Materials and Structures / Matriaux et Constructions, Vol.

36, August-September 2003, pp 461-471

RILEM TC 154-EMC: Electrochemical Techniques for Measuring Metallic Corrosion

Recommendations

Half-cell potential measurements Potential mapping on

reinforced concrete structures
Prepared by B. Elsener with contributions from C. Andrade, J. Gulikers, R. Polder and
M. Raupach
The text presented hereafter is a draft for general consideration. Comments should be sent to the TC Chairlady: Dr. Carmen
Andrade, CSIC - Instituto de Ciencias de la Construccion "Eduardo Torroja", Serrano Galvache s/n - Aptdo 19.002, 28033
Madrid, Spain; Tel.: +34 1 302 04 40; Fax: +34 1 302 07 00; Email: [email protected], by 31 December 2003.
TC Membership Chairlady: C. Andrade, Spain; Secretary: B. Elsener, Switzerland/Italy; Members: C. Alonso, Spain;
R. Cigna, Italy; J. Galland, France; J. Gulikers, The Netherlands; U. Nrnberger, Germany; R. Polder, The Netherlands; V.
Pollet, Belgium; M. Salta, Portugal; . Vennesland, Norway; R. Weydert, Germany/Luxemburg; Corresponding
members: C. Page, UK; C. Stevenson, South Africa.

Corrosion of Steel in Concrete, was established in 1980,

presenting a state-of-the-art report in 1988 on the generally
agreed knowledge about the processes causing corrosion
and the corrosion mechanism themselves. RILEM TC-124
on Strategies for the Repair of Concrete Structures
Damaged by Steel Corrosion published a draft
recommendation in 1994, where the importance of
condition assessment has been pointed out as the first step
in the repair process. To ensure a sound decision on the
type of repair work, it is essential to get first of all a
thorough assessment of the condition of the structure,
including the evaluation of:
- cause of damage or loss of protection
- degree and amount of damage
- expected progress of damage with time
- effect of damage on structural behaviour and
serviceability.
The acquisition of reliable information on level and rate
of deterioration is done during regular main inspections
relying on visual examination of the whole structure. If no
reliable conclusion on the state of the structure or part of it
can be obtained, an additional inspection (verification)
including non-destructive testing has to be performed until
the cause of damage is clearly identified.

1. PREFACE
Corrosion of steel in concrete is one of the major
problems with respect to the durability of reinforced
concrete structures. The majority of concrete structures
perform well even after a long period of use in normal
environments. However, there are various reinforced
concrete structures important for our infrastructure,
especially bridges and buildings, which exhibit premature
damage due to environmental actions (EN 206). In contrast
to mechanical actions (as load, wind, etc.) the
environmental actions are not reversible and accumulate
hazardous components (as e.g. chloride ions) in the
concrete. A high percentage of the damages is caused by
insufficient planning, wrong estimation of the severity of
environmental actions and by bad workmanship and thus
many of these structures need to be repaired after a short
service life. In most industrial countries, repair and
restoration of reinforced concrete structures today is nearly
as important as new constructions.
Due to the importance of the problem, RILEM set up an
initial committee on reinforcing steel corrosion in 1960. A
second committee related to the problem, TC 60-CSC

1359-5997/03 RILEM

461

RILEM TC 154-EMC
Following the work of TC-124, RILEM set up a
Technical Committee 154-EMC on Electrochemical
Techniques for Measuring Corrosion of steel in concrete.
During the first meeting of the committee it was decided to
prepare RILEM Technical Recommendations on
- Half cell potential measurements
- Test methods for on-site corrosion rate measurements
of reinforcement by means of the polarization
resistance
- Test methods for on-site measurement of resistivity of
concrete
- Embedded Probes for Corrosion monitoring
These documents are presently prepared for publication.
This Draft RILEM Technical Recommendation on Half-Cell
potential measurements intends to give a general description
of half-cell potential measurements for assessing the
corrosion risk of an existing structure (inspection), for
locating precisely corroding rebars prior to and during repair
work and for evaluating the effectiveness and durability of
the repair work. Special attention will be paid to factors
influencing potential measurements and to the interpretation
of potential readings.

Half-cell potential measurements are recommended and

may be used for one of the following main purposes:
1 Locate corroding rebars and thus assess the present
corrosion condition of the reinforcement during
inspection and condition assessment of a RC structure
2 Define the position for further destructive analysis
(cores for chloride analysis, inspection windows to
visually examine the corrosion state of the rebars etc.)
and of embedded sensors for further monitoring on the
basis of the results of potential measurements
3 Evaluate the corrosion state of the rebars after repair
work and thus evaluate the efficiency and durability of
repair work
4 Design an anode layout of cathodic protection systems
or electrochemical restoration techniques
Half-cell potential measurements do not provide
quantitative information on the actual corrosion rate of the
rebars. They should be interpreted in the context of
complementary data from the concrete structure (chloride
content, carbonation, porosity etc.) by specialists or skilled
engineers experienced in the field of corrosion testing and
structural evaluation.

2. SCOPE

4. DEFINITIONS

This RILEM Technical Recommendation intends to

provide the background, a description of the application and
guidelines for the interpretation of half-cell potential
measurements on reinforced concrete structures. It covers
both: point measurements (mostly used during inspection,
thus in the project phase of a restoration) and potential
mapping.

The definitions and terminology used in this

recommendation are in agreement with ISO 8044 (1999)
third edition.

