Construction and Building Materials 173 (2018) 368–380

Contents lists available at ScienceDirect

Construction and Building Materials

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

Rapid repair of severely damaged reinforced concrete columns under

combined axial compression and flexure: An experimental study
M. Chellapandian, S. Suriya Prakash ⇑
Department of Civil Engineering, Indian Institute of Technology Hyderabad, India

h i g h l i g h t s

Effectiveness of different repair schemes under combined compression and flexure is studied.

Repair effectiveness of quickset cement mortar, NSM and hybrid FRP technique is studied.

All the repaired columns are tested under concentric and eccentric compression (e/d = 0.63).

The effectiveness of repair is estimated through stiffness, strength and displacement indices.

Hybrid FRP technique fully restored the strength and ductility of columns.

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

Article history: Reinforced concrete (RC) columns are in general subjected to combination of axial compression and
Received 17 January 2018 bending loads. A rapid repair technique for effectively restoring the original strength and stiffness of
Received in revised form 26 March 2018 severely damaged RC columns under combined bending and compression loads is investigated. In total,
Accepted 4 April 2018
six RC columns are tested under uniaxial eccentric compression at an eccentricity (e) to depth ratio (d) of
0.63. The columns are initially subjected to severe damage corresponding to a concrete compressive
strain (ec) of 0.0035. Damaged columns had buckling of longitudinal reinforcement and concrete crush-
Keywords:
ing. After severe damage, the columns are repaired with three different techniques using (i) quick set
CFRP
Concentric compression damaged column
cement grout mortar (QS), (ii) near surface mounting (NSM) of CFRP laminates and (iii) hybrid FRP
Eccentric compression strengthening using a combination of NSM and external bonding (EB) of CFRP fabric. The specimens
Hybrid FRP technique are tested under high ratio of eccentric compression (e/d = 0.63) in a displacement controlled mode.
Rapid repair The performance of different repair techniques under eccentric compression is compared with the similar
RC column specimens tested under concentric compression. Test results revealed that hybrid FRP strengthening is
very efficient in fully restoring the original strength and failure displacement under combined compres-
sion and bending loads. Only NSM strengthening is found to be very efficient under eccentric compres-
sion than in concentric compression.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction extreme loading conditions [7–9]. Moreover, the immediate occu-

Damage of reinforced concrete (RC) columns under earthquakes
and in accidents due to fire and blast loading can lead to catas-
trophic failure of the entire RC structure [1–4]. Columns are the
main load bearing elements in the structural system and cannot
undergo severe damages in order for the buildings to remain func-
tional [5,6]. The current design code allows limited damages in RC
columns such as cracking and local concrete spalling under
⇑ Corresponding author.
E-mail addresses: ce15resch11005@iith.ac.in (M. Chellapandian), suriyap@iith.
ac.in (S.S. Prakash).
pancy after earthquakes is necessary for certain important build-
ings. Thus, it is necessary to restore the original capacity of
columns in a quick time before adopting global repair solutions.
The repair technique adopted can be deemed to be successful only
if the columns can be brought to service in a short time after severe
damage. Conventional strengthening strategies such as steel jack-
eting and concrete enlargement has been used in the past by many
researchers [10–13]. However, their high dead weight and
increased cost of labour have made FRP strengthening as effective
repair solution [14,15]. Various FRP strengthening techniques such
as external bonding (EB) [16–22], near surface mounting (NSM)
[23–31] have been used to improve the behavior of RC elements
under different loading conditions. However, only limited studies

https://doi.org/10.1016/j.conbuildmat.2018.04.037
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380 369

are available in the literature on evaluation of repair effectiveness
of RC columns using different FRP techniques under different com-
binations of bending and compression loading.
The effectiveness of FRP confined columns for restoring the
original capacity of RC columns under concentric compression
has been investigated by some researchers [32–39]. They con-
cluded that the FRP confinement helps in complete restoration of
the original capacity without much reduction in ductility under
different levels of compression. Elkin et al. [40] investigated the
seismic repair efficiency of FRP confined columns with moderate
and severe damage. They found that the repair using CFRP confine-
ment restored the strength and displacement capacity of moder-
ately damaged columns whereas the displacement capacity of
severely damaged columns is not restored. Saadatmanesh et al.
[41] studied the flexural repair efficiency of columns using CFRP
wrapping. They found that the repair with CFRP wrapping
improved the hysteretic behavior of columns and completely
restored the original strength and ductility. Repair efficiency of
CFRP wrapping technique for RC columns under torsion, combined
torsion and bending loads have been recently investigated by some
researchers [42–45]. They concluded that the damaged columns
repaired using CFRP wrapping technique performed better than
the original specimens. Seismic repair of RC bridge columns using
CFRP shells and epoxy anchored headed steel bars is carried out
recently [46]. The performance of column elements under com-
bined axial and bending loads has also been extensively studied
[47,48]. From the research work carried out in the past, it is clear
that the repair efficiency of severely damaged columns using FRP
under combined axial and bending loads has not been adequately
investigated. The present study focusses on filling the knowledge
gap on the effectiveness of different repair strategies for severely
damaged RC columns under combined compression and bending.

Fig. 1. Schematic details of RC column elements.

Table 1
Details of test matrix.

Specimen ID Dimensions (mm) Repair Technique Used

b d h
RCC-0 230 230 450 –
RCC-R-QS-0 Quick Set High Strength Grout
RCC-R-NSM-0 NSM CFRP Laminates
RCC-R-HYB-0 CFRP Laminates + CFRP Fabric
RCC-145 230 230 600 –
RCC-R-QS-145 Quick Set High Strength Grout
RCC-R-NSM-145 NSM CFRP Laminates
RCC-R-HYB-145 CFRP Laminates + CFRP Fabric
**
Note: The values 0 mm and 145 mm are used to represent the level of eccentricity in the test specimen. The details of test specimen under concentric compression are used
only for comparison purposes [32].
370 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380

Fig. 2. Test setup and instrumentation details under eccentric loading. 1. Test Specimen 2. LVDT for Measuring Strain Gradient Effect 3. LVDT for Axial Displacement 4. LVDT
for Lateral Displacement 5. DAQ 6. DAQ Operating System. 7. CTM Operating System.

