Construction and Building Materials 154 (2017) 632–643

Contents lists available at ScienceDirect

Construction and Building Materials

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

Recovering flexural performance of thermally damaged concrete beams

using NSM CFRP strips
Rami H. Haddad ⇑, Oubaida A. Almomani
Jordan University of Science and Technology, P.O. Box 303, 22110 Irbid, Jordan

h i g h l i g h t s g r a p h i c a l a b s t r a c t

Exposure of beams to 600 °C for 2 h The potential of recovering flexural performance of thermally damaged concrete beams using near sur-
degraded their mechanical properties. face mounted (NSM) carbon fiber reinforced polymer (CFRP) strips was satisfactory when enough devel-

Flexural behavior of post-heated opment lengths beyond critical stress region were provided. Intact/strengthened and heat-damaged/
beams with NSM CFRP strips was repaired beams showed improved load capacity and toughness, yet experienced reductions in ductility
recovered. and toughness as compared to control ones.

Ductility was recovered for heat-
damaged beams with NSM CFRP
strips.

Heat-damaged had significantly
reduced bond strength between
concrete and NSM CFRP.

End-concrete cover peeling off was
the dominant failure mode for
repaired and strengthened beams.

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

Article history: The potential of recovering the flexural performance of thermally damaged concrete beams using near
Received 29 April 2017 surface mounted (NSM) carbon fiber reinforced polymer (CFRP) strips was experimentally investigated.
Received in revised form 11 July 2017 Twenty reinforced concrete beams (150  250  1400 mm) were cast then cured for 28 days in moist
Accepted 29 July 2017
burlap. A set of ten beams were heated at 600 °C for 2 h using an electrical furnace whereas those of
the second set were left in laboratory air. Four pairs of beams from each set were repaired/strengthened
at their tension side using similar configurations of NSM CFRP strips. Duplicate beams of each set were
Keywords:
tested as controls. The mechanical performance of different beams was evaluated under four-point load-
Strengthening
NSMCFRP strips
ing test setup including measurement of strain in NSM CFRP strips and slippage between NSM CFRP strips
Configuration and concrete. Moreover, cracking and failure modes were monitored and characterized. Intact/strength-
Peeling off ened and heat-damaged/repaired beams showed improved load capacity and toughness, yet experienced
Analytical prediction reductions in ductility and toughness as compared to control ones. Different performance indicators
revealed good potential of repairing heat-damaged beams using NSM CFRP strips. End-cover separation
failure mode was observed for both strengthened as well as repaired beams. Analytical predictions of
ultimate load capacity for different beams confirmed experimentally obtained results.
Ó 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.
E-mail address: rhaddad@just.edu.jo (R.H. Haddad).