5. THEORETICAL BACKGROUND OF
HALF-CELL POTENTIAL MEASUREMENTS

3. SIGNIFICANCE AND USE

5.1 The electrochemistry of corrosion of steel

in concrete

Half-cell potential measurements as described in this

document can be performed on structures with ordinary or
stainless steel reinforcement. Corrosion of prestressing steel
in concrete can be assessed in the same way. Prestressing
steel in the ducts of posttensioned cables can not be
assessed: metallic ducts shield the strands from the concrete
surface, plastic ducts isolate the strands. Epoxy coated steel
is not treated in this recommendation because EP coated
bars usually are electrically isolated from each other and
thus cannot be measured. Galvanized steel can be measured
in the same way but different criteria applied.
Half-cell potential measurements are suitable mainly on
reinforced concrete structures exposed to the atmosphere.
The method (measurement and interpretation) can be
applied regardless of the depth of concrete cover and the
rebar size and detailing. Half-cell potential measurements
will indicate corroding rebars not only in the most external
layers of reinforcement facing the reference electrode but
also in greater depth.
The method can be used at any time during the life of a
structure and in any kind of climate providing the
temperature is higher than +2C. Half-cell potential
measurements should be taken only on a free concrete
surface, the presence of isolating layers (asphalt, organic
coatings or paints etc.) may make measurements erroneous
or impossible.

Corrosion of steel in concrete follows the

electrochemical mechanism of corrosion of a metal in an
electrolyte [1, 2]. Chemical thermodynamics determine
whether a particular metal tends to corrode in a particular
environment. According to the equilibrium potential / pH
diagram (or Pourbaix diagram), ordinary reinforcing steel
in the highly alkaline environment of concrete is protected
by a thin oxide film (the passive film) from further
corrosion. This protective passive film may be destroyed by
chlorides penetrating through the concrete cover to the steel
surface and / or by carbonation of the concrete, thus the loss
of alkalinity due to the reaction with CO2 from the
atmosphere - the steel is then depassivated. Active
corrosion will take place when the steel is depassivated and
both oxygen and humidity, are present at the steel surface.
Corrosion of a metal implies separate anodic and
cathodic processes occuring simultaneously on the same
metal surface. Corroding steel in concrete acts as a mixed
electrode with coupled anodic and cathodic reactions taking
place simultaneously on its surface [1, 2]. At the corroding
site (the anode), iron is dissolved and oxidized to iron ions
leaving electrons in the steel: Fe -> Fe2+ + 2 e-. Due to the
condition of electroneutrality, these electrons have to be
consumed by the cathodic reaction at the steel surface,

462

Materials and Structures / Matriaux et Constructions, Vol. 36, August-September 2003

where oxygen is reduced and hydroxyl ions are produced:
O2 + 2H2O + 4e- -> 4 OH-.
Depending on the spatial distribution of anodic and
cathodic reactions on the surface of the steel and the
conductivity of the medium, two different types of
corrosion can take place:

passive film on the steel. Under these circumstances the

metal behaves actively, the corrosion potential is depressed
to very negative values in the range of ca. -0.9 to -1.0 V
SCE. The rate of metal dissolution is very low owing to the
very low cathodic current density [1].

5.1.1 Uniform corrosion

In macro-cell or localized corrosion, the anodic and

cathodic areas of the corrosion process are separated in
space. Under these circumstances, a significant amount of
ohmic potential drop (IR-drop) separates the potential of
the anodic and the cathodic area and the Evans diagram is
modified (Fig. 1b). The corrosion potential varies with
position between the limits of Ecorr (anode) and Ecorr
(cathode). The positions of local anodes are indicated by
regions of more negative corrosion potentials - thus surveys
of corrosion potential can be used to locate sites of active
corrosion.
For steel in concrete, macro-cell corrosion is occurring
in the case of localized corrosion attack due to the action of
significant levels of chlorides in the concrete. The corrosion
takes the shape of corrosion pits, concentrated anodic areas,
surrounded by relatively large areas of passive steel (acting
as cathode).

5.1.2 Macro-cell corrosion

The simplest form of a corrosion process is one in which

anodic and cathodic reations are coupled to form numerous
micro-cells on the steel surface and occur on the same
areas. This type of corrosion can be represented by the
Evans diagram (Fig. 1a) showing the polarization curves for
the separate anodic and cathodic reactions intersecting at a
point P where the mean anodic and cathodic current
densities are equal and represent the corrosion rate icorr.
The electrode potential of this couple at this point is termed
the corrosion potential Ecorr.. It is constant over the whole
surface of the corroding zone. For steel in concrete, uniform
corrosion is normally encountered when the steel has lost
its passive state due to carbonation or very high chloride
concentration, and humidity and oxygen are present in
sufficient amounts to sustain the cathodic reaction.
A special case of uniform corrosion is the state of active,
low potential corrosion. In environments where the
availability of oxygen is very limited, as is sometimes the
case for fully submerged or buried reinforced concrete, the
cathodic current density may be too low to maintain the

5.2 Half-cell potentials

The corrosion potential Ecorr (half-cell rebar / concrete) is
measured as potential difference (or voltage) against a
reference electrode (half-cell). As a corrosion detection
technique this was first used by Stratful [3]. The numerical
value of the measured potential difference between the steel in
concrete and the reference electrode will depend on the type of
reference electrode used and on the corrosion condition of the
steel in concrete. In addition, half-cell potentials of steel in
concrete cannot be measured directly at the interface concrete /
rebar due to the presence of the concrete cover (Fig. 2), the
potentials are thus influenced by iR drop in the cover, by
macrocell current and possibly by junction potentials.

Fig. 2 - Schematic view of the electric field and current flow in

an active / passive macrocell on steel in concrete.

5.2.1 Reference electrodes

It is essential to quote always the reference electrode being
used for half-cell potential measurements. In practice the
following reference electrodes with a defined, constant and

Fig. 1 - Evans diagram for uniform corrosion (a) and for a corrosion
couple with ohmic drop (b). Ecorr = corrosion potential, Eeq =
equilibrium potential, icorr = corrosion current density.

