Materials and Structures (2014) 47:18391855

DOI 10.1617/s11527-013-0154-x

ORIGINAL ARTICLE

Cathodic protection with localised galvanic anodes

in slender carbonated concrete elements
Elena Redaelli Federica Lollini Luca Bertolini

Received: 30 November 2012 / Accepted: 23 July 2013 / Published online: 1 August 2013
 RILEM 2013

Abstract A combined experimental and numerical

investigation was carried out with the aim of determining whether few localised galvanic anodes per unit
length could protect the reinforcement of slender
carbonated concrete elements, exposed to atmospheric
conditions, which could not be repaired with traditional methods. Initially, the cathodic behaviour of
steel under galvanostatic polarisation was determined
on small-size specimens obtained from a real element.
A correlation of potential versus applied current was
obtained. The current distribution in slender elements
was then determined through finite elements simulations, considering various scenarios of carbonation
and humidity. Results showed that, in spite of the high
electrical resistivity of carbonated concrete, anodes
with spacing of 0.45 m are enough to protect corroding reinforcement in most exposure conditions, even
in thin parts of element. Estimated anode durations
were of the order of several years or even decades;
however, it was shown that also reinforcement in dry
(carbonated or alkaline) concrete, which does not need
to be protected, contributes to anode consumption.
Although other aspects play a role on the performance
of a cathodic protection system (such as the effectiveness of anode-encasing material and of electrical

E. Redaelli (&)  F. Lollini  L. Bertolini

Department of Chemistry, Materials and Chemical
Engineering Giulio Natta, Politecnico di Milano,
Piazza Leonardo da Vinci 32, 20133 Milan, Italy
e-mail: [email protected]

connection to reinforcement), the results obtained are

supportive of a repair strategy based on the use of
localised galvanic anodes and can be generalised to
slender elements exposed to atmospheric conditions
suffering carbonation induced corrosion.
Keywords Carbonation  Cathodic protection 
Concrete  Current distribution  Galvanic anode 
Localised anode

1 Introduction
The selection of the repair method for concrete
structures damaged by carbonation-induced corrosion
of reinforcement is a challenging task in those cases
where the need to preserve the original surface, shape
and materials makes a conventional intervention
impossible or unacceptable. This is the case, for
instance, of buildings and monuments of modern
architecture, for which issues related to conservation
are often of primary importance (as well as those
related to safety or costs) in the selection of the repair
method [10, 14, 15]. In other circumstances, the need
to limit the alteration of the structure stems from the
necessity to reduce the invasiveness of the intervention, for instance to keep the building in service during
the repair operations or to reduce the generation of
noise, dust and waste materials connected with
concrete removal.

1840

To this regard, electrochemical techniques may be a

valid alternative compared to conventional repair
because they allow to preserve non-protective carbonated concrete, provided it is not damaged (cracked
or spalled) by reinforcement corrosion yet [31].
Amongst these, cathodic protection (CP), that in the
past was mainly used for chloride-contaminated
concrete, is nowadays an option even for carbonation-induced corrosion thanks to advantages connected with the possibility of leaving carbonated
concrete in place and, so, minimising its replacement
[4, 13, 19, 20, 33].
Further advantages in terms of reduced invasiveness of the intervention may be achieved if small
localised anodes could be used instead of traditional
surface-distributed anode systems [3]. In particular,
localised galvanic anodes, that could be simply
connected to the steel reinforcement (ideally without
the need of electrical cabling), allow minimising the
repair operations and avoid or limit the subsequent
permanent monitoring system required by impressed
current CP. Such anodes could be shaped as small
cylinders to be inserted inside concrete or stripes to be
placed on the concrete surface. Galvanic anodes have
to be in contact with a proper electrolyte that prevents
metal passivation and allows current circulation [12].
They do not require an external current generator and,
unlike inert anodes used in impressed current systems,
accidental contact with steel reinforcement does not
impair the functioning of the CP system. Galvanic
anodes are self-regulating since the current they
supply is determined by the electrochemical condition
of steel and anode material (the latter is usually zinc
activated by a proper embodying material) and by the
electrical resistivity of concrete; both may change in
time as a consequence of variations in humidity,
carbonation, etc. [11, 27]. Advantages connected with
the use of localised anodes, however, may be partially
compensated by their intrinsic limitation related to
current distribution. In elements subject to carbonation, current distribution issues may be particularly
critical due to increased electrical resistivity of
carbonated concrete compared to alkaline concrete
[23].
The study that will be presented in this paper started
from the need to define a method for the repair of the
reinforced concrete newels on the facades of a
building exposed to urban environment [6]. The
elements are characterised by a slender, rather

Materials and Structures (2014) 47:18391855

complex shape, as shown in Fig. 1, and their function

is mainly decorative. Almost 2000 elements are
present on the facades of the building; they are not
identical due to slight differences in geometry. After
50 years since the construction, some elements show
the presence of cracks and local spalls of concrete
cover. A preliminary condition evaluation revealed
that damage is due to carbonation-induced corrosion
of reinforcement; many elements are expected to be in
a condition of high risk of falling of concrete debris,
which is unacceptable for safety reasons. The repair
method is required to maximise the preservation of the
original material, shape and surface of elements and, at
the same time, to fulfil adequate durability due to the
difficulty in accessing the facades for future maintenance. CP with localised galvanic anodes is considered as a possible repair option and a preliminary
design of the CP system is required, including an
estimation of the number of anodes that is necessary to
protect the reinforcement. To this purpose, the current
distribution needs to be evaluated, and this has been
done through different approaches.
Initially, the potential distribution that could be
achieved with a localised galvanic anode was tested on
a real-size specimen. One of the elements was taken
from the building and was available for experimental
tests in the laboratory. Results from these tests are
presented in [28]. They showed that a single galvanic
zinc anode positioned at one end of the specimen could
produce a cathodic polarisation higher than 100 mV
on steel reinforcement over a distance of 0.3 m both in
condition of water-saturation (where the current
supplied by the anode was 500 lA) and in condition
with 60 % RH (where the current supplied by the
anode was 80 lA). Hence, assuming the 100-mV
decay criterion the height of protection of the anode
was around 0.6 m, which means that protection of
reinforcement on a real full-scale element could be
achieved placing an anode every 0.6 m (i.e. 1.7 anodes
every meter) along the height in the front part of the
element. In the case analysed, this amount of anode per
length of element could be considered acceptable.
These encouraging results were, then, further
investigated by means of numerical models, that will
be presented in this paper, with the aim of optimising
the anodic configuration. First, small specimens were
obtained from the real element and were used to
characterise concrete and steel in relation to corrosion
and to provide input parameters for numerical models.