http://dx.doi.org/10.1016/j.conbuildmat.2017.07.211
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643
1. Introduction
Reinforced concrete (RC) structural members are designed to
have sufficient strength to resist all types of loads. In an event of
fire, however, these structures are exposed to very high tempera-
tures that affects negatively their mechanical and physiochemical
properties; causing the structural elements to deface and displace
and possibly spall under certain circumstances. As exposure tem-
peratures exceeds 400 °C, evaporable water is forced out from con-
crete’s capillary, gel and interlayer pores, in sequences, with
calcium hydroxide decomposed at about 400 °C and C-S-H broken
down and combined water lost in the temperature range of 500–
800 °C [1,2].
Usually reinforced concrete (RC) structures withstand low to
moderate fire exposure conditions because of their high thermal
capacity, and low thermal conductivity. However, upon exposure
to an accidental fire for periods exceeding 2 h, concrete structures
would show noticeable degradation, sever reductions in concrete’s
strength and stiffness, loss in steel reinforcements yield strength,
and degradation in bond between concrete and steel [3–6]. The
level of degradation in these structural elements is dependent
upon the type of structural element and extent and rate of fire,
type of load and support conditions. Documented fire accidents
showed that few concrete structures had collapsed as a result of
fire and that the majority of these structures are repairable.
Accordingly, the potential of repairing heat-damaged concrete
members had received a lot of interest; especially over the past
30 years or so [7–13].
The present database showed that reinforced concrete beams
are the most repaired structural elements, owning to their high
susceptibility to fire, which impacts tangibly their load resistance
mechanism. Until about 20 years ago, steel plates and elements
or reinforced concrete jacketing (RCJ) were mainly used in repair/
strengthening of beams; both of which had its disadvantages
[9,10]. Possessing better physical, chemical and mechanical prop-
erties than steel plates and RCJ, Fiber Reinforced Polymer (FRP)
plates and sheets have been used to strengthen/repair various con-
crete structural elements. The outcome of approximately twenty
years of research indicated many advantages of FRP materials over
conventional repair techniques; including higher repair efficiency,
and easier application within a relatively short period of time [11–
13]. However, the tendency of FRP composites in the forms of con-
ventional plates and sheets to detach from concrete at relatively
low loads undermined the benefit of their use in repair works.
To overcome the shortcomings associated with the use of con-
ventional FRP plates in repair, a new form of FRP composite is cur-
rently emerging, near surface mounted (NSM) Carbon FRP (CFRP)
strips. Those are inserted in man-created groove that is filled with
a strong adhesive. Compared to externally bonded FRP plates or
sheets, NSM CFRP strips are protected against accidental mechan-
ical damage, sun and moisture hence would be very attractive to
promote negative moment capacity in areas exposed to mechani-
cal damage or environmental attacks. Furthermore, NSM CFRP
strips are less prone to de-bonding from concrete substrate and
their application introduces no aesthetic changes to repaired struc-
tural elements [14–29]. Consequently, various studies were
recently undertaken to investigate the mechanical performance
and failure modes of beams repaired for shear and flexure using
NSM CFRP strips. The major findings indicated: (a) higher contribu-
tion of NSM CFRP strips to improving structural performance as
compared to externally attached strips of similar sectional area
and strength properties; (b) grove’s thickness and spacing to have
a significant impact on the efficiency of repair; and (c) premature
concrete cover delamination as a major disadvantage. Most studies
reported soffit cover delamination as the dominant failure mode in
633
shear strengthened beams; especially when relatively close strip
spacing is used [14–29].
2. Problem statement and objectives
Premature detachment of conventional Carbon FRP (CFRP)
plates in strengthened concrete beams before compression con-
crete develops its ultimate strain capacity had limited the benefit
of using such materials in repair works. Recently, NSM CFRP strips
had been manufactured and used for repair of concrete flexural
elements because their insertion in grooved concrete contributes
to an improved bond with concrete in addition to other benefits,
discussed earlier. In an event of exposure of fire, reinforced con-
crete elements may crack and loses its load capacity partially or
fully and hence the potential of regaining their structural capacity
using NSM CFRP strips becomes questionable. This work tends to
investigate the contribution of NSM CFRP strips to recovering the
original load capacity of heat-damaged and repaired beams.
3. Methodology of study
To achieve the objectives of the study, twenty reinforced con-
crete beams (150  250  1400 mm) were designed according to
ACI committee 318 as under reinforced, then prepared, and
moist-cured for 28 days [30]. The first set of ten beams were left
at laboratory air whereas the second set, represented the other
half, were subjected to an elevated temperature of 600 °C in an
electric furnace for a period of 2 h. Duplicate beams from each
set were either kept as controls or strengthened/repaired using
CFRP strips at different numbers and embedment lengths. The test-
ing program is summarized in Table 1 with strengthening/repair
configurations illustrated schematically in Fig. 1. Letter-number
designations were used to designate beams’ status (intact, heat-
damaged) and strengthening/repair configuration with regard to
number of NSM CFRP strips, and their extension and spacing.
All beams were tested under the effect of four-point loading
with applied load, mid-span deflection, strain in NSM CFRP and
slippage between NSM CFRP strips and concrete acquired using a
data acquisition system. The data were processed for load-
deflection and load-slip diagrams and their characteristics to
establish the significance of repairing heat-damaged beams using
the proposed techniques.
4. Experimental program
Material properties, mix design, concrete mixing and casting,
detailing of test beams and reinforcement, heat treatment protocol,
procedure of repair applications and testing setup and procedure
are presented in this section in details.
4.1. Concrete ingredients and proportions
Type I ordinary Portland cement, tap water, coarse limestone
aggregate with 19 mm maximum aggregate size, and a mixture
of fine limestone aggregate and silica sand at 70% and 30%, respec-
tively, were used in casting the concrete mixture used in this work.
A commercial superplasticizer was added to the mixture to achieve
the target workability. The coarse aggregate gradation meets the
requirements of the ASTM method C33 [31]. The physical proper-
ties for the different aggregate particles were carried out according
to ASTM specifications with results listed in Table 2, [31].
The proportions of concrete’s ingredients were determined
according to the mix-design method ACI-211.1 [32] such that a
cylinder compressive strength of 35 MPa is achieved at 28 days
634 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643

Table 1
Test parameters considered with designation of different beam specimens.