463

RILEM TC 154-EMC
reproducible potential versus the standard hydrogen electrode
(SHE) are used:
Copper / copper sulfate sat
CSE
Calomel (Hg / Hg2Cl2) KCl sat. SCE
Silver chloride (Ag / AgCl) KCl
SSCE
sat.

enhances the potential differences between the corroding and

passive zones. The moisture content of the concrete can be
checked with resistivity measurements (RILEM Technical
Recommendation Test methods for on-site measurements of
resistivity of concrete). Rusted rebars in very dry concrete
may not be well detected.

+ 0.318 V SHE
+ 0.241 V SHE
+ 0.199 V SHE

Influence of pH
No systematic trends are found with respect to the influence
of pH when using different cement types or additions of
microsilica. From experience it is known that the potentials of
steel in concrete containing slag cements are more negative
compared to OPC concretes.

For on site work, the saturated copper-copper sulfate

electrode is most robust and is sufficiently accurate, although
errors may arise due to contamination of the concrete surface
with copper sulphate [4]. Calomel and silver chloride
electrodes are used more in laboratory work. The base
potential of the reference electrodes depends on the
concentration of the electrolyte, thus care has to be taken to
operate in saturated conditions. The temperature dependence
of the reference electrodes has no practical influence on the
readings on site.

5.2.3 Macrocell and localized corrosion

In the case of macrocell corrosion and localized corrosion
attack the reference electrode placed on the concrete surface
does not measure the true corrosion potentials of anode and
cathode but a mixed value which depends on the position of
the electrode and the concrete resistivity. The most negative
values of the half-cell potentials always locate the local anode.
The absolute values of the potentials measured are influenced
by the cover depth and the resistivity of the concrete:

5.2.2 Corrosion state of steel in concrete (uniform

corrosion)
Half-cell potentials measure the corrosion potential Ecorr
which is related to the corrosion state of steel in concrete. As is
shown in subsequent paragraphs, the measured values are
influenced by concrete cover, concrete resistivity (moisture
content) and oxygen availability. Some typical ranges of
potentials of steel in concrete for different conditions are given
in Table 1.

Cover depth
The potential difference between the position above the
anode and a distant cathode becomes smaller with
increasing cover depth - thus the location of a small
corroding spot becomes more difficult with high cover
depth (Fig. 3) [5]. On the other side, these potential
gradients can indicate if a corroding area is located near the
surface or at rebars deep in the concrete.

Table 1 - Typical ranges of potentials of normal steel

in concrete (Volts CSE)
water saturated concrete without oxygen -0.9 .... -1.0 V
wet, chloride contaminated concrete

-0.4 .... -0.6 V

humid, chloride free concrete

+0.1 .... -0.2 V

humid, carbonated concrete

+0.1 .... -0.4 V

dry, carbonated concrete

+0.2 .... 0 V

dry concrete

+0.2 .... 0 V

The low potential active conditions (see 5.1.1) result in

very negative potentials (ca. -1.0 V CSE). Corroding steel in
chloride contaminated concrete shows potentials ranging from
-0.6 V to -0.4 V CSE. Passive steel in concrete reacts as an
oxygen electrode on variations in pH of the concrete pore
solutions, thus the corrosion potential depends with dE/dpH = 0.06 V (at 25C) on the pH. This means that potentials in
carbonated concrete are more positive. The corrosion potential
of passive steel in concrete further depends on the oxygen
availability (moisture content) and can vary over a wide range
of potentials. Passive stainless steels show similar potentials to
ordinary passive steel.

Fig. 3 - Influence of cover depth (distance from the steel

surface) on half cell potentials over an active / passive
macrocell [15]. Resistivity 1300 m, total length 30 cm, anode
0.5 cm.

Influence of concrete resistivity (moisture content)

A major influence on the potential readings is the condition
of the surface layer of the concrete (wet or dry). Changes in the
moisture content (e.g. by wetting of the concrete surface) shift
the potential values in the wetted areas to more negative values
[4], but the potential gradients and location of local potential
minima (corroding spots) do not change. Inhomogeneous
moisture distribution (e.g. dry/passive and wet/corroding)

Concrete resistivity
The macrocell current affects the potentials of the cathode
(passive steel): the anode is polarizing the passive rebars in the
vicinity of the corroding spot to more negative potentials
(Fig. 4). This shift of potentials to more negative values is

464

Materials and Structures / Matriaux et Constructions, Vol. 36, August-September 2003

6. PRACTICAL EXECUTION OF
MEASUREMENTS
6.1 Measurement arrangement
The measurement of half-cell potentials is based on the
electrical and electrolytic continuity between rebars in
concrete, reference electrode on the concrete surface and
voltmeter (Fig. 5).

Fig. 5 - Principle and main components of half-cell potential

measurements: Reference electrode, high impedance
voltmeter, connection to the rebar.

Fig. 4 - Influence of electrolyte resistivity on half cell potential

distribution measured on an active / passive model macrocell
[15]. Cover depth 20 mm, total length 30 cm, anode 0.5 cm.
Open symbols in aqueous electrolyte, closed symbols in very
wet, chloride containing mortar.