Materials and Structures (2014) 47:18391855

1841

impressed current. Test duration was of the order of

few weeks to few months.
The specimens had a simple prismatic geometry (as
it is shown in Fig. 2) with a height of 5570 mm. A
mixed metal oxide (MMO) activated titanium mesh
(of the type commonly used for impressed current CP)
embedded in a layer of mortar was applied on a lateral
surface of each specimen, parallel to rebars, in order to
work as counter-electrode in electrochemical tests.
The specimens were exposed to different conditions of
humidity and were subjected to galvanostatic cathodic
polarisation tests, with the aim of investigating the
correlation between applied current and steel potential
under cathodic polarisation. Free corrosion potential
and linear polarisation resistance of steel were measured on a reference specimen. The specimens and the
sequence of tests are reported in Table 1. All potentials were measured versus an external saturated
calomel electrode (SCE) reference electrode placed
on the top surface of the specimen, as indicated in
Fig. 2.
During the application of current, 4-h depolarisation tests were carried out in order to determine the on
potential (Eon, measured during application of the
current), the instant-off potential (Eoff, measured
within 1 s from switching off the current) and the
4-h potential (E4h, measured after 4 h from switching
Fig. 1 Geometry of the newel (dimensions in mm)

Then, calculations were carried out considering

different scenarios by varying conditions and parameters that could not be controlled during experimental
real-size tests, such as electrochemical conditions of
steel, number and position of anodes, and concrete
properties in terms of moisture content and carbonation. The results obtained will be discussed in relation
with the performance of two anodic configurations,
with anode spacing of either 0.9 m or 0.45 m, in
providing protection to corroding slender elements.

2 Characterisation of corrosion parameters

Several small specimens containing sections of two
longitudinal bars were obtained by cutting the newel
and were used to characterise the behaviour of steel in
alkaline and carbonated concrete, both in free corrosion conditions and under cathodic polarisation with

Fig. 2 Schematic geometry of specimens used in galvanostatic

CP tests

1842
Table 1 Summary of
results from tests on small
samples in different
exposure conditions:
applied cathodic current
density, average instant off
potential (Eoff) and average
feeding voltage (DV)

Materials and Structures (2014) 47:18391855

Exposure
condition

P-1 (passive)

30

50 % RH

10a

-0.427

14.8 (3.01/26.7)

10

Immersion

10

-0.456

1.039
0.791

A-1 (active)

A-2 (active)

A-3 (active)

A-4 (active)
a

Nominal value that could

not be supplied due to
saturation of generator

Average free corrosion

rate = 5.46 mA/m2

DV (min/max) (V)

Exposure
period (day)

P-2 (passive)

A-5 (active)

Current
density
(mA/m2)

Eoff
(V/SCE)

Specimen
(steel
condition)

40

Immersion

-0.463

20

Immersion

-0.377

30

Immersion

10

-0.534

1.142
1.025

10

Immersion

-0.470

20

Immersion

100

-0.711

2.610

20

Immersion

10

-0.446

1.124

20

Immersion

-0.330

28

Immersion

-0.613

1.127

10

50 % RH

-0.553

1.894

27

95 % RH

-0.575

1.566

28

Immersion

10

-0.743

1.321

10

50 % RH

10

-0.776

3.749

27
14

95 % RH
Immersion

10
10

-0.715
-0.628

2.345
1.165

14

Immersion

100

-1.080

10

50 % RH

100a

-2.240

2.39

27

95 % RH

100

-0.959

5.3

14

Immersion

100

-1.160

2.457

14

Immersion

10

-1.026

1.620

25.27 (18.74/26.82)

10

50 % RH

10

-0.770

3.885

27

95 % RH

10

-0.729

1.80

40

Immersion

-0.616b

off the current). The difference between E4h and Eoff

gives the 4-h potential decay.
Figure 3 shows, as an example, the results of
potential measurements obtained on specimen A-1,
which was subjected to a constant cathodic current of
5 mA/m2 in various environments. In this specimen
the steel was active since it was in contact with
carbonated concrete. Eon, Eoff and E4h potentials are
reported together with the applied cathodic current
density (cd). During the initial period of immersion,
Eon and Eoff potentials were -0.62 and -0.61 V/SCE,
respectively; the small difference between them
reflects a low ohmic drop contribution, due to the
relatively low value of applied current and to the
condition of immersion that makes concrete electrical
resistivity low. E4h potential was -0.58 V/SCE and
the 4 h decay was around 40 mV. When the specimen
was exposed to a dry environment (50 % RH), it
progressively dried out. Eon decreased to values as low

as -1.55 V/SCE, while Eoff decreased to -0.64

V/SCE and its average value during exposure to dry
environment was -0.55 V/SCE. E4h was around
-0.42 V/SCE and 4 h decay was about 135 mV.
Finally, after exposure to 95 % RH, both Eon and Eoff
became more positive, and the 4 h decay was around
80 mV.
The results obtained from all the specimens are
summarised in Table 1 in terms of instant-off potential
(Eoff) and feeding voltage (DV); average values are
reported, calculated from all values in each condition.
Eoff values were then plotted, together with literature
data, versus the cathodic cd, as reported in Fig. 4, with
the aim of investigating the cathodic behaviour of steel
in alkaline and carbonated concrete and, specifically,
highlighting possible differences according to the
conditions of carbonation of concrete. Although the
scatter of the results, no systematic trend was observed
in terms of potential versus cd values in carbonated

Materials and Structures (2014) 47:18391855

Fig. 3 Example of results of galvanostatic tests on specimen

A-1. On (Eon), instant-off (Eoff) and 4-h (E4h) potentials of steel
reinforcement and cathodic cd are shown

and alkaline concrete. Concrete carbonation did not

play a primary role in the determination of the
cathodic polarisation curve.
Also four plain concrete specimens without reinforcement (cubic geometry with side of approximately
45 mm) were cut from the newel and were used to
determine the humidity content and the electrical
resistivity in different conditions of humidity. After
initial immersion, they were exposed in sequence to
90, 50 and 95 % relative humidity (RH) at temperature

Fig. 4 Instant off values of steel potential as a function of the

applied cathodic cd, measured on steel in alkaline and
carbonated concrete in different exposure conditions (circles
refer to data from [5])

1843

of 20 C. The electrical conductance, C, was measured

between two parallel surfaces with a conductimeter;
concrete resistivity, q, was obtained through the
relation: q = A/(C L), where A is the surface and
L the thickness. During exposure to conditions of
immersion, 95 and 90 % RH, the moisture contents in
concrete measured at the end of each exposure period
were around 4, 1.5 and 1 %, respectively, and
corresponding values of electrical resistivity were
40770, 3507550 and 1107425 X m. Values of
concrete resistivity of 3,900710,000 X m were measured during exposure to 50 % RH. These results are
representative of alkaline concrete, since it was not
possible to obtain wholly carbonated specimens due to
the relatively low depth of carbonation. Figure 5
shows the correlation between moisture content and
resistivity measured on all the specimens. Values of
resistivity higher than 1,000 X m, which is usually
considered as a threshold value above which the
propagation of carbonation induced corrosion can be
neglected [1, 23], were obtained for moisture contents
lower than 0.5 %.