2. Research motivation

NSM strengthening performs well under flexure dominant load-

ing. Its performance can be improved further when their debond-
ing mode of failure is prevented by lateral confinement. The past
studies on hybrid FRP strengthening using a combination of NSM
CFRP laminates and EB CFRP fabric is shown to improve perfor-
mance of undamaged columns under compression and bending
loads [49–52]. The objective of the research work is to study the
effectiveness of different repair techniques. The columns are tested
under eccentric compression loading (e/d = 0.63). The performance
of these specimens are compared with the columns tested under
concentric compression to illustrate the effects of additional bend-
ing due to eccentric compression.

3. Experimental program

3.1. Details of test specimen Fig. 3. Load–displacement behavior comparison of RC columns.

The specimens are tested under uniaxial eccentric compression loading (e/d =
0.63). The chosen e/d ratio is expected to result in a balanced failure condition in
the axial compression and bending interaction diagram. The test specimens are
square in cross section of 230 mm dimension. The test zone at the centre is of
600 mm height (Fig. 1). Eight bars of 12 mm diameter are provided as the longitu-
dinal reinforcement. The ratio of longitudinal reinforcement ratio is 1.73%. Trans-
verse reinforcement of 10 mm diameter with volumetric reinforcement ratio of
0.89% is provided at a spacing of 100 mm c/c. Both the longitudinal and transverse
reinforcement ratio considered in the investigation is typically used in mid-rise
building columns. Some additional reinforcements are provided in the top and bot-
tom to avoid cracking and premature failure of specimens. The details of the test
matrix are provided in Table 1.
3.2. Material properties
in this study are of Fe500 grade. The mechanical properties such as yield strength,
ultimate stress and failure strain are found to be 512 MPa, 620 MPa and 7.8%,
respectively.
3.2.2. FRP properties
CFRP laminates of size 12.5 mm  4.2 mm is used for NSM repair technique. The
mechanical properties of laminates are found by testing the coupon specimens
under uniaxial tension and compression [54,55]. The tensile strength and elastic
modulus of CFRP are found to be 2300 MPa and 150 GPa, respectively. Uni-
directional CFRP fabric of 230 g per square meter (GSM) and 0.2 mm thickness is
used for EB strengthening. From the coupon test results, the tensile strength and
failure strain of CFRP fabric are found to be 1400 MPa and 1.3%, respectively. The
detailed description of the tensile testing of CFRP materials under tension and com-
pression loading is documented elsewhere [32,49].

