RELATIONSHIP BETWEEN CRACKING AND ELECTRICAL

RESISTANCE IN REINFORCED AND UNREINFORCED CONCRETE.

J. Pacheco
(1)
, B. avija
(1)
, E. Schlangen
(1)
and R.B. Polder
(1,2)
(1) Faculty of Civil Engineering and Geosciences, Delft University of Technology, The
Netherlands
(2) TNO Technical Sciences / Building Materials, The Netherlands



Abstract
Electrical resistivity of concrete is a material dependent property that has been used to
assess transport properties. Ions contained in the pore solution are responsible of carrying
electrical current through concretes pore network. Pore size and connectivity have an
influence on the transport properties of concrete. When concrete is subjected to environmental
and/or mechanical action, cracking of concrete cover is a serious threat for its durability.
Cracks represent fast routes for chloride ingress that lead to corrosion of reinforcement. In
concrete, cracks can be considered as spatial discontinuities between aggregates and concrete
hardened matrix. For concrete infrastructure, cracks are important for assessing the remaining
service life. Tensile cracks in bending have a tapered geometry, i.e. the widest crack occurs at
concretes surface with a decreasing opening as crack propagates towards the reinforcement.
This type of crack is critical for assessing the condition of a concrete structure. In this paper,
real-time measurements of concrete resistance during cracking of reinforced concrete
specimens were performed under laboratory conditions. A Modified Wedge Splitting Test
(MWST) was used in order to create controlled cracks in the specimens. Results show that the
load-displacement curves can be linked to the relationship between displacement and
resistance of concrete.

1. INTRODUCTION

Reinforced concrete is the most widely used construction material in the world. Because of
its cost-efficiency and capacity for withstanding severe environmental conditions, concrete
infrastructure is required. However, concrete durability is not permanent. Concrete is a porous
material that allows transport of matter through its pore structure. The connectivity and size of
pores have a significant influence on transport properties in concrete. Ions contained in the
pore solution are transported by the action of some mechanisms like diffusion, capillary
suction or convection. Also, these ions can carry electrical current through concretes pore
network. Electrical resistivity of concrete is a material dependent property that has been used
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands

to assess transport properties of concrete. Measured values are usually between 10
1
to 10
5
m
and are dependent on concrete composition, age and environmental conditions [1]. The
chemical composition of concretes pore solution depends on the cement type and
composition and the presence of other binders like GGBFS or fly ash. Usually, the variation
of measured values of resistivity from different concretes is high. The influence of both mix
de-sign parameters and/or environmental conditions are the cause of such variation. High
values of water-to-binder ratio are related to wider and more connected pores. In this sense, it
is reasonable to assume that resistivity and durability are strongly related. A higher resistance
to chloride or CO
2
penetration; and slower deterioration (i.e. corrosion rate) should be
expected when high values of resistivity are found. This relationship between resistivity and
durability performance has been investigated extensively but remains still under debate [2-4].
Measurements of on-site resistivity have been used to assess the in-situ resistance to chloride
penetration [5] or as a quality control assessment [6] show other applications of concrete
resistivity.
One of the most inherent deterioration aspects of concrete is the presence of cracks. The
causes for concrete cracking are numerous during service of concrete infrastructure. When
concrete is subjected to environmental and/or mechanical action, cracking of concrete cover
can occur. Cracks represent fast routes for chloride ingress that leads to corrosion of
reinforcement. In concrete, cracks can be considered as spatial discontinuities of the ITZ
(Interface Transition Zone) between aggregates and concrete hardened matrix. Research
focused on the transport properties of cracks has revealed their strong influence on concrete
durability. A review on cracking methods in laboratory, the influence of cracks on chloride
penetration [7] and its influence on reinforcement corrosion [8] has been published recently.
Corrosion of reinforcement is also affected by cracking of concrete cover. For concrete
infrastructure, cracks are important for assessing the remaining service life. Tensile cracks in
bending have a tapered geometry, i.e. the largest displacement occurs at concretes surface
with a decreasing opening as crack propagates towards the reinforcement. This type of crack
is critical for assessing the condition of a concrete structure.
Schiel and Raupach [9] showed that transverse cracks induce a macro-cell on corrosion of
reinforcement, in which reinforcement intersected by cracks acts as anode. These findings
were later investigated [10,11] with similar results. A study carried out by Boulay et al. [12],
showed that the electrical resistance of concrete decreased as cracks occurred in disks subject
to Brazilian splitting. Each disk was placed between two reservoirs that released a solution
into the crack opening as cracking occurred. Measurements of resistance, and its inverse
conductance, showed a linear dependence on COD (Crack Opening Displacement). As the
crack became wider, more solution was allowed to fill the void, thus increasing the current
flow. A study focused on the permeability of cracks found a distinction between parallel-wall
cracks obtained from Brazilian splitting and V-shaped cracks from bending [13].
In this paper, real-time measurements of concrete resistance of reinforced concrete
specimens were performed under laboratory conditions. A Modified Wedge Splitting Test
(MWST) was used in order to control the crack opening in the specimens. By means of this
test, a V-shaped crack that is representative of those found in structures was obtained. Results
show that the load-displacement curves can be linked to the relationship between
displacement and resistance of concrete.
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands


2. EXPERIMENTAL DESIGN

2.1 Concrete mix design and specimen fabrication
Reinforced concrete specimens were fabricated in standard cubic moulds of 150x150x150
mm
3
. Table 1 shows the materials and mix design for the concrete cubes. In total, 6 specimens
were tested: four with reinforcement and two without. Two reinforcing bars of 120 mm in
length were placed at the bottom of the mould as shown in Figure 1. Prior to casting the
specimens, a steel profile with a cross section of 40x40 mm
2
is mounted on the mould, with
two resistance electrodes attached to the sides to obtain the target geometry. The electrodes
were made from a strip of titanium mesh covered with non-conductive coating, leaving the tip
(2 mm in length) uncoated. Electrodes RA were placed at the level of the notch in all 6
specimens, while electrodes RB were only placed in unreinforced specimens, at the same
position as the reinforcement in the reinforced specimens. The horizontal distance between
the electrodes was 40 mm. After around 5 hours, the steel profile is removed.

Table 1. Concrete mix composition.
Material Quantity, kg/m
3
Portland Cement CEM I 52.5 R 368
Water 165.5
Fine aggregate (<4mm) 840
Coarse aggregate (4-16mm) 1027.5
Superplasticizer 5.1

a) b)

Figure 1. Reinforced concrete specimen (a) and tapered crack with reinforcement (b).

After 24 hours, the specimens were stored in a curing (fog) room (20C, >90% RH) for 27
days. At the end of the curing period, a notch (5 mm thick, 15 mm in depth) was sawn in the
specimen using a water-cooled diamond saw. The purpose of the notch is to guide the fracture
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands

process; namely, the crack should typically start from the notch. The depth of this notch can
also be adjusted in order to achieve certain concrete cover depth.

2.2 Cracking and resistance measurements
At the age of 36 days, cracking procedure was performed. In order to control the obtained
crack width, LVDTs are placed on both sides of the specimen at the bottom of the notch,
where the crack is expected to initiate. Their average is used as a feedback signal for the ma-
chine, and controls the whole loading process. In this way, a stable cracking procedure is
performed. The COD (Crack Opening Displacement) is the measurement of the horizontal
displacement by the LVDT, which is related to the opening of the crack itself.
When the desired crack width has been achieved, hard plastic wedges are placed inside the
notch, in order to reduce the crack closing (recovery) after unloading. That way, the final
crack width after unloading, i.e. the crack width used as an input in the durability study, is
closer to the final crack width obtained in the test. This is important since, even though the
crack somewhat closes upon unloading, the area around the tip of the crack contains some
microcracks which do not vanish when the specimen is unloaded, and are more permeable
than sound concrete. These plastic wedges were able to reduce the crack recovery, but not
fully: the ideal solution would be to sustain the load during the subsequent durability
investigation, which is not possible due to the expected duration of the test.
Measurements of electrical resistance were carried out at the same time as the loading. For
this, a LCR meter was used and the electrical resistance (120 Hz AC) was stored with a scan
rate of 1 measurement per second. For the reinforced specimens, resistance measurements
were performed over the electrodes RA: both at the level of the notch. For the unreinforced
specimens RA and RB electrodes were measured.