3 Current distribution
After characterising the corrosion conditions and the
cathodic behaviour of steel in concrete on small
samples, numerical simulations were performed with
the aim of determining the distribution of current in a

Fig. 5 Electrical resistivity of alkaline concrete as a function of

moisture content measured on four plain specimens exposed to
different conditions of RH

1844

newel subjected to CP with localised galvanic anodes.

The commercial package COMSOL was used. The
soundness of numerical solutions as well as nondependence on mesh size were checked for all models.
3.1 Description of element
The real newel is a precast concrete element reinforced with steel bars and wires. The shape is quite
complex, thinner in the central part compared to the
top and bottom ends (Fig. 1). The front and rear parts
are respectively 2.62 and 2.09 m high and 235 and
165 mm deep. The thickness varies between 60 and
120 mm; the lower thickness is in the centre of the
front part. The front part is exposed outside; the rear
part is embedded between adjacent elements such as
panels and pillars and is not accessible.
Thanks to cutting of the newel in the top and bottom
parts to obtain previously described specimens (Sect.
2), the position of the rebars and the carbonation depth
were measured directly on several freshly-cut sections.
The reinforcement consists of two adjacent longitudinal bars (/ = 6 mm) in the front part of the element,
and two (/ = 8 mm) in the rear part, anchored to each
other with several twisted and bent steel wires (/ =
5 mm) placed at various heights (stirrups). In the top
and bottom front parts, the longitudinal bars are bent up
over a length of approximately 300 mm. All reinforcement is unribbed. Neither longitudinal bars nor wires
are exactly parallel to the surface, giving rise to
variability in concrete cover thickness.
Carbonation depths were measured with phenolphthalein tests and ranged between 17 and 30 mm on
all sides; these values were somewhat lower compared
to those expected after condition evaluation [6],
indicating that in the newel most of reinforcement is
still passive, being in contact with alkaline concrete,
while many newels in the building are carbonated up
to the depth of the bars. Only longitudinal rebars in the
thinner part of the newel are in contact with carbonated concrete.
3.2 Geometry of the model
In spite of slight differences in the shape and
variability of the position of reinforcement, the
geometry determined on the newel was considered
representative of most elements, and, therefore, it was
used to define the geometry of the domain that was

Materials and Structures (2014) 47:18391855

considered to simulate current distribution (Fig. 6a).

The 3D domain reflected the geometry evaluated on
the newel, with minor simplifications (e.g. the position
of the longitudinal reinforcement and of the anchoring
wires). The concrete cover thickness on the front
surface was set to 27 mm.
3.3 Scenarios
Various scenarios in terms of moisture content and
carbonation of concrete and two different anodic
configurations were considered.
Regarding humidity conditions, it was assumed that
in the front part of the element, which is exposed to the
atmosphere, concrete could be dry (moisture content
in concrete around 0.5 %), wet (around 1.5 %) or
saturated (around 4 %), whilst in the rear part, which is
in contact with adjacent elements, concrete was
always considered dry. Regarding carbonation, it
was assumed that the element could be fully carbonated, fully alkaline, or alkaline in the rear part and
carbonated in the front part. The sharp division
between front and rear, although not fully representative of real conditions, stemmed from the need to
simplify the geometry of the subdomains.
The anodic configurations were defined in terms of
number and position of anodes, consisting of small
rods made of zinc and placed in the front part
transversally to the lateral surface, at different heights,
as indicated in Fig. 6b, c. The rod has a width of
30 mm, a height of 5 mm and a length equal to the
thickness of the newel. Experimental results on realsize specimen clearly showed that more than one
anode is necessary to protect the element. Since in
very aggressive (water-saturated) conditions the
required anode spacing was 0.6 m [28], two anodic
configurations were defined, with anode spacing of
0.9 m (i.e. three anodes per element) and of 0.45 m
(i.e. six anodes per element). Configurations with
anode spacing lower than 0.45 m, that require more
than six anodes per element, were not considered
realistic and, so, were not taken into account.
Selected combinations of the above-mentioned conditions of carbonation and moisture and anode configurations gave rise to the various scenarios considered,
that are reported in Table 2. These scenarios were
implemented by values of resistivity of concrete and
boundary conditions of steel and anode surfaces, as it
will be explained in the following section.

Materials and Structures (2014) 47:18391855

1845

Fig. 6 3D geometry of
element with reinforcement
considered in numerical
simulations (a),
configuration with three
anodes (b, bars not shown)
and configuration with six
anodes (c, bars not shown)

3.4 Input parameters

The numerical model requires the knowledge of the
parameters that describe the electrochemical behaviour of steel and anode, and the electrical properties of
the concrete, in the different scenarios. The parameters
were selected both on the basis of the results obtained
with experimental tests described in Sect. 2 and from
literature values [26].

The electrochemical behaviour of steel was

described through a ButlerVolmer type polarisation
curve, in the form of:
i icorr  10

EEcorr
bA

 icorr  10

EEcorr
bC

where icorr is the free corrosion cd (in mA/m2), Ecorr

the free corrosion potential (in V), and bA and bC the
slopes of the anodic and cathodic branches of the curve