3.2.1. Concrete and steel reinforcement
All the test specimens are designed for M40 design mix as per IS10262-2008
[53] to have a target cubic compressive strength of 48.25 MPa. The proportion of
concrete mix used is as follows: cement – 360 kg/m3; fine aggregate (sand) –
752 kg/m3; coarse aggregate – 1196 kg/m3 and super plasticizer (admixture) –
3.6 kg/m3, water – 144 kg/m3. Auxiliary specimens such as cubes and cylinders
are also cast to determine the compressive strength and stress-strain properties
of concrete after 28 days of water curing. Material characterization of concrete
and steel specimens are performed and minimum of five samples are used to char-
acterize the properties of these different materials used in this study. The average
cubic and cylinder compressive strength of concrete are found to be 55 MPa and
42 MPa, respectively. The longitudinal and transverse steel reinforcements used
4. Test setup and instrumentation details
The test setup and complete instrumentation details for RC ele-
ments under high eccentric compression is shown in Fig. 2. The
columns are tested using a servo controlled compression testing
machine (CTM) of 3000 kN capacity. Complete details of columns
tested under concentric compression is documented in the previ-
ous work of authors [32]. The columns under eccentric compres-
sion are tested under displacement control mode at a loading
rate of 0.03 mm/s. The bottom support is movable (free to deform
 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380
Fig. 4. Failure mode observation for original RC columns.
vertically) and the top support is fixed (only vertical displacements
are arrested). Two linear variable displacement transducers
(LVDTs) of 20 mm stroke are used for measuring the axial displace-
ments and strains. In addition, two LVDTs of 20 mm stroke are
used on the tension and compression face of the specimens to mea-
sure the strain gradient and is subsequently used to calculate cur-
vature. Two LVDTs of 50 mm stroke are used at the mid-span to
measure the lateral (transverse) displacement. Strain gauges of
120 X resistance installed on the longitudinal reinforcements are
used for observing the strain variation in the reinforcement with
the increase in load. All the strain and displacement data obtained
during the testing of columns are recorded using a HBM data
acquisition system (DAQ).
5. Tests to induce severe damage on columns
The columns without any strengthening are loaded under
uniaxial eccentric compression with an eccentricity of 145 mm
(e/d = 0.63). Out of six specimens, one specimen is tested until fail-
ure (complete collapse) and the remaining five specimens are loaded
until drop in peak load of up to 20% or minimum concrete compres-
sive strain of 0.0035. All the specimens reached concrete compres-
sive strain of 0.0035. All the columns are severely damaged at the
end of testing. Longitudinal bars just yielded on the compression
side. Buckling of longitudinal bars is also observed. The columns
had severe spalling and concrete crushing in the compression face.
The average load–axial displacement behavior of RC elements under
high eccentric compression is compared with the similar specimens
tested under concentric compression (e/d = 0) in Fig. 3. Due to eccen-
tricity level of 145 mm, peak strength of the column is only about
33.4% of similar specimen under concentric compression (RCC-0).
The failure mode of original RC columns (RCC-145) under high
eccentric compression is shown in Fig. 4.
6. Repair strategy for damaged RC columns
The column specimens that are severely damaged under eccen-
tric compression loading are repaired employing three techniques.
No reinforcements were treated/repaired after damage in order to
showcase the effectiveness of repair techniques for columns with
371
the reinforcements in damaged condition. It will be interesting to
consider the repair or treatment for these bars. It is scope for fur-
ther work. All the damaged specimens are initially repaired using
quick set cement grout mortar. Thereafter, two damaged columns
are repaired with NSM technique (RCC-R-NSM) and the remaining
two with hybrid FRP technique (RCC-R-HYB). In the past, numer-
ous work has focused on the behavior of FRP confined columns
and therefore repair with only FRP confinement is not carried out
in this study. Moreover, FRP confinement is expected to show only
marginal strength improvement under combined compression and
bending loads [18–20].
6.1. Repair of damaged columns using high strength cement grout
Step by step repair procedure for QS technique is depicted in
Fig. 5. After initial damage, the specimens are straightened by
applying transverse load. Thereafter, severely damaged locations
with excessive cracking and concrete cover spalling are chipped
off to obtain a strong substrate. Crack injection is not performed
since the cracked concrete is completely removed in the tension
face. The bond between the old and new concrete is ensured by
the application of bonding agent (araldite). To measure the contri-
bution of longitudinal reinforcement in load resistance, additional
strain gauges are installed in the reinforcement (Fig. 5(ii)). There-
after, the cement mortar of high strength and good bonding capac-
ity is prepared and poured [56]. The repaired specimens are water
cured for a minimum period of three days to achieve sufficient
strength. The compressive strength of the grout material on the
three, seven and twenty-eight days are 34 MPa, 45 MPa and 60
MPa, respectively. In the companion study of authors, bond
strength of high strength grout is investigated by embedding CFRP
laminates in the test material. The bond strength of high strength
mortar is found to be 4.5 MPa [56].
6.2. Repair using near surface mounting of CFRP laminates
Out of five specimens repaired using cement mortar, NSM
repair procedure is carried out for two column specimens as per
ACI 440 provisions [14]. The dimensions of the grooves [20 mm
deep  8 mm wide] are kept at least 1.5 times the dimensions of
372 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380
Fig. 5. Repair procedure for columns using high strength cement mortar.
the NSM laminates. The choice of providing NSM is made based on
the previous study from the authors [49]. The idea behind the
selection of particular NSM reinforcement ratio is to provide equal
distribution of about 1% cross sectional area of laminates as addi-
tional longitudinal reinforcement. To avoid edge failure in NSM
strengthened columns, a minimum edge distance is provided in
the columns as per ACI 440.2R provisions [14]. The grooves are
cut, cleaned thoroughly and primer is applied for better bonding.
Thereafter, the grooves are half filled with epoxy resin and the
NSM laminates are pressed against the epoxy. Once the CFRP lam-
inates are set in position, the grooves are completely filled with the
epoxy and the surface is finished (Fig. 6). The minimum curing per-
iod for epoxy to gain sufficient bond strength is about 48 h [49,57].
6.3. Repair using hybrid CFRP technique
In this technique, the NSM CFRP laminates are confined using
externally bonded (EB) CFRP fabric. The reason for providing this
lateral confinement is to prevent NSM CFRP laminates from buck-
ling or debonding under concentric or eccentric compression load-
ing. In eccentric compression, due to significant contribution of
NSM laminates in the tension face, there is high chance of lami-
nates to de-bond from the groove at peak load. Moreover, the lam-
inates in the compression face can buckle due to no restraints
available to support. This can be prevented in the hybrid repair
technique enabling the columns to undergo large strain before fail-
ure. Hybrid FRP strengthening is carried out as per ACI 440.2R pro-
visions [14]. The repaired columns are initially strengthened using
NSM technique after which they are confined using EB CFRP fabric.
Two layers of CFRP wrapping are provided for better confinement.
This is selected based on the previous studies in which the authors
performed a detailed parametric analysis [51]. It is found that FRP
confinement of concrete with more than three layers of wraps are
found to be less effective for improvement in the load carrying
capacity under concentric and eccentric compression. Moreover,
the strength improvement depends on a saturation point more
than which the increase in area of the material will not provide
any improvement in the strength [18–20]. The corners of the
NSM repaired specimens are ground to a minimum radius of 15
mm [14]. The column surface is cleaned thoroughly and applied
with primer. After 24 h of minimum curing time, two component
based epoxy resin is applied (Base:Hardener = 2.8:1). The CFRP
wrapping is carried out segment by segment with the minimum
overlap length as per ACI 440.2R [14]. The specimens are
tested after 48 h of minimum curing time. The hybrid FRP repair
procedure is shown in Fig. 7. The schematics of hybrid FRP
 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380 373

(i) Grooving (ii) Primer Application (iii) Epoxy Application (iv) Repaired Specimen
Fig. 6. Repair procedure for columns using NSM technique.

Fig. 7. Repair procedure for columns using hybrid FRP technique.