set # NS SCONF GN EL (mm) GS (mm) Beam designation

1 2 None N.A. N.A. N.A. C
2 2 NSM 2 150 7.5 C SN2 ELð15Þ SP7:5
2 NSM 2 250 7.5 C SN2 ELð25Þ SP7:5
2 NSM 2 450 7.5 C SN2 ELð45Þ SP7:5
2 NSM 3 150 5 C SN3 ELð15Þ SP5
3 2 None N.A. N.A. N.A. HD
4 2 NSM 2 150 7.5 HD SN2 ELð15Þ SP7:5
2 NSM 2 250 7.5 HD SN2 ELð25Þ SP7:5
2 NSM 2 450 7.5 HD SN2 ELð45Þ SP7:5
2 NSM 3 150 5 HD SN3 ELð15Þ SP5

SCONF, strengthening configuration; NS: Number of samples; SN, Strips number; GS: Grooves spacing; GN: Grooves number; SP: spacing between strips; NSM: near surface
mounted; Embedment length (EL): the distance between the end of maximum moment zone to the termination point of NSM strips; HD: heat-damaged.

Fig. 1. Schematic presentations illustrating longitudinal and transverse positioning of two and three NSM CFRP strips used to strengthen/repair different concrete beams. (a)
A schematic showing the detailing of main and lateral reinforcement of the beams. (b) A schematic of test setup for present beams with attached steel plates.

Table 2
Physical properties of aggregate used in present work.

Properties Aggregate
CL FL SS 70%FL + 30SS
BSG(D) 2.41 2.60 2.59 2.62
FM N. A. 3.25 1.6 2.8
Absorption (%) 2.1 1.7 0.5 N. A.
UW (kg/m3) 1401 1612 N. A. N. A.

CL, Coarse Limestone; FL, Fine Limestone; SS, Silica Sand; B.S.G (D), Bulk Specific Gravity; F.M, Fineness Modulus; UW, Unit Weight.

of moist curing, while attaining the required slump at 100 mm. The
proportions of the present concrete mixture, designed at a w/c
ratio of 0.50, were 430, 870, 506, 217, 236, and 3.23 kg/m3 for
cement, coarse aggregate, fine aggregate, silica sand, water, and
superplasticizer, respectively.
4.2. Test specimens
Twenty beams (150  250  1400 mm) were reinforced with
2u12 mm steel bars at the compression side (top), and 2u14 mm
at the tension side (bottom), with steel stirrups of u8 mm spaced
@ 75 mm (center-to-center): steel detailing is shown in the
schematics of Fig.2a. The mechanical and geometric characteristics
for the steel bars were determined at room temperature and after
being exposed to 600 °C for three hours then cooled at laboratory
air: results are listed in Table 3. Additional shear reinforcement
using two parallel steel plates of 10 mm thickness were provided
at both high shear zones of the present beams to ensure flexural
failure, as shown in Fig. 2b. The steel plates covered the shear
zones to a distance of 30 mm from the bottom surface of the
beams’ soffit where the NSM CFRP strips are inserted in concrete
grooves. In such a way the attached steel plates’ would have no
direct effect upon possible end-concrete cover separation at the
bottom of strengthened/repaired beams.
R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643 635

Fig. 2. Schematics of beams illustrating reinforcement detailing and testing setup.

636 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643

Table 3
Mechanical and geometric properties for reinforcing steel of different bar sizes before and after exposure to 600 °C.

Temperature BD (mm) YS (MPa) US (MPa) EAF (%) AD (mm) RT (mm) RW (mm) RA (mm)
23 °C 8 390 570 13
12 441 641 8
14 608 732 7.5 15.4 0.8 2.6 30
600 °C 8 234 364 10
12 308 474 8.8
14 440 561 8.2 14.7 0.7 2.2 30

BD, bar diameter; YS, Yield stress; US, Ultimate strength; EAF, Elongation at failure; AD, actual diameter; RT, Rib height; RW, Rib width; RA, Rib angle.

