Khan, S. Kumar, M.T.B. Owais, J. Jaafar, N. from the marine sediment–water interface II: Kinetic activity of anode materials,
S. Suhaimin, Y.A. Wahab, K. Dutta, M. Asif, S.F.W.M. Hatta, M. Uzair, Mapping the Biosens. Bioelectron. 21 (2006) 2058–2063, https://doi.org/10.1016/j.
field of microbial fuel cell: a quantitative literature review (1970–2020), Energy bios.2006.01.033.
Rep. 7 (2021) 4126–4138. [48] M. Rimboud, D. Pocaznoi, B. Erable, A. Bergel, Electroanalysis of microbial anodes
[17] J. Md Khudzari, J. Kurian, B. Tartakovsky, G.S.V. Raghavan, Bibliometric analysis for bioelectrochemical systems: basics, progress and perspectives, PCCP 16 (2014)
of global research trends on microbial fuel cells using Scopus database, Biochem. 16349–16366, https://doi.org/10.1039/C4CP01698J.
Eng. J. 136 (2018) 51–60, https://doi.org/10.1016/j.bej.2018.05.002. [49] C. Reimers, M. Wolf, Power from benthic microbial fuel cells drives autonomous
[18] P.M.D. Serra, E.-s. a,, Sourcing power with microbial fuel cells: A timeline, J. Power sensors and acoustic modems, Oceanography 31 (1) (2018) 98–103.
Sources 482 (2021), 228921, https://doi.org/10.1016/j.jpowsour.2020.228921. [50] C. Zhou, Y. Fu, H. Zhang, W. Chen, Z. Liu, Z. Liu, M. Ying, X. Zai, Structure design
[19] A.E. Franks, K.P. Nevin, Microbial Fuel Cells, A Current Review, Energies. 3 (2010) and performance comparison of large-scale marine sediment microbial fuel cells in
899–919, https://doi.org/10.3390/en3050899. lab and real sea as power source to drive monitoring instruments for long-term
[20] K. Tuutti, CORROSION OF STEEL IN CONCRETE, CBI ForskResearch. (1982). work, Ionics 24 (2018) 797–805, https://doi.org/10.1007/s11581-017-2251-2.
https://trid.trb.org/view.aspx?id=194388 (accessed January 10, 2022). [51] P.S. Schrader, C.E. Reimers, P. Girguis, J. Delaney, C. Doolan, M. Wolf, D. Green,
[21] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National Independent benthic microbial fuel cells powering sensors and acoustic
Association of Corrosion Engineers, 1974. communications with the MARS underwater observatory, J. Atmospheric Ocean.
[22] J.P. Broomfield, Corrosion of Steel in Concrete: Understanding, Investigation and Technol. 33 (2016) 607–617, https://doi.org/10.1175/JTECH-D-15-0102.1.
Repair, CRC Press, 2023. [52] C.E. Reimers, M. Wolf, Y. Alleau, C. Li, Benthic microbial fuel cell systems for
[23] C.L. Page, K.W.J. Treadaway, Aspects of the electrochemistry of steel in concrete, marine applications, J. Power Sources 522 (2022), 231033, https://doi.org/
Nature 297 (1982) 109–115, https://doi.org/10.1038/297109a0. 10.1016/j.jpowsour.2022.231033.
[24] R. François, S. Laurens, F. Deby, Corrosion and its Consequences for Reinforced [53] 越川義功, 北島洋二, 上野嘉之, 間宮尚, 構造物金属材の電気防食方法及び装置,
Concrete Structures, Elsevier, 2018. JP6071053B2, 2017.
[25] A. Abd El Fattah, I. Al-Duais, K. Riding, M. Thomas, Field evaluation of corrosion [54] W. Zou, C. Hou, W. Huangfu, W. Zhang, A. Zhang, Dispositif De Protection
mitigation on reinforced concrete in marine exposure conditions, Constr. Build. Cathodique Basé Sur Une Bioanode Pour Équipement De Grand Fond,
Mater. 165 (2018) 663–674, https://doi.org/10.1016/j.conbuildmat.2018.01.077. WO2021227839A1 (2021).
[26] A. Costa, J. Appleton, Case studies of concrete deterioration in a marine [55] M.G. Sohail, S. Laurens, F. Deby, J.P. Balayssac, N., Al Nuaimi, Electrochemical
environment in Portugal, Cem. Concr. Compos. 24 (2002) 169–179, https://doi. corrosion parameters for active and passive reinforcing steel in carbonated and
org/10.1016/S0958-9465(01)00037-3. sound concrete, Mater. Corros. 72 (2021) 1854–1871, https://doi.org/10.1002/
[27] U. Angst, F. Moro, M. Geiker, S. Kessler, H. Beushausen, C. Andrade, J. Lahdensivu, maco.202112569.
A. Köliö, K. Imamoto, S. von Greve-Dierfeld, M. Serdar, Corrosion of steel in [56] X. Zai, Z. Duan, W. Chen, J. Yu, Y. Fu, Pyrolyzed iron phthalocyanine-modified
carbonated concrete: mechanisms, practical experience, and research priorities – a multi-walled carbon nanotubes as composite anode in marine sediment microbial