strengthening for showing the position of laminates and wraps is
shown in Fig. 8.
7. Results and discussion
Test results of columns repaired after severe damage under con-
centric compression loading is used for comparison purposes to
illustrate the effectiveness of different repair techniques under
combinations of compression and bending loads. The axial dis-
placement measured from the specimens are used for comparing
the load–displacement behavior of columns under concentric and
eccentric loading conditions. The schematic representation of orig-
inal and repaired columns with different indices are shown in
Fig. 9. The effectiveness of different repair techniques is evaluated
in terms of three indices namely stiffness index (KI), strength index
(SI) and displacement index (DI) [58–62]. The relative effectiveness
of NSM and hybrid FRP techniques evaluated using percentage
increments in strength, stiffness and displacement indices are
summarized in Table 2.
Stiffness index (KI) is defined as the ratio of stiffness of repaired
specimen first 30% peak load to the stiffness of original specimen at
30% peak load as shown in Eq. (1).
Kr
KI ¼ ð1Þ
K0
Strength index (SI) is the ratio of peak strength of repaired spec-
imen to the peak strength of original undamaged specimen as
shown in Eq. (2).
374 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380
Fig. 8. Schematics of FRP strengthened columns. (**Note: Three CFRP laminates are provided in the compression face of the column as well similar to the tension face which is
not clear from this schematic view.)
Fig. 9. Representation of original and damaged columns for different indices.
Pr
SI ¼
P0
Displacement index (DI) is the ratio of displacement of repaired
specimen after 20% decay in peak load to displacement of original
specimen after 20% decay in peak load as shown in Eq. (3).
Dr
DI ¼
D0
7.1. Columns repaired using high strength cement mortar (RCC-R-QS)
The load displacement behavior of columns repaired using high
strength cement grout under concentric and eccentric compression
loading is compared in Fig. 10. In Table 2, the indices of RCC-0 and
RCC-145 are kept as 1.00 to understand the increase/decrease in the
effectiveness of repair specimen with respect to the original column
behavior. Due to initial damage, the peak strength of the repaired
columns is reduced by 80% and 74% under concentric and eccentric
compression, respectively. In terms of stiffness index (Kr), the
repaired columns with cement grout had stiffness reduction of up
to 64% when compared to their original specimen (RCC-0) under
concentric compression. However, the initial stiffness of similar
specimens under high eccentric compression loading (RCC-145) is
completely restored. The stiffness of damaged column is fully
restored as the original damage is localized due to strain gradient
effect under eccentric compression. However, complete section
undergoes damage under concentric compression loading. The
axial displacement of repaired specimen (RCC-145) reduces drasti-
cally to 59% of the undamaged columns under eccentric loading. For
similar specimens tested under concentric compression, the failure
displacement reduces only by 10% when compared to the original
column (RCC-0). The columns had reduced failure displacement
under eccentric compression specimens as it had localized failure
in the presence of additional bending moment.
7.2. Columns repaired using near surface mounting technique (RCC-R-
NSM)
The load–axial displacement behavior of columns repaired with
NSM CFRP laminates under concentric and eccentric compression
loading is compared with the original columns without any dam-
age (Fig. 11). The stiffness of NSM repaired columns under concen-
ð2Þ tric loading is restored only by 43% whereas the columns, under
eccentric loading had more than 100% restoration of stiffness
(Kr). This could be due to pre-existing cracks in the specimen
which helped the NSM laminates to effectively contribute to the
bending resistance right from the inception of applied loads. In
terms of peak load, NSM repair is able to completely restore the
ð3Þ strength under eccentric compression (102%). The peak strength
of two individual NSM repaired specimens are 824 kN and 717
kN respectively. The difference in peak load is due to difference
in the level of damages during initial damage mechanism. How-
ever, for the concentric compression specimens, the peak strength
is restored only to 83% when compared to the original columns
(RCC-0). This shows the NSM is more effective under bending dom-
inant loads compared to concentric compression. Ductility is an
important parameter in evaluating the effectiveness of any
strengthening technique. For NSM repaired columns under concen-
tric compression, the axial displacement is fully restored (100%).
However, it reduces to 57% for similar specimens under eccentric
compression (Fig. 11). This is due to sudden debonding of NSM
CFRP laminate and crushing of concrete soon-after reaching the
peak load.
 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380 375

Table 2
Experimental results of damaged RC columns with different repair techniques.

Loading Type Specimen Yield Yield Axial Peak Axial Failure Failure Axial
Load Disp. Stiffness Load Strength Load Displ. Displacement
(kN) (mm) Index (kN) Index (kN) (mm) Index
Concentric Compression (e/d = 0) RCC-0 2277 0.21 1.00 2418 1.00 1903 0.74 1.00
RCC-R-QS-0 1919 0.30 0.36 1955 0.80 1335 0.83 0.91
RCC-R-NSM-0 1975 0.30 0.43 2005 0.83 1359 0.85 1.00
RCC-R-HYB-0 2508 0.33 0.64 2591 1.06 1702 0.97 1.26
High Eccentric Compression (e/d = 0.63) RCC-145 676 0.22 1.00 760 1.00 608 0.65 1.00
RCC-R-QS-145 552 0.18 0.99 564 0.74 451 0.27 0.41
RCC-R-NSM-145 591 0.14 1.41 771 1.02 618 0.27 0.42
RCC-R-HYB-145 700 0.15 1.56 951 1.25 761 1.20 1.84
**
Note: The values represented correspond to average of two individual specimens. The comparison of different indices is made with respect to their original column specimen
(RCC). Yield load and its corresponding displacement are mentioned just to highlight whether the reinforcement reaches its yield strain value.

Fig. 10. Load-axial displacement comparison of columns repaired with QS Fig. 12. Load-axial displacement comparison of columns repaired with hybrid
technique. technique.

Specimen Set - I Specimen Set - II

2800
P a = 2418 kN P a = 2418 kN
2400
RCC-R-QS-0 RCC-R-QS-0
RCC-R-NSM-0 RCC-R-NSM-0
2000 RCC-R-HYB-0 RCC-R-HYB-0
Load (kN)

1600
RCC-R-QS-145 RCC-R-QS-145
RCC-R-NSM-145 RCC-R-NSM-145
1200
RCC-R-HYB-145 RCC-R-HYB-145

P e = 760 kN P e = 760 kN
800

400

0
0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0
Axial Displacement (mm)

Fig. 13. Overall load–displacement comparison of repaired RC columns. **Note:

Pa – Peak strength of original columns under concentric compression; Pe – Peak
strength of original columns under eccentric compression; Specimen set I and II
refers to comparison of two different specimens under same series.
Fig. 11. Overall comparison of columns repaired with NSM technique.

7.3. Columns repaired using hybrid FRP technique (RCC-R-HYB)
The load-displacement behavior of damaged columns repaired
with hybrid FRP technique is shown in Fig. 12. The stiffness of
repaired column under concentric compression is not fully
restored (64% of original column) due to rapid progression of
cracks and complete damage of the specimen. However, the
stiffness is completely restored (156% of original column) for
specimens tested under combined compression and bending
(e/d = 0.63) due to effective contribution of NSM laminates in crack
arresting on the tension side. The peak load capacity of columns is
completely restored for specimens under both loading conditions.
376 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380

The peak strength of two individual hybrid repaired specimens are

884 kN and 1018 kN respectively. The behavior of second specimen
strengthened using hybrid FRP is slightly different from the first
specimen in terms of post-cracking stiffness and post-peak behav-
ior. This is mainly due to difference in the level of damages
between them. Nevertheless, both the repaired columns attained
the original column capacity. The percentage of peak strength
restored for concentric and eccentric compression specimens are
106% and 125%, respectively of their original undamaged columns
(RCC). The important advantage of this hybrid strengthening is the
substantial improvement in strength without much compromise in
ductility under all combinations of bending and compression load-
ing. The failure displacement of columns under both loading condi-
tions are fully restored.