4.3. CFRP composites 4.4. Concrete mixing, casting, and curing

NSM CFRP strips and their prescribed adhesive from SIKA were A tilting drum mixer of 0.15 m3 was used for mixing the concrete
used in strengthening different specimens. Their properties, as pro- ingredients. At first, a little amount of water was added in the mixer
vided by the manufacturer (SIKA), are shown in Tables 4 and 5, to wet surface, then coarse aggregates were added and witted with
respectively. some water while the mixer was rolling. Afterwards, cement, fine
aggregate, silica and water were added gradually and alternatively,
before a superplasticizer was added to the mixture at 0.75% by
cement weight to achieve the required workability. Finally, the
ingredients were blended for about 1 min in the mixer to obtain a
Table 4
Physical and mechanical properties of SIKA NSM CFRP cohesive mixture. Concrete was placed in the molds in three layers;
strips. each compacted using a vibrator, before the beams surface finished
smooth by a trowel, as shown in Fig. 3. The specimens were
Fiber Type NSM CFRP
demolded after 24 h then covered with wet burlap for another
Fiber orientation 0° (unidirectional)
27 days. Standard cylinder specimens (100  200 mm) were also
Fiber Density 1.6 g/cm3
Strip width 15 mm cast, cured in wet burlap and tested for compressive strength after
Strip thickness 2.5 mm 28 days according to ASTM C 39 [31].
Cross sectional area 37.5 mm2
Mean tensile strength 3100 N/mm2
Tensile E-modulus 165000 N/mm2 (nominal) 4.5. Thermal treatment of beams
Strain at break >1.7% (nominal)
Concrete beams, in duplicates, with standard cylinder speci-
mens were exposed to high temperatures of 600 °C for a period
Table 5
of 2 h in an electrical furnace with temperature and heating time
Properties of SIKA epoxy used for bonding CFRP strips. controlled by an electronic panel, located next to the furnace.
Because of the beams’ relatively large size and high moisture con-
Color Cement grey
tent, the heating process was continued for few hours until the tar-
Mixed density at 25 °C 1.8 g/cm3 (approx.) get temperatures was reached: Fig. 4 shows the heating regime
Sag flow Non-sag on vertical surface
Flashpoint N/A
adopted. After heating, the beams were allowed to cool down
Shrinkage 0.04% inside the furnace at a rate of 24 °C/h before taken out and covered
Compressive strength-ASTM C579 85 N/mm2 at 3 days with plastic sheets to prevent possible self-healing by air moisture.
Flexural strength according to 25 N/mm2 at 7 days
ASTM
Tensile strength according to ISO 18 N/mm2 at 7 days 4.6. Bonding NSM CFRP strips
527
Shear strength according to ASTM 21 N/mm2 at 7 days
In order to install the NSM CFRP strips, grooves with a depth of
Bond to concrete >4 N/mm2 at 1 day (concrete fracture)
E-Modulus-ASTM& ISO 527 10000 N/mm2(compression and 22.5 mm and width of 8 mm were made using an electrical drill at
tension) different spacing values, as shown in Fig. 5. The dimensions for the
grooves were chosen upon the recommendation by Seo et al. [22]

Fig. 3. Beams cast in molds and surface finished, ready for moist curing.
 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643
Fig. 4. Heating regime inside the electric furnace as applied to present specimens.
and considering the geometric dimensions of the CFRP strips used.
The grooves were cleaned by air vacuuming and direct brushing,
before wetted with a volatile compound to reduce surface humid-
ity. Later, the bonding epoxy was squeezed into the grooves until
filled completely to expel air bubbles before the NSM CFRP strips
were pressed into the full depth of the grooves with the epoxy,
squeezed out, used to cover the strips by a special spatula.
4.7. Test setup
Control and NSM CFRP repaired/strengthened concrete (RC)
beams were simply supported over a span of 1.3 m then subjected
to a four-point loading, as shown in Fig. 2. The distance between
two point loads was kept constant at 400 mm. The supports were
designed to act as a roller from one end and as a hinge from the
other. One linear variable displacement transducers (LVDT) was
positioned at the mid span to measure vertical deflection, whereas
another LVDT was used to measure longitudinal strain in CFRP
strips. The latter was mounted on two small extensions of NSM
CFRP strips, attached to the main NSM CFRP strips at two points
spaced at 160 mm around the center of the beams. Another two
LVDTs was mounted on an extension from each free end of one
NSM CFRP strip with its knob touching a 30  30 mm plastic plate,
pre-adhered to concrete surface: Fig. 2b illustrates LVDTs distribu-
tion. The load was applied using a hydraulic testing machine of
250 KN capacity under loading rate of 0.02 kN/s. Measurements
of load versus vertical deflection, NSM CFRP elongation, and slip-
page between NSM CFRP strips and concrete were collected elec-
tronically by a data acquisition system.
Fig. 5. Creating single and multiple grooves on the tension side of the beam; Sawing procedure (left); a schematic showing groove dimensions (right).
637
5. Result and discussion
5.1. Effect of elevated temperature
Cracking extent and mode after heating: Exposure of concrete
beams to 600 °C for a period of 2 h had resulted in intense cracking
on their surfaces and in the vicinity of reinforcing steel, as shown
in Fig. 6. Cracks were small and spread all over the beam with an
average width of less than 0.08 mm. Those formed around steel
bars are expected to weaken steel bond to surrounding concrete.
Water vapor pressure and decomposition of both CH and C-S-H
are major causes of observed cracking [1,2].
Concrete strength: The compressive strength of heat-damaged
and control concrete, determined using standard cylinders
(100  200 mm), were found to be 10 and 36 MPa, respectively.
The percentage reduction in strength at 73% is consistent with
the amount of damage observed on the surface of concrete beams.
Reinforcing steel’s mechanical and geometric properties: the
mechanical properties of reinforcing steel bars having diameters of
14, 12 and 8 mm, obtained from the heat-damaged specimens,
were determined from their stress-strain diagrams before com-
pared to those of control steel specimens: the mechanical and geo-
metric properties of different bars are summarized in Table 3. As
can be deduced, the yield stress was reduced by about 40%, 30%
and 28% for bar sizes of 14, 12, and 8 mm, respectively, whereas
the ductility was increased upon heating. Furthermore, the mea-
sured post-heating diameter, and rib height and width for the
14-mm bars were reduced.
Mechanical performance of concrete beams: The load-
deflection curves for control and heat-damaged beams showed
similar trend behavior, as depicted in Fig. 7: curves were linear
up to certain point before became non-linear. Limited linear por-
tion can be clearly noticed for heat-damaged beams because of
concrete’s thermal cracking. Upon heating, the flexural load capac-
ity, stiffness and toughness were reduced by about 40%, 40% and
11%, respectively, whereas rotational ductility increased by 57%.
The degradation in mechanical properties and softening in heat-
damaged beams is referred to the reduction in: (a) the yield
strength of steel reinforcing bars; (b) bond strength between steel
reinforcement and concrete; and (c) concrete compressive
strength, which was reduced by about 73% of its intact value.
5.2. Mechanical response of beams strengthened/repaired with NSM
CFRP strips
5.2.1. Introduction
The mechanical behavior of NSM CFRP strengthened/repaired
beams was investigated considering the effect of key parameters,
638 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643