11
J. Dubuit et al. Construction and Building Materials 403 (2023) 133180

fuel cells and its electrochemical performance, J. Ocean Univ. China 18 (2019) Netherlands, Dordrecht, 2016: pp. 107–131. https://doi.org/10.1007/978-94-017-
1395–1401, https://doi.org/10.1007/s11802-019-4024-1. 7309-6_5.
[57] Y. Fu, Q. Xu, X. Zai, Y. Liu, Z. Lu, Low electrical potential anode modified with Fe/ [66] Stratégie d’évaluation Non destructive pour la Surveillance des Ouvrages en béton
ferric oxide and its application in marine benthic microbial fuel cell with higher - SENSO, Rapport final, 2009.
voltage and power output, Appl. Surf. Sci. 289 (2014) 472–477, https://doi.org/ [67] L. Soussan, B. Erable, M.-L. Delia, A. Bergel, The open circuit potential of
10.1016/j.apsusc.2013.11.011. Geobacter sulfurreducens bioanodes depends on the electrochemical adaptation of
[58] AFNOR, NF EN 1992-1-1 “Eurocode 2 - Design of concrete structures, Part 1-1: the strain, Electrochem. Commun. 33 (2013) 35–38, https://doi.org/10.1016/j.
General rules and rules for buildings,” (2005). elecom.2013.04.013.
[59] M. Aachib, M. Mbonimpa, M. Aubertin, Measurement and prediction of the oxygen [68] M.K. Stenstrom, R.A. Poduska, The effect of dissolved oxygen concentration on
diffusion coefficient in unsaturated media, with applications to soil covers, Water nitrification, Water Res. 14 (1980) 643–649, https://doi.org/10.1016/0043-1354
Air Soil Pollut. 156 (2004) 163–193, https://doi.org/10.1023/B: (80)90122-0.
WATE.0000036803.84061.e5. [69] R.W. Okey, O.E. Albertson, Evidence for oxygen-limiting conditions during tertiary
[60] D. Garcia, S. Laurens, S. Panin, A comprehensive study of the spatial distribution of fixed-film nitrification, J. Water Pollut. Control Fed. 61 (1989) 510–519.
the galvanic protection current supplied by zinc layer anodes applied to steel- [70] D.F.A. Koch, R. Woods, The electro-oxidation of acetate on platinum at low
reinforced concrete structures, Corros. Sci. 158 (2019), 108108, https://doi.org/ potentials, Electrochim. Acta 13 (1968) 2101–2109, https://doi.org/10.1016/
10.1016/j.corsci.2019.108108. 0013-4686(68)80119-7.
[61] G.E. Archie, The electrical resistivity log as an aid in determining some reservoir [71] R. Lacroix, S.D. Silva, M. Viaplana Gaig, R. Rousseau, M.-L. Délia-Dupuy, A. Bergel,
characteristics, Trans. AIME 146 (1942) 54–62, https://doi.org/10.2118/942054- Modelling potential/current distribution in microbial electrochemical systems
G. shows how the optimal bioanode architecture depends on electrolyte conductivity,
[62] M.A. Kominz, K. Patterson, D. Odette, Lithology dependence of porosity in slope PCCP 16 (2014) 22892–22902, https://doi.org/10.1039/C4CP02177K.
and deep marine sediments, J. Sediment. Res. 81 (2011) 730–742, https://doi.org/ [72] B. Erable, N. Byrne, L. Etcheverry, W. Achouak, A. Bergel, Single medium
10.2110/jsr.2011.60. microbial fuel cell: stainless steel and graphite electrode materials select bacterial
[63] R.L. Evans, Using CSEM techniques to map the shallow section of seafloor: from the communities resulting in opposite electrocatalytic activities, Int. J. Hydrog.
coastline to the edges of the continental slope, Geophysics 72 (2) (2007) Energy. 42 (2017) 26059–26067, https://doi.org/10.1016/j.
WA105–WA116. ijhydene.2017.08.178.
[64] Lcpc, Techniques et méthodes des LPC, Recommandations provisoires : “Maîtrise [73] C. Santoro, C. Arbizzani, B. Erable, I. Ieropoulos, Microbial fuel cells: from
de la durabilité des ouvrages d’art en béton, Application de l’approche fundamentals to applications. a review, J. Power Sources 356 (2017) 225–244,
performantielle” (2010). https://doi.org/10.1016/j.jpowsour.2017.03.109.
[65] H. Beushausen, M.G. Alexander, M. Basheer, V. Baroghel-Bouny, R. d’Andréa, A. [74] AFNOR, NF P18-459 - Béton - Essai pour béton durci - Essai de porosité et de masse
Gonçalves, J. Gulikers, F. Jacobs, M. Khrapko, A.V. Monteiro, S.V. Nanukuttan, M. volumique, (2022).
Otieno, R. Polder, R. Torrent, Principles of the Performance-Based Approach for [75] H.Z. Hamdan, D.A. Salam, Sediment microbial fuel cells for bioremediation of
Concrete Durability, in: H. Beushausen, L. Fernandez Luco (Eds.), Perform.-Based pollutants and power generation: a review, Environ. Chem. Lett. (2023), https://
Specif. Control Concr. Durab. State—Art Rep. RILEM TC 230-PSC, Springer doi.org/10.1007/s10311-023-01625-y.

12