(i) Comparison of Axial Stiffness Index (KI) 7.4. Comparison of repair techniques under concentric and eccentric
compression

A repair scheme which is effective under a particular loading

(ii) Comparison of Axial Strength Index (SI)
may not exhibit similar performance under differing loading condi-
tions. The importance of this research work is to propose a unique
repair scheme that works under different combinations of bending
and compression loading. Test results from this study has proven
that hybrid FRP repair technique is efficient in overall performance
improvement of damaged columns under different combinations
of compression and bending loads. Hybrid repair technique com-
pletely restored the axial strength and displacement of columns
under both concentric and eccentric compression (Fig. 13). NSM
repair scheme is found to be highly effective when the members
are exposed to bending dominant behavior (under high eccentric
compression). NSM technique is effective in improving the overall
load carrying capacity of the specimen under eccentric compres-
sion load but resulted in brittle failure. As expected, the repair of
columns only with quickset cement grout is found to be the least
effective when compared to the other two repair techniques.

7.5. Comparison of different repair techniques

The effectiveness of different repair techniques for RC elements

under concentric and eccentric compression loading are compared
with their respective original columns (Fig. 14). The increasing
trend for stiffness, strength and displacement improvement for dif-
ferent repair techniques is shown in Fig. 14. It is clear that hybrid
FRP technique is highly effective in improving the overall perfor-
mance of columns even after being subjected to severe initial dam-
age under both concentric and eccentric compression loading. In
order to clearly compare the effectiveness of FRP repaired columns,
their behavior in terms of different indices are compared with the
QS repaired specimens and shown in Table 3.

(iii) Comparison of Axial Displacement Index (DI) 7.6. Failure mode comparison under high eccentric loading

Fig. 14. Overall comparison of RC elements with different repair techniques.

The failure mode comparison of column elements repaired
using different techniques and tested under high eccentric com-

Table 3
Comparison of different indices for FRP strengthened columns with RCC-R-QS.

Specimen ID Loading Type Strength Index (SI) Stiffness Index (KI) Displacement Index (DI)
RCC-R-NSM Concentric 3.8 19.4 10.0
Eccentric 42.4 37.8 2.4
RCC-R-HYB Concentric 32.5 77.7 27.7
Eccentric 57.6 68.9 348.7
 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380 377

a) Cracking – Tension Face b) Crushing - Compression Face c) Spalling – Compression Face

Fig. 15. Failure mode observation for repaired RC columns (QS).

a) Cracking – Tension Face b) Crushing - Compression Face c) Debonding – Tension Face

Fig. 16. Failure mode observation for NSM repaired RC columns.

pression is represented in Figs. 15–17. The failure progression is
represented in stages in terms of tension and compression stress
states observed in the specimen.
7.6.1. RC columns repaired using quickset cement mortar (RCC-R-QS-
145)
The column specimen which is repaired using only quickset
cement mortar had initial cracking in the tension face of the spec-
imen. At peak load, the crack propagated rapidly from the tension
face towards the compression face. A sudden drop in the behavior
is observed due to the severe crushing of concrete observed in the
compression face (Fig. 15). The failure is also due to the severe con-
crete cover spalling. The interface behavior between the old con-
crete and the repair mortar layer can be further enhanced by
providing additional shear keys or through novel anchorage sys-
tem. Investigating the effect of this interface behavior would be
interesting and scope for further work.
7.6.2. RC columns repaired using NSM technique (RCC-R-NSM-145)
The column specimens which are repaired using NSM technique
had initiation of cracking in the mid-span of the tension side
(Fig. 16) followed by development and propagation of cracks
towards the other sides of the column. At peak load, there is a sud-
den crushing of concrete observed in the compression face due to
the high stiffness of FRP and good bonding by the epoxy resin. In
the post peak regime, the NSM repaired specimen had ultimate
378 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380

a) Cracking – Tension Face b) FRPRupture – Compr ession Face

Fig. 17. Failure mode observation for hybrid FRP repaired RC columns.

failure due to debonding of FRP laminates in the bottom of the col-

umn on tension face (Fig. 16c). The column failed in sudden brittle
manner.

7.6.3. RC columns repaired using hybrid FRP technique (RCC-R-HYB-

145)
The failure propagation of hybrid FRP repaired columns under
eccentric compression loading is shown in Fig. 17. The column
had first crack initiation in the form of FRP rupture in the tension
face at a high compressive load of 750 kN. At increased loading
levels, these cracks opened up and propagated rapidly. After peak
load, the CFRP fabric which confined the NSM laminates opened
up due to sudden rupture (Fig. 17). This attributed to a sudden load
drop in the behavior of the specimen. The ultimate failure of the
specimen occurred due to significant rupture of CFRP fabric in
the compression face. However, due to effective restraining of
NSM laminates by EB CFRP confinement, the overall capacity is (a) Load – Longitudinal Steel Strain Behavior of Repaired RC Columns
enhanced significantly.
160
1050
(-) Compression (+) Tension
7.7. Load–strain behavior comparison of columns 140
900

The strain behavior of steel reinforcement and FRP is observed 120

from the strain gauges installed at the mid-height of the column. 750
Moment (kN.m)

100
Load (kN)

The initial strain in the damaged specimen is not added to the

600
repaired specimens in order to quantify the actual contribution 80
of the damaged specimens even after the steel yielded initially.
450
To make it clear, the strain data obtained for the original column 60
(RCC-0, RCC-145) are shown in Fig. 18. It is clear that hybrid FRP
300
repair technique is very effective under eccentric compression 40
RCC-R-NSM-145
RCC-R-NSM-145
loading. The column repaired using quick set cement mortar indi-
150 RCC-R-HYB-145-Fabric 20
cated yielding of longitudinal rebar in the tension face followed by RCC-R-HYB-145- Laminate
RCC-R-HYB-145-Laminate
the compression rebar yielding. Yield strain of reinforcement is 0 0
considered as 2500 lm/m from the material characterization of -12000 -8000 -4000 0 4000 8000 12000
steel rebar. NSM technique reduced the overall strain in the ten- -6
Longitudinal Strain ( x 10 )
sion/compression steel reinforcement due to additional contribu- (b) Load – CFRP Laminate/Fabric Strain of Repaired RC Columns
tion of CFRP laminates (Fig. 18(b)). Strain in the longitudinal
reinforcement is reduced further when the NSM CFRP laminates Fig. 18. Load–steel/CFRP strain comparison of repaired RC columns.
 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380 379

are prevented from debonding because of lateral confinement from suddenly. However, the hybrid FRP technique improved the curva-
EB CFRP fabric in the hybrid repair scheme (Fig. 18). ture at failure of the specimen significantly (Fig. 20).