Fig. 6. Post-heating cracking pattern in concrete beams and around steel reinforcement.

stages of loading before turned nonlinear behavior thereafter. Sim-

ilar trend behavior was noticed for intact/strengthened having
identical NSM CFRP strips’ configurations, as shown in Fig. 9.
Load-deflection curves, pertaining to concrete beams with 3 strips
of NSM CFRP, reflected either limited improvement and even
degradation in the mechanical response of intact/strengthened
and heat-damaged/repaired, respectively, as compared to control
beams, as depicted in Fig. 10. To establish the significance
of repairing heat-damaged beams using NSM CFPR strips,
corresponding mechanical characteristics, as obtained from
load-deflection curves, namely ultimate load capacity, rotational

Fig. 7. Load-deflection curves for control (C) and heat-damaged (HD) beams.

namely NSM CFRP strips’ embedment length, and number. It is

important to mention that the repair/strengthening configurations,
considered in this work, were chosen based upon a primary study
by the present authors, as well as most recent literature works. To
establish the potential of using NSM CFRP strips, as a repair tech-
nique of heat-damaged beams, relevant mechanical characteristics
and failure modes are compared to those of intact-strengthened
ones.

5.2.2. Load-deflection diagram Fig. 9. Load versus deflection for beams strengthened with two NSM CFRP strip at
Fig. 8 shows load-deflection curves for heat-damaged beams, different embedment lengths at a constant spacing of 75 mm.
repaired with two NSM CFRP strips at embedment length of 150,
250 and 450 mm. The curves showed linear behavior at the initial

Fig. 8. Load-deflection curves for heat-damaged and repaired beams with two NSM Fig. 10. Load versus deflection for heat-damaged beams, repaired with three NSM
CFRP strip at different embedment lengths at a constant spacing of 75 mm. CFRP strip at constant embedment length of 150 mm and spacing of 50 mm.
 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643 639