6.1.1 Voltmeter
The voltmeter should have a high input impedance so
that the current flowing through the reference electrode
does not disturb the stability of the reference electrode
potential. In general the current should not exceed about
10 pA. For measurements on normal outdoor concrete the
minimum acceptable input impedance is 10 M:.
The connections to the voltmeter should be made as
follows: the reference electrode is connected to the negative
terminal, the reinforcing steel to the positive terminal of the
voltmeter. In this arrangement half-cell potential readings
generally will be negative. The occurrence of positive
potentials is possible on passive rebars in dry concrete.

pronounced in low resistive concrete (wet, chloride

contaminated) and it helps to detect even very small corrosion
spots with a comparatively large grid size (e.g. 0.15 m) [5, 6].
In high resistive concrete (dry, carbonated structures) the most
pronounced differences between anode and cathode are found,
but only a very small area of passive rebars is polarized and the
detection of small corroding spots becomes more difficult.
Further factors influencing the distance of macrocell
action are the availability of oxygen at the cathode (cathodic
current density) and the amount of passive rebar surface
around the anode.

6.1.2 Electrolytic contact to the concrete surface

Electrical conduction between the reference electrode and
the concrete is established by the transport of ions. This can
be effected by placing a conductive bridge between the
reference electrode and the concrete. In practice, frequently a
wet sponge is placed between the reference electrode and the
concrete surface. The sponge has to be wetted periodically
with a diluted solution of detergent. This conductive bridge
lowers the resistance between the reference electrode and the
concrete and reduces the liquid junction potential between
the electrolyte in the reference electrode and concrete. It is
important that the sponge is not allowed to dry out and it
should be cleaned or replaced regularly to prevent
contamination of the reference electrode.
If potential wheels or multiple electrode devices with
wheels are used (e.g. Fig. 6), the contact to the concrete
shall be established in a similar way. The wheels shall
provide good electrolytical contact to the concrete surface.
On dry concrete, during warm periods etc. more water is
needed to establish the electrolytic contact. Excessive
prewetting of the concrete surface changes the natural
conditions, and a potential shift to more negative values

Contact with buried or submerged parts of RC

structures
Buried or submerged parts of RC structures show very low
potentials due to the restricted oxygen availability (see 5.2.1).
The potential of passive rebars near the buried or submerged
ones is polarized to low values, which should not be
misinterpreted as corroding steel [6, 7].
5.2.4 Junction potentials
At interfaces between solutions of different composition
(chloride content, pH etc.) junction potentials can be created
due to the different transport numbers (product of
concentration and ion mobility) of ions present. The
electrolytic connection between the reference electrode on the
surface and the concrete occurs by ionic flow, thus junction
potentials might in principle arise. Differences can be expected
due to changes in surface pH or different chloride
concentrations in the concrete. The magnitude of such junction
potentials and their significance is under discussion. For
practical purposes their influence on half-cell potential
measurements is negligible.

465

RILEM TC 154-EMC

6.3 Coordinate system

The primary goal of potential measurements on
reinforced structures is to locate areas of corroding
reinforcement. Thus before starting to take measurements a
coordinate system has to be put onto the structure in order
to be able to relate readings with the point on the structure
where they were taken. The coordinate system may be
oriented towards characteristic features (appearing
equidistant on the structure) or consist of equidistant lines
marked onto the concrete surface. During automatic
measurements with a potential wheel that measures also the
distance the agreement with the coordinate system on the
structure has to be checked frequently.
Fig. 6 - Multiple wheel electrode half-cell potential measuring
instrument with computer assisted data acquisition [5, 10].
Note the slight wetting of the concrete surface at the wheels in
order to achieve a good electrolytic contact between reference
electrode and concrete.

6.4 Procedure
Potential measurements can be performed with a single
electrode (point measurements) or with one or several
wheel electrodes (potential mapping).
Inhomogeneous surface conditions as e.g. wet and dry
areas or not embedded rebars that are encountered should
be located on a sketch in order to facilitate the
interpretation of potential readings.
For a statistical evaluation of the data (cumulative
probability plot, histogram) the grid spacing must be kept
constant.

will result (see 5.2.2). In conditions of very dry, carbonated

concrete corrosion could be activated. However, this might
be done intentionally in order to be able to locate corroding
steel. It should not be misinterpreted in terms of high
corrosion risk.
6.1.3 Electric contact to the reinforcement
The electrical connection to the rebars requires a
metallic, electrically sound contact. If crocodile clips are
used the rebars have to be cleaned from rust. A secure
connection can be obtained by drilling a 25 mm hole into
the concrete above the rebar, then drill a 4 mm hole into the
rebar and insert a self tapping screw with attached lead into
the hole in the steel. Sometimes it may be convenient to
make the electrical connection to an external metallic
feature of the structure such as a pipe or parapet. In such a
case it is essential to check that there is electrical continuity
with the reinforcement.

6.4.1 Point measurements

Measurements with a single electrode will usually be
performed with a large grid size (0.5 m up to 2 m) e.g. in
the project phase of a restoration or condition assessment
(in order to get a rapid and cost effective overview on the
corrosion state of a structure) or on comparatively small
objects. These measurements can be made on vertical, on
top or bottom horizontal surfaces or any other orientation.
When measuring horizontal surfaces from the bottom, care
has to be taken that the reference electrode is filled with
sufficient electrolyte solution.
6.4.2 Potential mapping
Potential mapping is usually performed with a multiple
wheel arrangement and a small grid size of ca. 0.15 m on
large surfaces (bridge decks, walls, parking decks etc).
With single or multiple electrode instruments and computer
assisted data acquisition (Fig. 6) several 100 m per hour
can be measured. About 30-50 single measurements are
taken per square meter of concrete surface. For this reason
the precision of the individual reading is less important, a
higher drift (see below) can be accepted.
Potential mapping is most easy when measuring
horizontal surfaces (bridge decks, parking decks). Vertical
surfaces or horizontal surfaces from the underside can be
measured with the same experimental arrangement,
provided a suitable construction that presses the (multiple)
wheels onto the concrete surface is available.