7.8. Axial load–lateral displacement behavior 8. Summary and conclusions

The axial load–lateral displacement behavior of columns under
high eccentric compression is shown in Fig. 19. The column
repaired using quickset cement mortar had a maximum lateral dis-
placement of about 13 mm. Due to repair of columns using NSM
technique, the lateral displacement at failure increased up to 24%
irrespective of its brittle failure mode. Hybrid FRP technique
increased the lateral displacement at failure of the specimens by
98% when compared to similar specimens repaired with quick
set cement grout mortar.
7.9. Moment–curvature behavior
The moment–curvature behavior of repaired columns under
high eccentric compression is shown in Fig. 20. The moment capac-
ity (M) is calculated from the corresponding load (P) and the eccen-
tricity level (e) in the column (M = P * e). The curvature of the
specimen is calculated from the displacement measurements
obtained from the tension and compression sides of the test spec-
imen (from strain gradient). Due to brittle debonding failure of
NSM laminates, the curvature of the NSM repaired column reduced
1200
160
1000
140
120
Moment (kN.m)
The overall objective of this work is to assess the effectiveness
of different rapid repair techniques for severely damaged columns
under combined compression and bending loads. Six columns are
severely damaged and repaired using (i) high strength cement
grout (ii) NSM technique and (iii) Hybrid FRP technique. Test
results of repaired specimens under eccentric and concentric com-
pression are compared. The following major conclusions can be
drawn from the results presented in this paper.
The presence of additional bending moment significantly
reduced the axial strength and stiffness of the repaired col-
umns. Repair of columns using quick set cement mortar is
found to be least effective under both loading combinations.
Repair of severely damaged columns using hybrid FRP tech-
nique is found to be the most efficient in restoring the original
strength, stiffness and displacement. It improved the overall
performance of RC columns in terms of initial and post cracking
stiffness, peak strength and failure displacement under both
concentric and eccentric compressions.
Hybrid FRP repair technique prevented the debonding failure of
NSM laminates and led to increase in peak strength and failure
displacement. The failure of hybrid specimens occurred due to
rupture of externally bonded fabric in the compression face
under eccentric compression.
Repair of RC columns using NSM CFRP technique exhibited a
good improvement in the strength and failure displacement
under high eccentric loading (e/d = 0.63). For similar specimens

800
Load (kN)

tested under concentric compression, the original strength of

100 column is not fully restored.
600 RCC-145
80
RCC-R-QS-145
Conflict of interest
400 60
RCC-R-NSM-145 There is no conflict of interest to declare.
40
200 RCC-R-HYB-145
20 Acknowledgements
0 0
0 4 8 12 16 20 24 28 32 This research work was sponsored by ‘‘FAST – Center for Sus-
Lateral Displacement @ Mid-Span (mm) tainable Development” funded by Ministry of Human Resource
Development, India. The authors gratefully acknowledge the gen-
Fig. 19. Load–lateral displacement comparison of repaired RC columns. erous support provided. The authors would also acknowledge the
help of lab technicians at the structural engineering laboratory,
IIT-H during casting, strengthening and testing of columns.
160
1050
140
RCC - 145 References
900
RCC - R-QS -145 [1] A. Belarbi, P.F. Silva, Retrofit using CFRP composites of RC bridge columns
120
under combined axial, shear, flexure, and torsion, in: Torres Marques (Ed.),
Moment (kN.m)

RCC - R-NSM-145 750 Challenges for Civil Construction, FEUP, Porto, 2008, p. 10.
100
Load (kN)