1846
Table 2 Scenarios
considered in the numerical
simulations
(sat. = saturated)

Materials and Structures (2014) 47:18391855

Reinforcement

Completely passive steel

Active steel in the front part
Completely active steel

(in V/dec). These four parameters were chosen as a

function of the properties of concrete with which steel
was in contact, in terms of carbonation and humidity
content (expressed as resistivity).
A single cathodic curve was chosen for all cases,
with a slope of 0.2 V/dec (without any limiting current
of oxygen diffusion), in accordance with results
obtained on small specimens (Sect. 2). For active
steel, after selecting a given combination of free
corrosion potential and corrosion rate (lying on the
cathodic curve), the slope of the anodic branch of the
curve was determined by intersection with the point
with potential of -1 V/SCE and cd of 0.01 mA/m2, i.e.
roughly the equilibrium potential and exchange cd of
the reaction of iron oxidation, as it is shown in Fig. 7.
This is consistent with the anodic resistive control that
characterises the corrosion of steel in carbonated
concrete [9]. Specifically, for active steel in carbonated
and wet concrete a free corrosion potential of -0.36
V/SCE and a free corrosion cd of 2 mA/m2 were
selected, that resulted in an anodic slope of 0.28 V/dec,
while for active steel in carbonated and dry concrete a
potential of -0.1 V/SCE and cd of 0.1 mA/m2 were
selected, and the resulting anodic slope was 1 V/dec. In
carbonated and water-saturated concrete the corrosion
cd was 5 mA/m2 and the resulting cathodic slope was
0.21 V/SCE. For passive steel, the anodic slope was set
to 1,000 V/dec, giving a vertical anodic branch that
was intended to represent the condition of passivity.
All the parameters are summarised in Table 3.
For the boundary conditions on the galvanic anode,
a constant potential value was assumed, equal to
-1.05 V/SCE, which is a typical potential value for
zinc in contact with an alkaline environment [30].1
1

Galvanic anodes made of zinc are encased in a highly alkaline

electrolyte in order to keep the metal active. The potential

Concrete in
the rear part

Concrete in the
front part

Anode spacing
0.90 m

Alkaline, dry

Alkaline, dry

Alkaline, dry

Alkaline, wet

Alkaline, dry
Carbonated, dry

Carbonated, wet
Carbonated, dry

9
9

Carbonated, dry

Carbonated, wet

Carbonated, dry

Carbonated, sat.

0.45 m
9
9
9
9
9
9

Fig. 7 Determination of anodic slopes of steel in carbonated

concrete in dry, wet and water-saturated conditions according to
the cathodic curve (grey line) and to the free corrosion
conditions (circles). Slopes are in V/dec

Finally, the electrical resistivity of concrete was

specified as a function of concrete carbonation and
moisture content. According to literature data the
resistivity of carbonated concrete was assumed to be
410 times greater than that of alkaline concrete in the
same condition of exposure [2, 7, 16, 17, 21]. So, the
resistivity of alkaline concrete was set to 250 X m in
wet condition and 400 X m in dry condition; the
resistivity of carbonated concrete was set to
1,000 X m in wet condition, 10,000 X m in dry
condition and 200 X m in water-saturated condition
(Table 3).

Footnote 1 continued
chosen for zinc anodes reflects the contact with such alkaline
environment.

Materials and Structures (2014) 47:18391855

1847

Table 3 Input parameters used in numerical simulations in different conditions of carbonation and humidity: for steel, parameters
used in Eq. 1; for concrete, electrical resistivity q
Concrete

Condition

Label

Steel

Concrete

Ecorr
(V/SCE)

icorr
(mA/m2)

bA
(V/dec)

bC
(V/dec)

q
(X m)

Carbonated

Wet

CW

-0.36

0.28

0.2

Carbonated

Saturated

CS

-0.45

0.21

0.2

1,000
200

Carbonated

Dry

CD

-0.1

0.1

0.2

10,000

Alkaline

Wet

AW

-0.15

0.2

1,000

0.2

250

Alkaline

Dry

AD

-0.15

0.2

1,000

0.2

400

3.5 Results
Figures 8 and 9 show examples of the distributions of
potential (a) and cd (b) along the height of reinforcement in the front part and in the rear part of a newel
protected with three and six anodes (bent front
indicates the bent part of longitudinal bars in the front
part, as depicted in Fig. 6a; the potential of stirrups is
not shown). These results refer to the scenario in which
the newel is fully carbonated and it is wet in the front
part and dry in the rear part.
The longitudinal rebars in the front part of the
newel show an uneven potential distribution: in the

presence of three anodes they are cathodically polarised only in the vicinity of the anodes, where their
potential is about -0.5 V/SCE, while between the
anodes their potential is close to -0.36 V/SCE, i.e. the
free corrosion value (dashed black line); bent reinforcement shows somewhat lower values of potential
due to its closeness to the anode. Also in the case with
six anodes the potential is -0.5 V/SCE in the vicinity
of the anodes, while it is -0.4 V/SCE far from the
anodes. The longitudinal rebars in the rear part, that
are farther from the anodes compared to those in the
front, show a more uniform potential distribution, with
values around -0.25 V/SCE close to the anodes and

Fig. 8 Distribution of potential (a) and cd (b) on steel reinforcement connected with three galvanic anodes placed at heights of 0.43,
1.31 and 2.29 m. Front part: carbonated and wet concrete; rear part: carbonated and dry concrete

1848

Materials and Structures (2014) 47:18391855

Fig. 9 Distribution of
potential (a) and cd (b) on
steel reinforcement
connected with six galvanic
anodes placed at heights of
0.22, 0.66, 1.1, 1.54, 1.98
and 2.42 m. Front part:
carbonated and wet
concrete; rear part:
carbonated and dry concrete

-0.2 V/SCE in between anodes in the configuration

with three anodes and almost constant with values of
-0.337-0.3 V/SCE in the configuration with six
anodes (for these rebars the free corrosion potential is
-0.1 V/SCE as indicated by the dashed grey line).
These rebars do not require to be protected from
corrosion, since they are in contact with dry concrete,
anyway they have to be considered since they receive a
current due to the electrical connection with the bars in
the front part of the element.
The cd on the front bars approaches values of
10 mA/m2 in the vicinity of the anodes in both
configurations, while far from the anodes it is of the

order of 0.1 mA/m2 or even less when three anodes are

used and 1 mA/m2 when six anodes are used; the
reinforcement in the rear part receives a much more
even cd in the range 0.270.4 mA/m2 in both cases.
The results of all simulations are summarised in
Tables 4 and 5. As a general trend, the most negative
potential and the maximum cd are always obtained on
reinforcement portions close to the anodes. Moreover,
potential and cd distributions on the rear reinforcement are more uniform compared to those on the front
reinforcement. Decreasing anode spacing from 0.90 to
0.45 m considerably affects the most positive value of
steel potential on the front reinforcement (which

Table 4 Minimum and maximum values of steel potential on front, bent front, rear reinforcement and stirrups in the various
scenarios
Spacing (m)

0.90

0.45

Front:
Rear:
Reinf.