6.1.4 Reinforcement continuity

In all cases, the electrical continuity of the reinforcement
has to be checked prior to perform half-cell potential
measurements. On large structures like bridge decks, on
structures with different construction intervals or on very old
structures, continuity may not be present overall or locally.
The resistance determined with AC measurements between
remote points on a structure should be < 1 Ohm (without
resistance of the connecting cable) in order to get reliable
measurements. If the resistance values are higher, multiple
electrical contacts shall be made to the reinforcement. Note: on
structures with epoxy coated reinforcement (not welded), the
individual bars are isolated and the electrical continuity
necessary to measure half-cell potentials is not given.

6.2 Calibration and standardization

6.5 Source of errors in measurements

The reference electrode(s) used for half-cell potential

measurements have to be checked against a standard (e.g.
laboratory Calomel electrode) in the lab or on site before
starting to take measurements. The difference should be
less than 10 mV.

The requirements for stability of a single half-cell

potential reading are higher for point measurements with a
large grid size than for potential mapping because every
point is representative of a large area.

466

Materials and Structures / Matriaux et Constructions, Vol. 36, August-September 2003

o

Drift of half-cell potential readings

If the half-cell potential reading shows a systematic
change over time (called drift), this may be due to water
entering into dry concrete changing the moisture content
(and resistivity) at the point of measurement. Rates of drift
of less than 20 mV per minute can be ignored. On some
commercial instruments the drift tolerance can be selected
and an acoustic signal indicates the operator when readings
are stable and are taken by the instrument.
o

Randomly fluctuating readings

If the half-cell potential readings are not stable but
fluctuating randomly, a loss of electrical continuity is usually
the cause. This can be due to air bubbles or lack of enough
electrolyte in the reference electrode, sponge being dried out,
bad electrical connections to the rebars or to the voltmeter.
Not yet detected concrete delaminations may result in similar
effects. Fluctuations may be induced also by the presence of
stray currents (DC rail- or tramway, CP systems etc.).

6.6 Data presentation

The way of representing half-cell potential measurements
may vary according to the number of readings taken: if a
large grid size was used or only a few readings were taken,
the half-cell potential values may be given as table or be
written directly into the plan view. A large number of data
can be presented as potential map and / or in a statistical
representation.

Fig. 7 - Examples of half-cell potential maps (Riding deck in

the Tunnel San Bernardino) [5, 6]. Data representation: colour
plot (right) and equicontour line plot (left). DF: dilatation joint
(every 25 m).

6.6.1 Half-cell potential map (potential field)

Representation of half-cell potential data in order to locate
areas with different potential can be performed by colour plots,
equipotential contour plots, 3D plots or others. In a colour plot
(Fig. 7) each individual data point is represented without
smoothing or interpolation by a coloured square of adequate
size on a suitably scaled plan view. The colour is related to the
potential interval (e.g. red = -0.4 - -0.35 V CSE). The
maximum potential interval associated to a colour shall usually
not exceed 0.05 V. Colour plots allow to identify erroneous
readings or potential shifts in one electrode when multiple
wheel electrodes are used. Erroneous readings (e.g. electrode
occasionally not well contacted) are visible but do not
influence the map as a whole.
In an equipotential contour plot (Fig. 7) lines of constant
potential are calculated and plotted through points of equal or
interpolated equal values. Contour plots react very sensitive on
single erroneous readings (e.g. very negative potentials) because
fictitious contour lines are plotted around the erroneous point.
The maximum contour interval should be 50 mV.
3D plots represent the potential readings on the z axis
versus the x/y position. This representation usually is of
limited value when the measured potential maps are very
inhomogeneous.

Histograms or frequency distributions are obtained when

all potential readings are grouped in intervals of potential
and the number of readings in each group are counted. If
the intervals chosen are quite large (e.g. 50 mV), a
histogram of potential readings is obtained. A frequency
distribution (Fig. 8) is obtained when the group interval is

6.6.2 Statistical representation of the data

The data to be represented and treated statistically (histogram,
frequency distribution or cumulative probability plot) have to be
measured and recorded on a regular grid size (a smaller grid size
e.g. in areas with corrosion would erroneously increase the
percentage of corroding reinforcement on the structure). In the
statistical representations the information on position of the
reading is lost.

Fig. 8 - Frequency distribution of half cell potential data

(retaining wall at Rofla, alpine highway N13 in Switzerland).
Original data , curve fitted assuming gaussian distribution for
active and passive potentials [22].

467

RILEM TC 154-EMC
small (5 mV or 10 mV, according to the total number of
readings) and the number of readings N(E) is plotted versus
the potential E. The frequency distribution gives a first
impression on mean value of potential, distribution and
skewness. It may be used to compare different parts of a
structure (e.g. two lanes of a bridge deck, different oriented
facades etc.). From the frequency distribution the presence
and percentage of actively corroding and passive rebars can
be estimated.
Cumulative frequency distributions are obtained by
summing up the frequency distribution and plotting the values
6 N(E) versus the potential E on a probability plot (Fig. 9). A
single gaussian distribution would appear as one straight line.
Usually, two or three straight lines with different slopes are
obtained indicating the passive and active distribution of
potentials. Due to the (usually) much higher number of
readings of passive potentials, this distribution is defined
better. From the cumulative probability plot the border value
of potential (90% or 95% passive) can be determined.