ductility, stiffness, and toughness, were obtained and compared to
those of intact/strengthened ones, as shown in Table 6.
5.2.3. Characteristics of load-deflection diagrams
Load capacity: heat-damaged beams, repaired with two NSM
CFRP strips at embedment length of 150, 250 and 450 mm,
achieved residuals load capacity of 69, 83 and 100%, respectively,
as compared to 60% for heat-damaged ones, as indicated in Table 6.
This suggested that heat-damaged beams had recovered fully their
original load capacity only for case when NSM CFRP strips were
embedded at a length of 450 mm beyond the critical stress region.
Intact and strengthened beams achieved higher percentage
increase of their load capacity, stiffness, and toughness yet attained
lower percentage increase in their rotational ductility as compared
to heat-damaged and repaired ones. The percentage increases in
load capacity upon repair/strengthening achieved as high as 40%
for heat-damaged as compared to 54% for intact ones. It is clear
that the percentage improvement in load capacity was enhanced
when the embedment length was increased from 150 mm to 250
and 450 mm; owning to the expected improvement in bond
between NSM CFRP strips and concrete. As may be expected, using
three strips of NSM CFRP imparted a slight increase to strength-
ened beams yet decrease to heat-damaged ones; because of pre-
mature concrete cover peeling off, noticed at failure.
Other mechanical characteristics: The residual rotational duc-
tility of heat-damaged/repaired beams exceeded, by far, those of
intact/strengthened beams; except for those repaired with three
NSM CFRP strips at an embedment length of 150 mm; owning to
premature concrete cover peeling off. The stiffness, on the other
hand, was recovered for intact/strengthened with residuals rang-
ing from 105 to 113% but not for heat-damaged/repaired beams
with residual ranging from 42 to 67%. Toughness results indicated
much lower residuals for heat-damaged/repaired than correspond-
ing intact/strengthened beams. Upon the application of two NSM
CFRP strips at varying embedment lengths from 150 to 450 mm,
residual toughness was reduced for heat-damaged by as much as
34% yet was increased by as high as 46% for intact beams. In gen-
eral, the contribution of three NSM CFRP strips at an embedment
length of 150 mm to the mechanical performance of strengthened
or repaired beams was very limited and sometime negative, hence
deemed as inappropriate repair/strengthening alternative.
where SF is the strength factor calculated as the ratio of strength of
strengthened/repaired to that of intact beams; and DF is the defor-
mity factor calculated as the ratio of rotational ductility of strength-
ened/repaired to that of intact beams.
The values of performance factor (PF) of Table 7 showed that
the present level of heat-damage had determinately affected the
efficiency of the present repair techniques: PF ranged from 1.34
to 1.45 for intact/strengthened as compared to 0.80–1.05 for
heat-damaged/repaired beams. Furthermore, it can be deduced
that the degradation in the performance factor of heat-damaged
and repaired beams is referred to the reduction in the strength
factor.
The strain, induced in NSM CFRP strips at failure and reported in
Table 6, were consistent with the structural capacity values. The
ratios of strain in the double NSM CFRP strips, embedded in intact
beams at extensions of (150, 250, and 450 mm), to corresponding
strips, embedded in heat-damaged beams, were 1.26, 1.76, and
1.87, respectively. As can be expected, close strain values were
obtained in triplicate NSM CFRP strips, inserted in either intact or
heat-damaged to an embedment length of 150 mm beyond the
critical stress region.
5.4. Bond behavior between concrete and NSM CFRP strips
The bond-slip relationship between NSM CFRP strips and con-
crete was obtained for certain CFRP strengthened/repaired RC
beams, as depicted in Fig. 11. The curves showed bilinear behavior
prior to failure; regardless of the beam’s type. The curves charac-
teristics, namely bond strength and stiffness as well as slippage
at ultimate stress were obtained and are listed in Table 8.
As can be deduced form Table 8, bond strengths and stiffness
were improved as embedment length for the two CFRP strips
was increased from 150 to 250 mm; with much more pronounced
improvements for the case with heated-damaged beams. Further-
more, the obtained bond characteristics of Table 8 supported the
already stated fact that the efficiency of repair with NSM CFRP
was higher for intact and strengthened than that for heat-
damaged/repaired beams. The former beams achieved higher bond
strength and stiffness yet lower slippage at ultimate stress than
those of the latter ones, repaired using the same NSM CFRP strips’
configuration.

5.3. Efficiency of repairing heat-damaged beams using NSM CFRP

strips 5.5. Mode of cracking and failure under loading

To evaluate the overall performance of the repaired/strength-
ened beams, the performance factor was calculated according to
Eq. (1) given as:
PF ¼ SFxDF
Intact versus heat-damaged beams: Intact and heat-damaged
beams, subjected to four-point loading, showed a flexural type fail-
ure. Cracks initiated in the middle of the span at the tension side at
loads of 40 and 68 kN, respectively, before extended to the com-
ð5:1Þ
pression side with further load increase until failure at loads of

Table 6
Mechanical characteristics of control, heat-damaged, strengthened/repaired beams.