AD
AW
AD
AW
Emin/Emax (mV/SCE)

CW
AD

CS
AD

CD
CD

CW
CD

CS
CD

Front

-521/-357

-347/-117

-522/-359

Bent front

-599/-364

-430/-127

-600/-365

Rear

-320/-245

-204/-119

-273/-203

Stirrups

-461/-259

-275/-121

-463/-220

Front

-412/-249

-612/-461

-554/-416

-655/-513

-350/-183

-527/-394

-695/-466

Bent front

-489/-265

-684/-484

-630/-435

-717/-531

-429/-198

-618/-411

-736/-498

Rear

-249/-224

-330/-304

-427/-398

-338/-313

-185/-160

-333/-287

-338/-294

Stirrups

-375/-236

-577/-324

-520/-412

-313/-170

-494/-302

-642/-311

Materials and Structures (2014) 47:18391855

1849

Table 5 Minimum and maximum values of cathodic current density on front, bent front, rear reinforcement and stirrups in the
various scenarios
Spacing (m)

Front:
Rear:
Reinf.

AD
AW
AD
AD
imin/imax (mA/m2)

CW
AD

CS
AD

CD
CD

CW
CD

CS
CD

0.90

Front

0/9.1

0/1.2

0/9.2

Bent front

0.1/29.0

0/4.3

0.1/29.1

Rear

0.3/0.9

0/0.2

0/0.5

Stirrups

0/4.3

0/0.5

0/4.5

Front

0.2/2.9

3.3/30.5

0.7/8.9

3.6/33.3

0.1/1.3

0.7/9.5

4.0/26.1

Bent front
Rear

0.3/13.8
0.2/0.3

5.1/123.7
0.6/5.0

1.2/37.6
0.2/0.4

5.6/137.0
0.7/1.3

0.1/6.2
0.07/0.14

1.2/35.9
0.2/0.3

9.2/134.2
0.3/0.6

Stirrups

0.2/1.8

0.6/20.2

0.3/6.0

0.1/0.8

0.2/6.5

0.2/17.2

0.45

decreases), while the effect on the most negative value

is negligible.
Table 6 shows the current supplied by each anode
and the total anodic current; in each scenario differences in the current supplied by each single anode are
mainly due to geometric differences inside the
element, related to anode dimension and position. In
the cases where calculations for both anodic configurations are available, the current supplied by each
anode is slightly lower in the configuration with anode
spacing of 0.45 m, and so the total current supplied by
six anodes is slightly less than twice the total current
supplied by three anodes. In general, for a given
anodic configuration, when the concrete in the front
part is dry, the current supplied by the anodes is lower
compared to wet or water-saturated concrete. With
anode spacing of 0.45 m, the highest current is
supplied when the concrete in the front part is watersaturated, and it is of the order of 3,0003,150 lA. A
high anodic current (2750 lA) is also supplied when
the element is fully alkaline, and wet in the front part,
whilst if it is fully dry the current is much lower
(250 lA). When the element is completely carbonated, the total current is around 800 lA when the
concrete in the front is wet and 110 lA when the
concrete in the front is dry.

4 Discussion of results
The performance of a galvanic CP system with
localised anodes needs to be investigated in relation
with two main aspects: the distribution of potential,
that determines the protection to steel reinforcement,

and the consumption of anodes, that determines their

working life. The numerical simulations allow to
determine current and potential distributions in the
presence of galvanic anodes and, therefore, to evaluate
the effectiveness of the two anodic configurations in
protecting steel reinforcement from corrosion in slender elements. In the following, protection conditions
will be addressed with regard to the reinforcement in
the front part of the element; the reinforcement in the
rear part is characterised by a low corrosion risk, since
it is either passive (being in contact with alkaline
concrete), or active (being in contact with carbonated
concrete) in conditions of lack of humidity that prevent
corrosion propagation. The presence of the rear
reinforcement, however, cannot be completely
neglected since it affects the current the anodes supply,
and, as a consequence, their performance in time.
4.1 Protection of reinforcement
In Sect. 3 results of numerical simulations were
presented in terms of distribution of potential and cd
on the reinforcement. However, the effectiveness of
the protection provided by a cathodic current is not
directly correlated with these parameters: measurements carried out on the reinforcement of a real
element showed that, for a given applied cathodic
current, steel potential strongly depends on concrete
moisture content, and it is much more negative in wet
concrete compared to dry concrete, although not
indicating a higher protection level [28]. So, the results
were analysed in terms of cathodic polarisation, i.e.
the difference between the free corrosion potential and
the potential in the presence of the anodes, and in

1850

Materials and Structures (2014) 47:18391855

Table 6 Current supplied by each anode and total anodic current (I, in lA), theoretical consumption rate (r, in g/year) and duration
of the anode (t, in year), obtained from numerical simulations in the various scenarios

Spacing (m)

Front:
Rear:
H (m)

0.90

0.43

148

20

1.31

100

13

98

2.29

206

28

205

Total

454

61

448

0.22

64

681

193

776

28

204

745

0.66
1.10

31
31

336
345

94
97

385
395

13
14

97
100

372
377

1.54

31

342

96

391

14

99

371

1.98

41

447

126

510

18

131

477

2.42

57

604

172

686

25

182

658

Total

255

2,756

779

3,143

112

814

3,000

Spacing (m)

H (m)

r (g/year)

0.90

0.43

1.58

0.21

1.55

1.31

1.07

0.14

1.05

2.29

2.2

0.3

2.19

0.22

0.68

7.27

2.06

8.28

0.3

2.18

7.96

0.66

0.33

3.59

1.01

4.11

0.14

1.04

3.97

1.10

0.34

3.69

1.03

4.22

0.15

1.07

4.03

1.54

0.33

3.65

1.03

4.18

0.14

1.06

3.96

1.98

0.44

4.78

1.35

5.44

0.19

1.4

5.09

2.42

0.61

6.45

1.83

7.33

0.27

1.94

7.03

Spacing (m)

H (m)

t (year)

0.90

0.43

81

603

83

1.31

83

621

84

2.29

58

428

59

0.22

189

18

62

16

428

59

16

0.66

271

25

88

22

622

85

22

1.10

264

24

86

21

606

83

22

1.54

266

24

86

21

611

83

22

1.98

291

27

95

24

668

92

25

2.42

212

20

70

18

479

66

18

0.45

0.45

0.45

AD
AD
I (lA)

AW
AD

CW
AD

CS
AD

CD
CD

CW
CD

CS
CD

145

H is the position of the anodes (height in m in the front part). Labels are defined in Table 3

terms of anodic activity of steel, i.e. the cd evaluated

on the anodic polarisation curve at the potential value
reached by the steel (Fig. 10). The cathodic polarisation has a practical relevance, since it is related to the
cathodic depolarisation usually measured in CP systems (as previously proposed by [22]); the anodic
activity is a measure of the corrosion rate of steel

under polarisation, and can be useful to estimate the

expected delay of cracking of concrete and, hence,
increase of service life after the intervention.
As an example, Fig. 11 shows the distribution of
cathodic polarisation on the front reinforcement in the
case of fully carbonated element, wet in the front part;
the bar chart on the right indicates areas with cathodic