7. INTERPRETATION AND USE OF HALFCELL POTENTIAL DATA

7.1 Interpretation of half-cell potential data
7.1.1 Potential range indicating active and passive state
Half-cell potential measurements taken from the concrete
surface are influenced not only by the corrosion state of the
rebars (active or passive) but - as has been shown in chapter 5 by the concrete cover and the resistivity of the concrete. The
resistivity itself varies with temperature, with concrete
moisture and chloride content or carbonation. Thus the surface
state and moisture content of concrete affect the numerical
value of potential readings as is documented by a lot of
research work [4-9] and practical application of potential
mapping [10-12] where the corrosion state of rebars
(corroding, passive) has been related to the potential measured
at the concrete surface (Fig. 10). Very different potential
ranges indicate corroding steel on different bridge decks or
substructures. The values of potential known as van Daveer
Criteria [13] and used also in the ASTM Standard C876 (under
revision) should not be used as absolute criteria to determine
the condition of steel in concrete.

Fig. 9 - Cumulative frequency distribution (probability plot) of

half-cell potentials measured on different bridge decks [6, 10,
15]. Note the curve San Bernardino: due to the abrupt change
in concrete humidity (corroding = wet, passive = dry) the
distribution curves of active and passive potentials are shiftet by
ca. 150 mV.

7.1.2 Half-cell potentials and concrete resistivity

(moisture content)
Changes in the moisture content (e.g. by wetting of the
concrete surface) lead to a shift of the whole potential field to
more negative values, e.g. a shift of ca. 100 mV was found on
a bridge deck measured in dry and wet conditions after rainfall.
The potential gradients and location of potential minima did
not change, differences were found only in the magnitude of
potential readings. This further indicates that the potential
gradient rather than the absolute value of potential is an
indicator of corroding rebars.
In structures with chloride induced corrosion, corroding
areas usually coincide with low resisitivity areas (Fig. 11) as
has been found on a part of a bridge deck where both potential
and resistivity mapping have been performed [14, 15]. This is

Fig. 10 - Experimentally determined potential range indicating

active corrosion on different bridge decks compared to the
ASTM C876 standard [6, 10, 15].
1 Cugnertobel, 2 San Bernardino, 3 Rhinebridge Tamins,
4 Caslertobelbrcke, 5 Morbio bridge, 6 Column in seawater,
7 ASTM C876 Standard.

468

Materials and Structures / Matriaux et Constructions, Vol. 36, August-September 2003

potential readings (Fig. 10), therefore very different
potential values may indicate corroding steel on different
RC structures.
Locating areas of actively corroding steel is best
achieved by considering the spatial variation of potential
(local minima, gradients in the potential field, Fig. 7) and
not the absolute values of the potentials. Rigidly using fixed
boundary limits of potentials to distinguish active and
passive zones (e.g. -0.35 V CSE as proposed in ASTM
C876-91) can be misleading.
Due to the easy polarization of passive rebars close to
the actively corroding spot in low resistive concrete the
location of very small corroding spots is much easier in
cases of chloride induced corrosion than in carbonated
concrete [10, 12].

Fig. 11 - Relation between half-cell potential and concrete

resistivity measured at the underside of a chloride contaminated
bridge deck [14, 15].

7.2.2 Define position for further destructive analysis

The half-cell potentials measured have to be interpreted
in the context of additional information as e.g. corrosion
state of the rebars (visual examination, loss of cross
section), chloride content at the rebar, resistivity of the
concrete, cover depth etc. This requires further nondestructive or destructive analyses. The location of the most
significant points of analyses can be determined in a
rational, cost effective way based on the potential field.

due to the fact that chlorides are transported more easily into
concrete with a lower resistivity (more open pore structure),
high chloride concentrations depassivate the rebars and active
potentials are measured. Points with low resistivity but passive
potentials (Fig. 11) can be interpreted as areas with a high
corrosion risk for the future. Areas with high concrete
resistivity and potentials in the passive range are not affected
by chloride ingress.
The moisture content of the concrete may vary from one
part of the structure to another, an example is given from the
carriageway in an alpine tunnel (Fig. 7). The dilatation joints
are all leaking, thus the concrete in the vicinity of the joints is
wet, highly chloride contaminated and the reinforcement is
actively corroding. Further away from the dilatation joints the
concrete is very dry, no chlorides penetrate and the rebars are
passive. Inhomogeneous moisture distribution enhances the
potential differences between the corroding and passive zones.
As shown above, half-cell potential and resistance data
may be combined to give an overview of various factors
relevant to corrosion risk. However, interpretation requires
taking into account particular features of the structure under
study and cannot be considered straight forward.
The influence of carbonation is twofold: due to the loss of
the concentrated alkaline pore solution the pH of the pore
solution is much lower and carbonated concrete, often
sheltered from rain, shows high resistivity. This influences
both the absolute values of the potentials measured from the
concrete surface and the possibility of locating small corroding
areas. In high resistive media as carbonated concrete, the
location of small corroding areas will be more difficult. Halfcell potential measurements taken on concrete structures with
high resistive surface layers due to carbonation show more
positive potentials of both, passive and corroding rebars.
As mentioned before, high resistive surface layers may
make correct interpretation very difficult.