Beam Designation Pu (kN) Ppre (KN) RD Stiffness (MN/m) Toughness (J) f (m) f
fu

C 165 (100%)* NA 2 (100%) 21.9 (100%) 1891 (100%) NA NA

C SN2 ELð15Þ SP7:5 190 (115%) 214 2.4 (120%) 23.5 (107%) 2567 (136%) 4300 24%
C SN2 ELð25Þ SP7:5 219 (133%) 241 1.9 (95%) 23.8 (109%) 2140 (113%) 6700 38%
C SN2 ELð45Þ SP7:5 254 (154%) 282 1.7 (85%) 24.8 (113%) 2089 (111%) 11400 64%
C SN3 ELð15Þ SP5 188 (114%) 214 2.5 (125%) 23 (105%) 2760 (146%) 3400 20%
HD 99 (60%) NA 3.14 (157%) 13.2 (60%) 1700 (89%) NA NA
HD SN 2 ELð15Þ SP 7:5 113 (69%) 108 2.35 (118%) 13.3 (61%) 1250 (66%) 3800 22%
HD SN 2 ELð25Þ SP 7:5 137 (83%) 153 2.46 (123%) 13.8 (63%) 1417 (75%) 6100 34%
HD SN 2 ELð45Þ SP 7:5 165 (100%) 178 2.10 (105%) 14.7 (67%) 1290 (69%) 7500 42%
HD SN 3 ELð15Þ SP 5 92 (56%) 95 1.31 (66%) 10.5 (48%) 422 (22%) 3400 20%

*, Residual properties; Pu, Ultimate load capacity; Ppre: Predicted ultimate load capacity; RD, Rotational ductility;f: Strain in NSM CFRP at beam failure; fu: Ultimate strain
capacity of NSM CFRP strips; NA, not applicable.
640 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643

Table 7 zone prior to end cover separation failure mode, which initiated
Performance factors for intact and heat-damaged beams with different configurations at a load of 65 kN then propagated rapidly toward the high tension
of NSM CFRP strips.
zone until crushing of compression concrete.
Beams SF, % DF, % PF Similar failure modes were observed for beams, strengthened
C 100 100 1.00 with two NSM CFRP strips at embedment lengths of 150, 250
C SN2 ELð15Þ SP7:5 115 124 1.43 and 450 mm, as shown in photos of Fig. 13. Flexural cracks initi-
C SN2 ELð25Þ SP7:5 133 101 1.34 ated at the middle zone at loads of 50, 52 and 59 kN, respectively,
C SN2 ELð45Þ SP7:5 154 94 1.45
before spread along the entire span until end cover separation ini-
HD 60 157 0.94 tiated by the formation of vertical crack across the concrete cover
HD SN2 ELð15Þ SP7:5 68 118 0.80
HD SN2 ELð25Þ SP7:5 83 123 1.02
at the termination of the NSM CFRP strips at loads of 75, 100 and
HD SN2 ELð45Þ SP7:5 100 105 1.05 175 kN, respectively. This was followed by the propagation of a
major horizontal crack at the level of the tension reinforcement
SF, Strength factor calculated as the ratio of strength of strengthened or repaired to
towards the high tension zone, until crushing of concrete compres-
that of intact beams; DF, Deformity factor calculated as the ratio rotational ductility
of strengthened or repaired to that of intact beams; PF, performance factor calcu- sion zone. As may be expected, intact or heat-damaged beams,
lated as SF x DF. strengthened or repaired with three strips of NSM CFRP at
150 mm embedment length, respectively, showed failure patterns
similar to the above beams but at much lower cracking and end-
cover separation loads of those reported for the case with two
strips. End-cover separation loads reached 61 kN for intact/
strengthened as compared to and 57 kN for heat-damaged/
repaired beams.

6. Prediction of load capacity

An analytical prediction of ultimate load capacity for NSM CFRP

repaired/strengthened beams was carried out according to ACI 318
[30]. First, strains in concrete, and top and bottom steel were deter-
mined upon Eq. (1). Those are expressed in terms of the location of
the neutral axis, c, cross sectional geometric characteristics, and
ultimate strains measured in NSM CFRP strips and listed in Table 6.
The compressive forces in concrete were then computed according
Fig. 11. Bond Strength-slip curve for intact/heat-damaged beams strengthened/ to Hognestad model, incorporated into Eq. (2), with (eo and eu)
repaired using NSM CFRP strips.
taken as (0.002 and 0.003) for intact concrete and (0.025 and
0.035) for heat-damaged concrete, respectively [33,34]. The degra-
dation in the elasticity modulus for reinforcing steel was estimated
Table 8 according to the Euro code at a residual of 40% or original vale [33].
Bond characteristic of heat-damaged and intact beams with different configuration of
The forces in compression and tension steel and NSM CFRP
NSM CFRP strips.
strips are expressed in terms mechanical and geometric properties
Beams Bond Stiffness MPa/mm su kPa S u mm as well as strains, as shown in Eq. (3). After that, the neutral axis
C SN 2 ELð15Þ SP7:5 11.21 1300 0.62 location, c, is determined by trial and error based upon equilibrium
C SN 2 ELð25Þ SP7:5 39.08 1400 0.52 of the different forces, given by Eq. (4). Accordingly, the resultant of
HD SN 2 ELð15Þ SP 7:5 3.76 752 1.04 compression in concrete, C, and tension, T, forces, acting at dis-
HD SN 2 ELð25Þ SP 7:5 7.85 1294 0.90
HD SN 3 ELð15Þ SP 5 2.12 486 1.39
tances dc and dt from the top fiber of the concrete were obtained,
respectively. Finally, the theoretical ultimate moment and load
su , Ultimate bond strength; Su , Ultimate slippage. were computed using Eqs. (5) and (6).