Materials and Structures (2014) 47:18391855

Fig. 10 Determination of cathodic polarisation and anodic

activity of steel

1851

Fig. 12 Intervals of cathodic polarisation of steel bars in the

front part of the newel in the various scenarios considered.
Circles indicate the height of anode position (placed in the front
part of the element)

Fig. 13 Intervals of anodic activity of steel bars in the front part

of the newel in the various scenarios considered. Circles
indicate the height of anode position (placed in the front part of
the element)

Fig. 11 Example of determination of ranges of cathodic

polarisation on the reinforcement in the front part of a
completely carbonated element (wet in the front, dry in the
rear). The bar on the right shows heights where the cathodic
polarisation is lower than 50 mV (black), between 50 and
100 mV (grey) and higher than 100 mV (white). Circles
indicate anodes

polarisation higher than 100 mV (white), lower than

50 mV (black) and intermediate (grey). Close to each
anode (indicated by a circle) there is a portion of

reinforcement where the cathodic polarisation is

higher than 100 mV; each of these portions is
surrounded by an area where the cathodic polarisation
is between 50 and 100 mV and, finally, in positions far
from the anodes, the polarisation is lower than 50 mV.
Although protection conditions on real structures are
usually evaluated only through cathodic polarisation,
simulations also allow an estimation of the actual
anodic activity of steel along the protected bar. The
corrosion rate was evaluated in a similar way and it was
classified as lower than 1 mA/m2 (negligible value),
between 1 and 2 mA/m2 (intermediate) and higher than
2 mA/m2 (high corrosion rate). The results obtained
from these classifications are shown in Figs. 12 and 13

1852

that depict intervals of cathodic polarisation and

corrosion rate of steel, respectively, in all the scenarios.
The configuration with anode spacing of 0.90 m
cannot be considered adequate to protect corroding
elements. As a matter of fact, in the scenarios where
concrete is carbonated and wet, the height with a
cathodic polarisation higher than 100 mV around each
anode is 0.0570.1 m, which means an overall height
of 10 % with respect to the total length of the
reinforcement. The percentage where the cathodic
polarisation is higher than 50 mV is 25 %. Considering the corrosion rate, values lower than 1 mA/m2
characterise 15 % of reinforcement, while values
lower than 2 mA/m2 characterise 8090 % of reinforcement. So, both criteria of cathodic polarisation
and corrosion rate indicate an incomplete protection of
actively corroding steel reinforcement.
When anode spacing is reduced to 0.45 m, the
amount of reinforcement with cathodic polarisation
higher than 100 mV increases in each scenario, not
only because the number of anodes is higher, but also
because their reduced spacing is such that reinforcement at intermediate position between two anodes
receives current from both. For instance, considering
the case when the element is carbonated and wet in the
front and alkaline and dry in the rear, the height with
cathodic polarisation higher than 100 mV around a
single anode is 0.170.2 m. This can be attributed to a
more even distribution of current (that can also be
seen, for instance, comparing Figs. 8 and 9), and not to
a higher amount of current supplied by each anode. In
fact, as already mentioned, the current supplied by
each anode is slightly lower when there are six anodes
instead of three (Table 6).
Neglecting scenarios in which concrete is fully dry
(either alkaline or carbonated), where cathodic polarisation is always higher than 50 mV and corrosion rate
is always lower than 1 mA/m2, it can be observed that
if the carbonated element is wet in the front there are
regions with cathodic polarisation lower than 50 mV,
which are however characterised by corrosion rate
lower than 2 mA/m2. However, when concrete in the
front is temporarily water-saturated, the corrosion rate
reaches values well higher than 2 mA/m2 in spite of a
relatively favourable distribution of cathodic polarisation that is mainly higher than 50 mV.
Comparing the conditions of water-saturated concrete and wet concrete, two contrasting factors clearly
emerge: on one hand, the promotion of current

Materials and Structures (2014) 47:18391855

circulation in saturated concrete due to low electrical

resistivity, and, on the other hand, the hindrance of
protection conditions due to the high corrosion rate of
steel in saturated concrete. The former results in a
higher percentage of steel polarised more than 100 mV
in the more severe condition of water-saturation; the
latter results in a higher rate of propagation of
corrosion, due to the higher corrosion rate of steel.
The condition of concrete in the rear part affects the
distribution of potential in the front part as a consequence of variation in electrical resistivity: for
instance, in the scenarios where concrete in the front
part is carbonated and saturated, the cathodic polarisation is always higher than 50 mV if concrete in the
rear part is alkaline (resistivity of 400 X m), while it
shows areas lower than 50 mV if concrete in the rear
part is carbonated (resistivity of 10,000 X m). Conversely, the condition of steel bars in the rear part
(active or passive) only affects the current circulating
between them and the anode.
The protection conditions depicted in Figs. 12 and 13
are conservative since the evaluation of both cathodic
polarisation and corrosion rate according to Fig. 10
neglects possible long-term effects of CP such as the
repassivation of steel due to the alkalinity produced by
the applied current, which is expected to be beneficial
even for relatively low applied current [5, 25]. On the
short term, such changes have been observed as well [18].
Moreover, in accordance with the polarisation curves
reported in Fig. 7, values of cathodic polarisation equal
to or higher than 100 mV guarantee that the corrosion
rate of steel is below 2 mA/m2 even in very aggressive
(water-saturated and carbonated) conditions. In less
aggressive environments, such as in carbonated and wet
concrete, even a cathodic polarisation of 50 mV is
representative of protection of steel. To this regard it
should be noted that although the standard on CP of steel
in concrete does not explicitly require a depolarisation
higher than 100 mV for galvanic anodes, it warns that
the corrosion risk should be assessed [8].
Figure 12 also shows that should a condition of
cathodic polarisation higher than 100 mV be required
on the entire front reinforcement in all the exposure
conditions considered, the anode spacing should be
lower than 0.45 m. However, this requirement could
be relaxed considering that:

the height of protection obtained from numerical

simulations is more conservative compared to that

Materials and Structures (2014) 47:18391855

1853

measured on a real-size element, owing to several

factors as the presence of an alkaline core with lower
resistivity that enhances current distribution (this was
verified through numerical simulations carried out on
the same real-size element, not presented here),
the most severe conditions (water-saturation in the
front part of a completely carbonated element) are
unlikely to occur on real elements,
even wet conditions are reached only for limited
(rainy) periods,
the long-term beneficial effects of the cathodic
current, such as the production of alkalinity, are
neglected in the numerical simulations (that only
consider static short-term scenarios).