Inspection windows to determine the state of corrosion

Half-cell potential measurements allow to locate corroding
rebars on concrete structures when corrosion is ongoing. In a
first interpretation no direct information on the state of
corrosion of the rebars (reduction in cross section) can be
obtained. At selected areas where large gradients 'E/'x in
potential exist, the rebars can be inspected visually after
concrete removal [10, 16]. The inspection window should
preferably be about 0.3 - 0.5 m large and extend in length from
completely passive zones to heavily corroding zones (Fig. 7).
The boundary limit of active respectively passive potentials are
determined by associating the rebar state with the measured
half-cell potential values (Fig. 10).
With the information from the inspection windows, a rough
relation between the corrosion state of the rebar and the
measured half-cell potential at the concrete surface can be
obtained for a specific concrete structure [10]. This relation is
not universal and changes from one structure to another.
o

Position of cores for chloride analysis

The position where cores for chloride analysis have to be
taken can be determined in a more rational way based on the
potential field: cores shall be taken in selected areas with
uniform potential, covering the centre of actively corroding
zones, passive zones and in the transition region. In a number of
chloride contaminated reinforced structures [10, 16, 17] a
relation between potential and chloride content at different depth
has been found. This allows to interpret the potential field at
least qualitatively and for the specific structure as chloride
distribution map, thus a more complete picture of the chloride
contamination can be obtained with only few cores.
Potential mapping results can be combined in a useful way
with other types of quick and cheap non-destructive
measurements such as resistivity mapping. Potential maps can
be used to select areas for further analysis using more
elaborated non-destructive testing (e.g. corrosion rate
measurements based on polarization resistance, RILEM TC

7.2 Use of half-cell potential data

7.2.1 Locate corroding rebars
The primary purpose of half-cell potential measurements
is to locate corroding rebars in a concrete structure during
condition assessment or repair work [10-12]. It has been
outlined in the paragraph above, that the surface state and
moisture content of concrete affect the numerical value of

469

RILEM TC 154-EMC
154-EMC) or for sample locations for destructive analyses.
Finally, they can be used to select the positions for embedding
sensors . In all cases, the number of measurements or sensors
can be minimised and the results integrated in a rational way
into the overall picture of the structure.
7.2.3 Control the effectiveness and durability of a
restoration work
There are different repair options for corroding reinforced
concrete structures. Some of them intend to bring the steel in
a non-corroding or passive condition. Thus half-cell potential
measurements can be used in principle to evaluate the
effectiveness and the durability of the repair work. Usually
this requires a potential survey on the same areas as
performed during inspection (prior to repair). The
interpretation of the overall success of the repair work can be
evaluated best on the basis of the cumulative probability plot;
plotting the potential map made after repair will show areas
where steel is still corroding.
o

Traditional repair: replace chloride contaminated

concrete
Half-cell potential mapping can be used to evaluate the
effectiveness of repair works on bridge decks [18]. Potential
readings taken on the repaired bridge deck some months after
repair (but before the membrane and the asphalt will be applied)
have to be compared to the potential map recorded prior to
repair. If the potential values in repaired and old concrete (with
passive rebars) are similar, the repair strategy chosen (removal
of all chloride contaminated concrete in the areas where
corroding rebars were found) was successful [18, 21].
o

Electrochemical Chloride Removal

Electrochemical chloride removal was used as a nondestructive restoration technique on a heavily chloride
contaminated bridge abutment, where corrosion of the rebars
just started. The inspection prior to the repair work as well as the
evaluation of the efficiency of the chloride removal was done
with half-cell potential mapping on the abutment [19]. After the
first chloride removal, the potential values became more positive
by about 80 - 100 mV indicating the repassivation of the rebars.
Some areas with too high chloride contents after the first
chloride extraction were treated a second time after several
month, the half-cell potentials became more positive again. An
evaluation five years after completion of the repair work showed
that no corrosion had initiated and the potential field became
more homogeneous (Fig. 12). In contrast to some isolated
chloride analysis, half-cell potential measurements allow to
evaluate the whole surface of the abutment and they provide
clear information on the repassivation of the rebars [19-21].

Fig. 12 - Cumulative frequency distribution (probability plot) of

half-cell potential data before, after first and second chloride
removal treatment and control after five years [15, 20, 21].

success of the treatment in stopping corrosion should therefore

be related to the establishment of a uniform potential field.
o

Inhibitors
Testing the efficiency of inhibitors penetrated from the
concrete surface with half-cell potential measurements is
difficult [21]. Changes in the pore water composition and thus
the concrete resistivity as well as different possible
mechanisms of the inhibitor action (anodic, cathodic, mixed,
blocking...) can lead to shifts of the half-cell potential measured
from the concrete surface that cannot be interpreted straight
forward. More research is needed before potential mapping may
allow unambiguous evaluation of inhibitor treatment.
In consequence, the following points should be taken
into account when assessing the efficiency and durability of
repair work with half-cell potential measurements:

Electrochemical realkalization
Half-cell potential measurements can be used to evaluate
the corrosion state of the rebars after electrochemical
realkalization. The results on a test site [20, 21] have shown
that the potential values after the treatment (after waiting three
months in order to allow depolarization) were shifted by about
0.2 V to more negative values. Any straightforward, simple
interpretation more positive potentials = repassivation fails.
The decrease in potentials is due to two effects: after the
realkalization treatment, the concrete resistivity is much lower
due to the presence of a quite concentrated sodium
(bi)carbonate solution in the pores and the alkalinity around
the rebars has strongly increased (from about pH 9 to 13). The

Time of measurement
The waiting time prior to perform measurements after
repair work has to be defined according to the type of repair:
electrochemical restoration techniques (ECR, ER) lead to a
strong cathodic polarization of the rebars, the time to
depolarize can be between one and several months according
to the oxygen access to the rebars. New concrete or repair
mortar instead require time to hydrate and dry out partially.