3. RESULTS AND DISCUSSION
Figure 2 shows the results of the load-displacement curves obtained from the MWST and
the behaviour of resistance during the measurement of a reinforced specimen (Specimen 1).
With regard to the load-displacement curve Fig 2.a, results show that a maximum load of
nearly 2 kN is reached during the test. At this loading condition, the average displacement of
the LVDT is about 50 m. Some variations in resistance were found during the initial loading
part of the curve (COD < 50 m), which can be attributed to the formation of microcracks. As
the load is reaching the maximum level, these variations are more evident. This behaviour
corresponds then to the propagation of the crack towards the reinforcement. After reaching its
maximum point (the softening part of the curve), a combination of changes in both resistance
and applied load are observed. The behaviour in this part of the curve may be attributed to the
development of secondary cracks (debonding at the reinforcement interface). Finally, the part
between 150 m and 300 m is believed to only increase the width of cracks, but not their
depth. After the maximum displacement has been reached, the unloading of the specimen is
carried out. After recovery the crack width at the notch tip was about 180 m.
In the same figure, the relationship between COD and electrical resistance is shown. When
cracking occurs, the current flow is disturbed by the presence of a non-conductive medium
(crack void). In the period below the maximum load (COD ~ 50 m), there is almost no
variation in resistance. However, after the maximum load is reached, there is a sudden
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands

increase in resistance that corresponds to the appearance of the first crack. This sudden
alteration on resistance is more evident in Figure 2.b After this, a gradual increase in
resistance is observed (COD 50 to 110 m), followed by another sudden increment. This
behaviour is repeated until the maximum resistance is obtained. This behaviour was also
found in Figure 2.b: sudden increments of resistance followed by gradual ones. It seems that
two distinct processes are occurring: gradual increase in resistance corresponding to crack
opening (increase in width); whereas sudden increase to crack propagation (debonding). Since
cracks obtained from the MWST had a V-shape (see Figure 1.b), the electrical resistance
increases as the cracked zone increases. Overall, its clear that there is a linear relationship
between cracking and the resistance. The equation that describes this relationship has the form
of:
y = ax+b (1)
where y is the measured electrical resistance (k), a is a parameter that describes the size
and connectivity of the pore structure, x is the disruption of the pore network (cracks) and b is
the initial resistance (k). There is a good correlation factor (R
2
) between load and resistance
in this curve. This linear relationship corresponds to that found by Boulay [12].


a) b)
Figure 2. Load-COD-Resistance (a) and Load-Resistance (b) curves of reinforced concrete.

Results of load-displacement and resistance for the four reinforced specimens are shown in
Table 2. In all four cases, the maximum load is on the range between 1.9 and 2.3 kN. The
COD at the maximum load was found in the range between 20 and 50 m. It seems that there
is no direct correlation between the maximum load and the COD: the maximum COD was
found at the lowest load and the maximum load had a narrow COD. For cracks below 300
m, the recovery after unloading was found to be above 40% or more. It seems logical that
wider cracks correspond to permanent damage that cannot be recovered during unloading
(Specimen 3, 20% of recovery). As for resistance increase, in three cases it was found to be
below 10%. Only in Specimen 3 the increase was above this level (15%). It seems to be that
as the damage becomes permanent, the increase in resistance will be higher.
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands


Table 2. Results of load-displacement and resistance of reinforced specimens.
Spec. Max.
Load
(kN)
COD
Max
load
(m)
Max.
COD
(m)
Final
COD
(m)
Recovery
(%)
Initial
Resistance
(k)
Final
Resistance
(k)
Increase
(%)
R
2