[2] S.S. Prakash, A. Belarbi, Y.M. You, Seismic performance of circular RC columns
RCC - R-HYB-145 600 subjected to axial, bending, and torsion with low and moderate shear, Eng.
80 Struct. (Elsevier) 32 (1) (2010) 46–59.
[3] V.K.R. Kodur, Spalling in HSC exposed to fire: concerns, causes, critical
450 parameters and cures, in: Proceedings of ASCE Structures Congress,
60
Philadelphia, PA, 2000, p. 12.
300 [4] V.K.R. Kodur, F.P. Cheng, T.C. Wang, Effect of strength and fiber reinforcement
40 on the fire resistance of high strength concrete columns, ASCE J. Struct. Eng.
129 (2) (2003) 253–259.
20 150 [5] A. Belarbi, S.S. Prakash, P.F. Silva, Incorporation of decoupled damage index
models in the performance-based evaluation of RC circular and square bridge
0 0 columns under combined loadings, ACI Spec. Publ. SP271 271 (2010) 79–102.
0.00000 0.00005 0.00010 0.00015 0.00020 0.00025 0.00030 [6] S.S. Prakash, A. Belarbi, Towards damage-based design approach for RC bridge
columns under combined loading using damage index models, J. Earthquake
Curvature ( x 10-6 mm-1) Eng. 14 (3) (2010) 363–389.
[7] IS 1893, Criteria for Earthquake Resistant Design of Structures – Part 1:
Fig. 20. Moment–curvature comparison of repaired RC columns. General Provisions and Buildings, Bureau of Indian Standards, 2016.
380 M. Chellapandian, S.S. Prakash / Construction and Building Materials 173 (2018) 368–380
[8] IS 1893, Criteria for Earthquake Resistant Design of Structures – Part 3:
Bridges and Retaining Walls, Bureau of Indian Standards, 2014.
[9] AAHSTO, American Association of State and Highway Transportation Officials
– LFRD Bridge Design Specifications, 2017.
[10] J.M. Adam, I. Salvador, F.J. Pallares, E. Gimenez, P.A. Cadderon, Axially loaded
reinforced concrete columns strengthened by steel caging – finite element
modeling, Constr. Build. Mater. 23 (2009) 2265–2276.
[11] J. Garzon-Roca, J.M. Adam, P.A. Calderon, Behaviour of RC columns
strengthened by steel caging under combined bending and axial loads,
Constr. Build. Mater. 25 (2011) 2402–2412.
[12] Y. Ghernouti, A. Li, R. Rabehi, Effectiveness of repair on damaged concrete
columns by using fiber-reinforced polymer composite and increasing concrete
section, J. Reinforced Plast. Compos. 31 (23) (2012) 1616–1629.
[13] Y.H. Chai, M.J.N. Priestley, F. Seible, Analytical model for steel jacketed
reinforced concrete circular bridge columns, ASCE J. Struct. Eng. 120 (8) (1994)
2358–2376.
[14] ACI 440.2R-17, Guide for the Design and Construction of Externally Bonded
FRP System for Strengthening Concrete Structures, ACI Committee 440,
Farmington Hills (MI), 2017, p. 45.
[15] A. Nanni, A.D. Luca, H.J. Zadeh, Reinforced Concrete with FRP Bars – Mechanics
and Design, CRC Press Taylor and Francis Group, 2014. ISBN: 13: 978-0-203-
87429-5.
[16] Q. Han, J. Wen, X. Du, J. Jia, Experimental and numerical studies on seismic
behavior of hollow bridge columns retrofitted with carbon fiber reinforced
polymer, J. Reinforced Plast. Compos. 33 (24) (2014) 2214–2227.
[17] M.N.S. Hadi, Comparative study of eccentrically loaded FRP wrapped columns,
Compos. Struct. 74 (2) (2006) 127–135.
[18] E.T. Maaddawy, Strengthening of eccentrically loaded reinforced concrete
columns with fiber reinforced polymer wrapping system – experimental
investigation and analytical modeling, ASCE J. Compos. Constr. 13 (1) (2009)
146–152.
[19] T. Maaddawy, M. Sayed, B.A. Magid, Effect of cross sectional shape and loading
condition on performance of reinforced concrete members confined with
carbon fiber reinforced polymers, Mater. Des. 31 (2013) 2330–2341.
[20] M.N.S. Hadi, H. Zhao, Experimental study of high strength concrete columns
confined with different types of mesh under eccentric and concentric loading,
ASCE J. Mater. Civ. Eng. 23 (6) (2013).
[21] P. Kankeri, S.S. Prakash, S.K.S. Pachalla, Analytical and numerical studies on
hollow core slabs strengthened with hybrid FRP and overlay techniques,
Struct. Eng. Mech. J. Technopress 65 (5) (2018) 535–546.
[22] S.K.S. Pachalla, S.S. Prakash, Load resistance and failure modes of GFRP
composite strengthened Hollow core slabs with openings, J. Mater. Struct. 50
(1) (2016).
[23] P. Kankeri, S.S. Prakash, Efficient hybrid strengthening for precast hollow core
slabs at low and high shear span to depth ratios, Compos. Struct. J. 170 (15)
(2017) 202–214.
[24] J.A.O. Barros, R.K. Varma, M. Sena-Cruz, A.F.M. Azevedo, Near surface mounted
FRP strips for the flexural strengthening of RC columns – experimental and
numerical research, Eng. Struct. J. 30 (12) (2008) 3412–3425.
[25] M. Badawi, K. Soudki, Control of corrosion-induced damage in reinforced
concrete beams using carbon fiber-reinforced polymer laminates, ASCE J.
Compos. Constr. 9 (2) (2005) 195–201.
[26] D.A. Bournas, T.C. Triantafillou, Flexural strengthening of RC columns with
NSM FRP or stainless steel, ACI Struct. J. 106 (4) (2009) 495–505.
[27] B. Charalambidi, T.C. Rousakis, A.I. Karabinis, Fatigue behavior of large-scale
reinforced concrete beams strengthened in flexure with fiber-reinforced
polymer laminates, ASCE J. Compos. Constr. 20 (5) (2016).
[28] P. Kankeri, S.S. Prakash, Experimental evaluation of bonded overlay and NSM
GFRP bar strengthening on flexural behavior of precast prestressed hollow
core slabs, Eng. Struct. 120 (2016) 49–57.
[29] B. Almassri, F. Al-Mahmoud, R. Francois, Behaviour of corroded reinforced
concrete beams repaired with NSM CFRP rods, experimental and finite
element study, Compos. Part: B Eng. 92 (2016) 477–488.
[30] G.G. Triantafyllou, T.C. Rousakis, A.I. Karabinis, Corroded RC beams patch
repaired and strengthened in flexure with fiber-reinforced polymer laminates,
Compos. Part: B Eng. 112 (1) (2017) 125–136.
[31] G.G. Triantafyllou, T.C. Rousakis, A.I. Karabinis, Effect of patch repair and
strengthening with EBR and NSM CFRP laminates for RC beams with low,
medium and heavy corrosion, Compos. Part B: Eng. 133 (2018) 101–111.
[32] S. Jain, M. Chellapandian, S.S. Prakash, Emergency repair of severely damaged