470

Materials and Structures / Matriaux et Constructions, Vol. 36, August-September 2003

[3] Stratful, R.F., The corrosion of steel in a reinforced concrete
bridge, Corrosion 13 (1957) 43-48.
[4] Vassie, R.P.W., Evaluation of techniques for investigating the
corrosion of steel in concrete, TRRL Supplementary Report 39,
Transport and Road Research Laboratory, London, 1978.
[5] Elsener, B. and Bhni H., Location of corrosion in reinforced
concrete: potential mapping Possibilities and limitations,
Schweizer Ingenieur und Architekt 105 (1987) 528-533 [in German].
[6] Elsener B., Mller, S., Suter, M. and Bhni, H., Corrosion
Monitoring of Steel in Concrete Theory and Practice,
Corrosion of Reinforcement in Concrete, ed. C.L. Page,
K.W.J. Treadaway, P.B. Bamforth (Elsevier Applied
Science, 1990) 348-357.
[7] Menzel, K. and Preusber, H., A non-destructive technique to
detect corrosion of rebars, Bauingenieur 64 (1989) 181-186
[in German].
[8] Elsener, B. and Bhni, H., Potential mapping and corrosion
of steel in concrete, in Corrosion rate of steel in concrete,
ASTM STP 1065 , N.S. Berke, V. Chaker, D. Whiting eds.
American Society for Testing and Materials, Philadelphia
(1990) 143-156.
[9] Arup, H., The Mechanism of Protection of Steel in Concrete,
Corrosion of Reinforcement in Concrete Construction, ed.
A.P. Crane, The Society for Chemical Industry, Ellis Horwood
Ltd., London (1983) 151-157.
[10] Elsener, B., Flckiger, D., Wojtas, H. and Bhni, H,
Methods for the evaluation of corrosion of steel in concrete,
Research Program for Bridge Maintenance, Swiss Federal
Department of Traffic and Energy EVED (1996). Published
by VSS Zrich [in German].
[11] Hunkeler, F., Inspection of RC structures with potential
mapping, Schweizer Ingenieur und Architekt 109 (1991) 272
[in German].
[12] SIA Documentation D 0126 Half-cell potential measurements
an important tool for the evaluation of RC Structures, Swiss
Association of Engineers and Architects, Zrich (1995) [in
German].
[13] Van Daveer, J.R., Techniques for evaluating reinforced concrete
bridge decks, J. American Concrete Inst. 72 (1975) 697-704.
[14] Brchler, D., Electrical resistivity of cement based
materials, PhD Thesis No. 11876 ETH Zurich (1996) [in
German].
[15] Elsener, B. and Bhni, H., Half-cell potential measurements
From theory to condition assessment of RC structures, Proc. Int.
Conference Understanding Corrosion Mechanisms of Metals in
Concrete - A Key to Improving Infrastructure Durability,
Massachusetts Institute of Technology, MIT (Cambridge, USA)
27-31 July 1997, Keynote lecture session No. 3.
[16] Suter, M., Potential mapping at the Lorraine Bridge, in [12]
p. 69 [in German].
[17] Elsener, B. and Bhni, H., Condition evaluation of reinforced
concrete bridges The benefits of potential mapping, Proc. 6th
Int. Conf. Structural Faults + Repair, London (1995) 47-52.
[18] Schiegg, Y., Potential Mapping after Bridge Maintenance,
in [12] p. 79 [in German].
[19] Elsener, B., Molina, M. and Bhni, H., Electrical removal of
chlorides from reinforced concrete, Corrosion Science 35
(1993) 1563-1570.
[20] Elsener, B., Zimmermann, L., Brchler, D. and Bhni, H., Repair
of reinforced concrete structures by electrochemical techniques Field experience, in Corrosion of Steel in Concrete, ed. J. Mietz,
B. Elsener and R. Polder, EFC Publication No. 25, The Institute of
Materials, London (1998) 125-140.
[21] Elsener, B., Half-cell potential mapping to assess repair work
on RC structures, Constr. Build. Materials 15 (2001) 133-139.
[22] Elsener, B. and Bhni, H., Electrochemical methods for
inspection of reinforcement corrosion in concrete structures - Field
experience, Materials Science Forum 111/112 (1992) 635-64.

Shift in potential
The shift in potential (before and after repair) depends on
the type of repair technique applied and its mechanistic action.
Half-cell potential mapping can be used to evaluate the
effectiveness and durability of repair works with the normal
interpretation only when the concrete resistivity is not changed
too much by the treatment.
o

Homogeneity of potential field

More important than the change in the absolute value of
the potential is that after a successful repair a uniform
potential field without hot spots shall be obtained.
7.2.4 Design of anode layout of CP or electrochemical
repair systems
The distribution of current from a homogeneous anode
on the concrete surface is governed by the concrete
resistivity and the corrosion state of the rebars:
o

Corrosion state
Corroding areas of the structure require more protection
current than areas with passive steel, due to the lower
polarization resistance the current will flow preferentially in
the corroding regions.
o

Concrete resistivity
The protection current will follow the low resistive
pathways, thus in areas with very humid or wet concrete
will flow more current.
Based on half-cell potential maps the corroding zones can
be identified and the layout of the anodes can be optimized:
critical areas can be protected by isolated anodes and
connected separately to the rectifier or the amount of anode
material (two or three layers of anode mesh) can be increased in
order to comply with the high current density expected. The
final draft version of European Standard EN12696 states that
potential mapping must be carried out on representative areas, in
particular for locating spots for reference electrodes to be placed.

8. REPORT
-

A report on half-cell potential measurements should contain:

date of testing, weather conditions (temperature,
humidity) at the time of testing and preferably over a
few days before testing
description of the structure, any visual observations at the
concrete surface (spalling, rust staining, exposed rebars etc.)
reference electrode used and calibration data
measuring grid used
results of half-cell potential measurements (preferably
in form of potential map with clear and scaled relation
to the structure)
chloride or carbonation analysis (if available).

REFERENCES
[1] RILEM TC 60 CSC, Corrosion of Steel in Concrete, ed. P.
Schiessl, Chapman and Hall, London (1988).
[2] Page, C.L. and Treadaway, K.W.J., Aspects of electrochemistry
of steel in concrete, Nature 297 (1982) 109-115.

471