1 1.93 50 300 180 40 6.4 7.0 9 0.94
2 2.02 30 200 120 40 7.1 7.4 4 0.94
3 2.26 27 500 400 20 6.2 7.1 15 0.97
4 2.04 23 150 80 47 5.7 6.2 8 0.83

Figure 3 shows the results of the load-displacement and resistance curves of an
unreinforced specimen. Figure 3.a shows that the maximum load is similar to those found on
reinforced specimens. However, it seems that for unreinforced specimens variations in the
load during the loading period (COD < 50 m) are absent. A reason for this behaviour is that
there are no grip disturbances between reinforcement and concrete. After the maximum load
has been reached, the softening part of the curve is similar as in Specimen 1. Also, the
recovery had roughly the same behaviour as in the reinforced case.


a) b)
Figure 3. Load-COD-Resistance (a) and Load-Resistance (b) curves of unreinforced
concrete. RA (R
2
= 0.98) is shown in both 3.a and 3.b . RB (R
2
= 0.25) only in curve in Fig
3.a

For resistance measurements, electrodes RA showed a behaviour somewhat similar to
Specimen 1, while electrode RB remained undisturbed over the whole measuring period. This
means that even after the maximum cracking occurs, the disturbed zone does not reach 100
mm from the top of the reinforcement. It seems that the first sudden increase in resistance did
not correspond to the displacement at the maximum load, but occurred at about 45 m. This is
also observed in Figure 3.b, the increase in resistance for electrode RA occurs when the load
in the specimen has already decreased. Compared to Specimen 1 the sudden changes in
resistance, which were attributed to crack propagation, were not so evident in unreinforced
concrete. In this case, the crack propagation and crack opening are closely related probably
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands

occurring simultaneously. It seems that in reinforced concrete, the crack propagates towards
the reinforcement almost immediately, whereas in unreinforced penetrates slowly. This was
also found by Pease [11]. For electrodes RA the relationship between the electrical resistance
and COD had the same linear dependency as in reinforced concrete. Table 3 shows the results
of load-displacement and resistance in unreinforced specimen.

Table 3. Results of load-displacement and resistance of unreinforced specimens.
Spec. Max.
Load
(kN)
COD
Max
load
(m)
Max.
COD
(m)
Final
COD
(m)
Rec.
(%)
Initial
RA
(k)
Final
RA
(k)

(%)
R
2
Initial
RB
(k)
Final
RB
(k)

(%)
R
2

5 1.90 30 300 198 34 5.9 7.0 18 0.98 3.7 3.7 0 0.25
6 2.28 35 400 300 25 7.5 9.0 20 0.98 7.1 7.1 0 0.93

Results of unreinforced specimens show that for electrodes at the level of the notch (RA),
the increase in resistance was up to 20%. This is higher than in reinforced specimens. It seems
that unreinforced concrete is subject to higher deterioration from wide cracks (above 300 m).
For resistance at 100 mm deep (RB) there is no influence on resistance, meaning that for a
crack width of 300 m at the level of the notch the crack depth has not reached 100 mm from
the top of the specimen. In other words, cracking of unreinforced concrete occurs from the
combination of crack opening and propagation acting simultaneously. In the case of
reinforced specimens, this behaviour is not expected because cracks propagate first towards
the reinforcement and along it (Figure 1.b).

3.1 Forthcoming work

Future research on the relationship between concrete durability and cracks involve the
exposure and monitoring of transport properties, corrosion deterioration and different
materials (particularly CEM III/B). Research focused on studying the debonding of concrete
and steel will be included in future publications.
4. CONCLUSIONS