reinforced concrete column elements under axial compression: an
experimental study, Constr. Build. Mater. 155 (2017) 751–761.
[33] K.D. Dalgic, M. Ispir, E. Binbir, A. Ilki, Effects of pre-damage on axial behavior of
CFRP jacketed non-circular members, in: Proceedings of Conference on Civil
Engineering Infrastructure based on Polymer Composites 2012; Poland, 22nd–
23rd November, 2012.
[34] A. Ilki, N. Kumbasar, Behavior of damaged and undamaged concrete
strengthened by carbon fiber composite sheets, Struct. Eng. Mech. 13 (1)
(2002) 75–90.
[35] G.C. Manos, D.S. Stavroy, M.A. Dimosthenous, B. Kourtides, Experimental and
analytical investigation of repaired and strengthened reinforced concrete
structural elements utilizing CFRP, in: 13th World Conference on Earthquake
Engineering, Vancouver, Canada, 2004, p. 91.
[36] M. Karantzikis, C.G. Papanicolaou, C.P. Antonopoulos, T.C. Triantafillou,
Experimental investigation of nonconventional confinement for concrete
using FRP, ASCE J. Compos. Constr. 9 (6) (2005) 480–487.
[37] T.C. Rousakis, Reusable and recyclable nonbonded composite tapes and ropes
for concrete columns confinement, Compos. Part B: Eng. 103 (2016) 15–22.
[38] K.K. Sideris, E. Manikas, A. Michalakopoulos, Effectiveness of different repair
materials and methods used for the repair of chloride corroded reinforced
concrete structures, in: Concrete Solutions – Proceedings of Concrete
Solutions, 6th International Conference on Concrete Repair, 2016, pp. 301–
304.
[39] A. Ilki, O. Peker, E. Karamuk, C. Demir, N. Kumbasar, FRP retrofit of low and
medium strength circular and rectangular reinforced concrete columns, ASCE
J. Mater. Civ. Eng. 20 (2) (2008) 169–188.
[40] S.J. Elkin, A.M. Nacamuli, D.E. Lehman, J.P. Moehle, Seismic performance of
damaged bridge columns, Earthquake Eng. Eng. Seismol. 1 (1999) 39–50.
[41] H. Saadatmanesh, M.R. Ehsani, L. Jin, Repair of earthquake-damaged RC
columns with FRP wraps, ACI Struct. J. (1997) 94. 206–X.
[42] R. He, S. Grelle, L.H. Sneed, A. Belarbi, Rapid repair of a severely damaged RC
column having fractured bars using externally bonded CFRP, Compos. Struct.
101 (2013) 225–242.
[43] Y. Yang, L.H. Sneed, A. Morgan, M.S. Saiidi, A. Belarbi, Repair of RC bridge
columns with interlocking spirals and fractured longitudinal bars – an
experimental study, Constr. Build. Mater. 78 (2015) 405–420.
[44] R. He, L.H. Sneed, A. Belarbi, Rapid repair of severely damaged RC columns with
different damage conditions: an experimental study, Int. J. Concr. Struct.
Mater. 7 (2013) 35–50.
[45] R. He, L.H. Sneed, A. Belarbi, Torsional repair of severely damaged column
using carbon fiber-reinforced polymer, ACI Struct. J. 111 (2014) 705–716,
https://doi.org/10.14359/51686627.
[46] R.Y. Wu, C.P. Pantelides, Rapid seismic repair of reinforced concrete bridge
columns, ACI Struct. J. 114 (5) (2017) 1339–1350.
[47] L. Bisby, M. Ranger, Axial-flexural interaction in circular FRP-confined
reinforced concrete columns, Constr. Build. Mater. 24 (9) (2010) 1672–1681.
[48] M.N.S. Hadi, T.D. Le, Behaviour of hollow core square reinforced concrete
columns wrapped with CFRP with different fibre orientations, Constr. Build.
Mater. 50 (2014) 62–73.
[49] M. Chellapandian, S.S. Prakash, A. Sharma, Strength and ductility of innovative
hybrid NSM reinforced and FRP confined short reinforced concrete columns
under axial compression, Compos. Struct. 176 (2017) 205–216.
[50] M. Chellapandian, S.S. Prakash, Behavior of FRP strengthened RC columns
under pure compression – experimental and numerical studies, Proc. Struct.
Eng. Convent. SERC-Chennai (2016) 1730–1735.
[51] M. Chellapandian, S.S. Prakash, A. Rajagopal, Analytical and finite element
studies on hybrid FRP strengthened RC column elements under axial and
eccentric compression, Compos. Struct. 184 (2017) 234–248.
[52] M. Chellapandian, S.S. Prakash, Behavior of hybrid NSM reinforced and
externally confined short RC columns under eccentric compression –
experimental and numerical research, in: 71st RILEM Week – ICACMS
Conference, Chennai, 2nd–7th September, vol. 2, 2017, pp. 519–528.
[53] Indian Standards 10262, Recommended Guidelines for Concrete Mix Design,
Bureau of Indian Standards, New Delhi, 2009.
[54] ASTM D3039M-17, Standard test Method for Tensile Properties of Polymer
Matrix Composite Materials, West Conshohocken, PA, 2017.
[55] ASTM D6484M-14, Standard Test Method for Compressive Strength of
Polymer Matrix Composite Laminates, West Conshohocken, PA, 2014.
[56] V.S. Kuntal, M. Chellapandian, S.S. Prakash, Investigation on NSM FRP shear
strengthening effect on prestressed concrete beams with different bonding
materials, in: Proceedings of ICCMS Conference, Hyderabad, India, 2017, pp.
2310–2318.
[57] V.S. Kuntal, M. Chellapandian, S.S. Prakash, Efficient near surface mounted
CFRP shear strengthening of high strength prestressed concrete members – an
experimental study, Compos. Struct. 180 (2017) 16–28.
[58] A. Vosooghi, Post-earthquake evaluation and emergency repair of damaged RC
bridge columns using CFRP materials (Ph.D Dissertation), Center for Civil
Engineering Earthquake Research, Dept. of Civil Engineering, Univ. of Nevada,
Reno, NV, 2010.
[59] A. Vosooghi, M. Saiidi, Design guidelines for rapid repair of earthquake-
damaged circular RC bridge columns using CFRP, J. Compos. Constr. 18 (9)
(2013) 827–836.
[60] A. Vosooghi, M. Saiidi, Post-earthquake evaluation and emergency repair of
damaged RC bridge columns using CFRP materials. Rep. No. CCEER-10-05,
Center for Civil Engineering Earthquake Research, Dept. of Civil Engineering,
Univ. of Nevada, Reno, NV, 2010.
[61] A. Vosooghi, M. Saiidi, Seismic damage states and response parameters for
bridge columns, in: P.F. Silva, R. Valluvan (Eds.), Structural Concrete in
Performance-Based Seismic Design of Bridges, American Concrete Institute,
Farmington Hills, MI, 2010, pp. 27–44.
[62] A. Vosooghi, M. Saiidi, Shake table studies of repaired RC bridge columns using
CFRP fabrics, ACI Struct. J. 110 (1) (2013) 105–114.