If so, given the necessity to limit the number of

anodes, the configuration with six anodes can be
considered suitable to protect the steel reinforcement
from carbonation-induced corrosion, provided possible differences in exposure conditions and extent of
carbonation are properly taken into account.
4.2 Consumption of anodes
In addition to the protection of reinforcement from
corrosion, an effective CP system should be able to
work properly for the intended service life, typically
1020 years, during which the anodic system has to
supply sufficient current [24]. Therefore, also the
behaviour in time of the anodic system has to be taken
into account, in particular for galvanic anodes whose
working life may be limited by their consumption [29,
32]. Table 6 reports the theoretical consumption rate
of each anode r (obtained by Faradays law supposing
that the anodic current is constant in time and
assuming 100 % anode efficiency) and its theoretical
Table 7 Cathodic current
(I, in lA) received by
reinforcement in the front
part of the element, in the
rear part and on the stirrups

working life t (obtained as the ratio between the mass

of each anode and its consumption rate).
Referring to the configuration with anode spacing
of 0.45 m, the lowest durations are around
1618 years and are obtained in carbonated and
water-saturated concrete. Such durations could be
increased, to some extent, increasing the anode mass:
given the small anode size, its volume can be increased
leaving current distribution practically unchanged.
When concrete is carbonated and wet the durations are
6095 years. When the element is completely dry,
either alkaline or carbonated, very long durations of at
least 200 years are obtained. However, even if the
alkaline element is wet in the front, the consumption
rates are such that the durations are 1827 years, i.e.,
for the same conditions, much lower than for fully
carbonated element. This indicates that galvanic
anodes are self-regulating, but not so smart to supply
current only when needed, since they consume even
when the steel does not need to be protected. This
suggests that a convenient repair strategy should be
based on the preliminary determination of elements
that really need protection, rather than on the systematic application of anodes to all elements.
Reinforcement in the rear part, which is always
in contact with dry concrete, also contributes to
anode consumption. Table 7 reports the cathodic
current received by the reinforcement, distinguishing between reinforcement in the front part (continuous and bent longitudinal), reinforcement in the
rear part and stirrups (which are roughly half in the
front and half in the rear, see Fig. 6). The fraction
of current on the rear reinforcement is 1520 %.
This contribution to anode consumption cannot be
eliminated due to the electrical contact amongst
reinforcement.

Spacing (m)

Front:
Rear:

AD
AD

0.90

Front

0.45

AW
AD

CW
AD

156

CS
AD

CD
CD

CW
CD

21

162

CS
CD

Bent front

118

16

120

Rear
Stirrups

46
65

7
8

27
71

Front

85

1,031

280

1,166

37

297

1,165

Bent front

76

762

222

860

34

223

837

Rear

28

98

36

110

12

29

48

Stirrups

43

552

157

19

165

588

1854

Materials and Structures (2014) 47:18391855

Table 8 Duration of the anode (t, in year) in fully carbonated

concrete in variable conditions of exposure (concrete in the
front is dry for 205 days, wet for 150 days and saturated for
10 days each year)
H (m)

t (year)

0.22

100

0.66

144

1.10

140

1.54

141

1.98

156

2.42

113

Previous estimates are static since they refer to a

hypothetical condition that does not change in time. In
real conditions of exposure, the elements are subject to
wettingdrying cycles and, as a consequence, their
moisture content changes in time. Also micro-climatic
conditions (e.g., local shelter from rain) affect the
actual moisture content in the element and its evolution in time. Based on data from the weathering station
close to the building considered in this work, it can be
assumed that each year there are 160 rainy days, 10 of
which are consecutive with intensity higher than
2.5 mm. As a consequence, the front part can be
assumed to be dry for 205 days, wet for 150 days and
saturated for 10 days each year. The duration of the
anodes in real exposure conditions can be estimated
with these average annual values. For instance,
referring to the case when the element is fully
carbonated and is protected with six anodes, results
shown in Table 8 are obtained. Values equal to or
higher than 100 years are obtained even for the endanodes. It should be reminded that these values only
refer to the consumption of the anodic material; in real
conditions, further factors (not considered here) may
affect the duration of the anodic system, such as the
effectiveness of the encasing material that keeps the
anode active, or the electrical connection to the steel.

5 Conclusions
A combined experimental and numerical study was
carried out to investigate the possibility of protecting
slender reinforced concrete elements from carbonation-induced corrosion using few small localised
galvanic anodes.

Galvanostatic cathodic polarisation tests on steel in

concrete in various conditions of carbonation and
humidity allowed to determine the cathodic polarisation
curve that was used as an input in numerical simulations.
The current distribution obtained allowed to evaluate the effectiveness of two anodic configurations
(anode spacings of 0.90 and 0.45 m) in protecting steel
reinforcement and to make predictions on the longterm behaviour of the anodes in relation with their
consumption.
The configuration with anode spacing of 0.90 m
(three anodes in the case considered) is not suitable
due to incomplete protection of portions of reinforcement that are far from the anodes.
The configuration with anode spacing of 0.45 m
(six anodes) is suitable for the protection of the
reinforcement from corrosion: although a cathodic
polarisation higher than 100 mV can not be obtained
in all the corroding scenarios considered, when the
concrete is wet the corrosion rate is kept lower than
2 mA/m2; when it is water-saturated, portions of steel
with corrosion rate higher than 2 mA/m2 are present,
however these conditions are unlikely to occur for
more than few days each year.
Estimated anode durations are of the order of several
tens or even hundreds of years, except when concrete is
water-saturated or when it is alkaline and wet, where
durations are lower than 20 years. Also reinforcement
in contact with dry concrete, even in the rear part of the
element, that does not need to be protected from
corrosion, contributes to anode consumption. Hence,
in order to avoid extra costs of repair and waste of
materials, it seems appropriate to use a galvanic anode
CP system only where and when corrosion has initiated
and propagates due to carbonation.
Acknowledgments

This study was financed by CIS-E, Milan.