Reinforced and unreinforced concrete specimens were subject to loading in a Modified
(MWST) and ordinary Wedge Splitting Tests (WST). Displacement and electrical resistance
between two electrodes were monitored during loading. Results showed that resistance
variations found in reinforced concrete were attributed to crack opening and propagation,
independently. For unreinforced concrete, these two mechanisms were found to be acting at
the same time, leading to smaller variations of resistance and loading during the experiment.
The most important findings include:
1. Maximum load in reinforced and unreinforced specimens was about 2 kN. There was
no influence of reinforcement on the load at cracking
2. Electrical resistance showed a linear relationship with COD at the level of the notch
(electrodes RA) for reinforced and unreinforced concrete; no effect on resistance was
found at 100 mm deep (unreinforced, electrodes RB).
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands

3. For reinforced concrete, sudden increments of resistance are probably attributed to
crack propagation towards and along the reinforcement. This will be confirmed with
further experiments.
4. In unreinforced concrete, electrical resistance did not show sudden increments but a
rather gradual behaviour.
5. Less recovery of COD at unloading was found in unreinforced specimens. Also, in the
same specimens a higher increase in resistance was found near the crack opening.
6. Monitoring electrical resistance provides a useful way of characterising cracking in
concrete.

ACKNOWLEDGEMENTS
Financial support by the Dutch Technology Foundation (STW) for the project 10978-
Measuring, Modelling, and Monitoring Chloride ingress and Corrosion initiation in Cracked
Concrete (M3C4) is gratefully acknowledged.
REFERENCES
[1] Polder, R.B. 2001. Test methods for on-site measurement of concrete resistivity a RILEM TC-
154 technical recommendation, Construction and Building Materials, 15(3): 125-131.
[2] Gulikers, J. 2005. Theoretical considerations on the supposed linear relationship between concrete
resistivity and corrosion rate of steel reinforcement. Materials and Corrosion 56(6): 393-403.
[3] Raupach, M. 2006. Models for the propagation phase of reinforcement corrosion - an overview.
Materials and Corrosion 57(8): 605-613.
[4] Andrade, C. 2010. Types of models of service life of reinforcement: the case of resistivity.
Concrete Research Letters 1(2): 73-80.
[5] Bryant Jr., J.W., Weyers, R.E., de la Garza, J.M. 2009. In-place resistivity of bridge deck concrete
mixtures. ACI Materials Journal, 106(2): 114-122.
[6] Rooij, M. de, Polder, R.B., Oosten, H. van, 2007, Validation of durability of cast in situ concrete
of the Groene Hart railway tunnel, HERON Vol. 52 (4), 225-238.
[7] avija, B. & Schlangen, E. 2011. Chloride ingress in cracked concrete a literature review. In C.
Andrade, J. Gulikers (eds.), Advances in Modeling Concrete Service Life: Proc. Of the 4
th

International RILEM PhD Workshop, Madrid, Spain, 19, November 2010. New York: Springer.
[8] Pacheco, J. & Polder, R.B. 2011. Corrosion initiation and propagation near cracks a literature
review. . In C. Andrade, J. Gulikers (eds.), Advances in Modeling Concrete Service Life: Proc. Of
the 4
th
International RILEM PhD Workshop, Madrid, Spain, 19, November 2010. New York:
Springer.
[9] Schiel, P. & Raupach, M. 1997. Laboratory studies and calculations on the influence of crack
width on chloride-induced corrosion of steel in concrete. ACI Materials Journal, 94(1): 56-60.
[10] Otieno, M.B., Alexander, M.G., Beushausen H.D. 2010. Corrosion in cracked and uncracked
concrete influence of crack width, concrete quality and crack reopening. Magazine of Concrete
Research, 62(6): 393-404.
[11] Pease, B.J., 2010. Influence of concrete cracking on ingress and reinforcement corrosion, PhD
Thesis, Technical University of Denmark.
[12] Boulay, C., Dal Pont, S., Belin, P. 2009. Real-time evolution of electrical resistance in cracking
concrete, Cement and Concrete Research 38(9): 825-831.
[13] Wang. K., Jansen, D.C., Shah, S.P. 1997. Permeability study of cracked concrete. Cement and
Concrete Research, 27(3): 381-393.
Second International Conference on Microstructural-related Durability of
Cementitious Composites, 11-13 April 2012, Amsterdam, The Netherlands