References
1. Alonso C, Andrade C, Gonzales JA (1988) Relation
between resistivity and corrosion rate of reinforcements in
carbonated mortar made with several cement types. Cem
Concr Res 18(5):687698
2. Anstice DJ, Page CL, Page MM (2005) The pore solution
phase of carbonated cement pastes. Cem Concr Res
35(2):377383. doi:10.1016/j.cemconres.2004.06.041
3. Bennett JE, Turk T, Tettamanti M, Manghi M, Tremolada S
(2008) Testing STARGARDTM discrete anodes for concrete. In: Proceedings of Eurocorr 2008, 711 September,
Edinburgh UK

Materials and Structures (2014) 47:18391855

4. Bertolini L, Bolzoni F, Pastore T, Pedeferri P (1998)
Cathodic protection of reinforcement in carbonated concrete. In: Proceedings of corrosion/98, paper 639, NACE,
Houston, pp 113
5. Bertolini L, Pedeferri P, Redaelli E, Pastore T (2003) Repassivation of steel in carbonated concrete induced by
cathodic protection. Mater Corros 54(3):163175. doi:10.
1002/maco.200390036
6. Bertolini L, Lollini F, Redaelli E (2009) Corrosion assessment of structural and decorative reinforced concrete of
Torre Velasca in Milan. In: Mazzolani FM (ed) Proceedings
of international conference on PROHITECH 09, 2124
June, Rome. Taylor and Francis Group, London, pp 495500
7. Dias WPS (2000) Reduction of concrete sorptivity with age
through carbonation. Cem Concr Res 30(8):12551261
8. EN ISO 12696 Standard (2012) Cathodic protection of steel
in concrete
9. Glass GK, Page CL, Short NR (1991) Factors affecting the
corrosion rate of steel in carbonated mortars. Corros Sci
32(12):12831294
10. Heinemann HA, van Hees RPJ, Nijland TG (2008) Concrete:
Too young for conservation? In: Fodde E (ed) Structural
analysis of historic construction: preserving safety and significance. Proceedings of the VI international conference on
SAHC08, 24 July. CRC Press, Bath, pp 151159
11. Holmes SP, Wilcox GD, Robins PJ, Glass GK, Roberts AC
(2011) Responsive behaviour of galvanic anodes in concrete
and the basis for its utilisation. Corros Sci 53(10):
34503454. doi:10.1016/j.corsci.2011.06.026
12. Jordan LC, Page CL (2003) Mortars for encapsulating
sacrificial zinc anodes in reinforced concrete. Mater Corros
54(6):387393. doi:10.1002/maco.200390088
13. Lambert P (1995) Cathodic protection of reinforced concrete. Anti-Corros Methods Mater 42(4):45
14. Macdonald S (2003) The investigation and repair of historic
concrete. NSW Heritage Office, Parramatta
15. Matthews S (2007) CONREPNET: performance-based
approach to the remediation of reinforced concrete structures: achieving durable repaired concrete structures. J Build
Apprais 3(1):620. doi:10.1057/palgrave.jba.2950063
16. Mo L, Panesar DK (2012) Effects of accelerated carbonation on the microstructure of Portland cement pastes
containing reactive MgO. Cem Concr Res 42(6):769777.
doi:10.1016/j.cemconres.2012.02.017
17. Ngala VT, Page CL (1997) Effects of carbonation on pore
structure and diffusional properties of hydrated cement
pastes. Cem Concr Res 27(7):9951007
18. Pacheco J, Polder RB, Fraaij ALA, Mol JMC (2011) Shortterm benefits of cathodic protection of steel in concrete. In:
Grantham M, Mechtcherine V, Schneck U (eds) Proceedings of concrete solutions. Taylor and Francis, Dresden,
pp 147156
19. Pedeferri P (1996) Cathodic protection and cathodic prevention. Constr Build Mater 10(5):391402
20. Polder RB (1998) Cathodic protection of reinforced concrete structures in The Netherlandsexperience and

1855

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

developments, cathodic protection of concrete10 years

experience. Heron 43(1):314
Polder RB (2001) Test methods for on site measurement
of resistivity of concretea RILEM TC-154 technical
recommendation. Constr Build Mater 15(23):125131.
doi:10.1016/S0950-0618(00)00061-1
Polder RB, Peelen WHA, Lollini F, Redaelli E, Bertolini L
(2009) Numerical design for cathodic protection systems for
concrete. Mater Corros 60(2):130136. doi:10.1002/maco.
200805056
Polder R, Andrade C, Elsener B, Vennesland , Gulikers J,
Weidert R, Raupach M (2000) Test methods for on site
measurement of resistivity of concrete. Mater Struct
33(10):603611
Polder RB, Peelen WHA (2011) Service life aspects of
cathodic protection of concrete structures. In: Grantham M
(ed) Concrete repair. A practical guide. Taylor and Francis,
Abingdon, pp 248261
Redaelli E, Bertolini L (2011) Electrochemical repair
techniques in carbonated concrete. Part II: cathodic protection. J Appl Electrochem 41(7):829837. doi:10.1007/
s10800-011-0302-3
Redaelli E, Bertolini L, Peelen W, Polder R (2006) FEMmodels for the propagation of chloride induced reinforcement corrosion. Mater Corros 57(8):628635. doi:10.1002/
maco.200603994
Redaelli E, Carsana M, Gastaldi M, Lollini F, Bertolini L
(2011) Electrochemical techniques for the repair of reinforced
concrete suffering carbonation-induced corrosion. Corros
Rev 29(56):179190. doi:10.1515/CORRREV.2011.008
Redaelli E, Lollini F, Bertolini L (2013) Throwing power of
localised anodes for the cathodic protection of slender carbonated concrete elements in atmospheric conditions.
Constr Build Mater 39:95104. doi:10.1016/j.conbuildmat.
2012.05.014
de Rincon OT, Hernandez-Lopez Y, de Valle-Moreno A,
Torres-Acosta AA, Barrios F, Montero P, Oidor-Salinas P,
Montero JR (2008) Environmental influence on point anodes
performance in reinforced concrete. Constr Build Mater
22(4):494503. doi:10.1016/j.conbuildmat.2006.11.014
de Rincon OT, de Romero MF, de Carruyo AR, Sanchez M,
Bravo J (1997) Performance of sacrificial anodes to protect the
splash zone of concrete piles. Mater Struct 30(9):556560
Schiessl P (ed) (1994) RILEM Technical Committee
124-SRC, Draft recommendation for repair strategies for
concrete structures damaged by reinforcement corrosion.
Mater Struct 27(7):415436
Sergi G (2011) Ten-year results of galvanic sacrificial anode
in steel reinforced concrete. Mater Corros 62(2):98104.
doi:10.1002/maco.201005707
Sitton I, Powers RG, Costa JE (1998) Galvanic metalized
zinc cathodic protection system for a carbonated reinforced
concrete structure. In: Proceedings of corrosion/98, paper
645. NACE, Houston, pp 114