c dc c  c0
2c ¼ 2f 2s ¼ 2f 20s ¼ 2f ð1Þ
df  c df  c df  c
165 and 98 kN, respectively. The steel jacket, affixed to the beams
throughout the high shear zone, prevented premature shear fail- where 2c ; 2s ; and 20s are strains in concrete, tension and compres-
ure, hence forced flexural failure. sion steel, respectively, in terms of distance between NSM CFRP
Repaired/strengthened beams: The failure modes for heat- strips and top concrete surface, df, depth of neutral axis from the
damaged beams, repaired with two NSM CFRP strips at embed- top surface of beams, c, and distance between compression steel
ment lengths of 150, 250 and 450 mm, are shown in Fig. 12. With center and upper concrete surface, c’.
load application, flexural cracks initiated at loads of 70 and 94 kN 8   2 
>
> 2 0 2c
throughout the middle zone for beams with NSM CFRP strips of >
< C ¼ 3
 f c  22c
2o
 2o
bc 0 < 2c < 2o
250 and 450 mm embedment lengths, respectively, before spread 
>   2  ð2Þ
along the beams spans. With further application of the load, end >
> C ¼ 2 f 0c 1  0:15 ð2c  2o Þ bc 2o < 2c < 2u
: 3 2u 2o
cover separation failure initiated by the formation of vertical crack
across the concrete cover at the termination points of the NSM 0
CFRP strips at loads of 95 and 100 kN, respectively, followed by where f c is the compressive strength of concrete whereas b is the
the propagation of a major horizontal crack at the level of the ten- widths of the beams.
8 0 0
sion reinforcement towards the high tension zone before crushing 0
Es 20s 6 f y
< C s ¼ As Es 2s
>
of compression zone. Hence, end cover separation was identified.
T s ¼ As Es 2s Es 2s P f y ð3Þ
For beams with NSM CFRP strips at an embedment length of >
:
150 mm limited cracking was observed within the high moment T f ¼ Af Ef 2f
 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643 641

Fig. 12. Failure modes for heat-damaged beams repaired with two CFRP strips at different embedment lengths (side view: left; bottom view: right).

Fig. 13. Mode of failure for intact beams strengthened with different embedment lengths of two NSM CFRP strips (side view: left; bottom view: right).
642 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643
where C 0s & T s are compression and tension forces in top and bottom
steel, respectively, and T f is the tension force in NSM CFRP strips. A0s
& As are cross-section areas of top and bottom steel reinforcement,
whereas Af is the cross-sectional area of NSM CFRP strips. E0s & Es are
elasticity moduli of top and bottom steel reinforcement, whereas
Ef is the elasticity modulus of NSM CFRP strips.
X X
F¼ ðT  CÞ ¼ 0
where C and T are compression and tension forces acting on the
beam cross section.
M ¼ ðC or TÞðdt  dc Þ
where M is the moment capacity of the beams, and dt and dc are
computed as follows:
3c
Cc þ C 0s ðc0 Þ
dc ¼ 8 [1] Y. Zhai, Z. Deng, N. Li, R. Xu, Study on compressive mechanical capability of
C c þ C 0s
T s ðdÞ þ T f ðdf Þ
dt ¼
Ts þ Tf
2M
Pu ¼
x [4] E.G. Choi, Y. Shin, H.S. Kim, Structural damage evaluation of reinforced
where P u the ultimate load capacity and x is the distance from
either support the nearest point load.
Theoretical load capacities for different strengthened beams are
listed in Table 6 along with those experimentally measured. The
error of prediction ranged from 3% to 13% with an error average
of 9%. These showed that the measured load capacity for different
beams were accurate and proportional to reinforcement quantities
and configuration.
7. Conclusion
Based on the above discussion, the following conclusions can be
made:
1. Exposure of present concrete beams to 600 °C for two hours led
to significant reductions in their load capacity, stiffness, and
toughness, yet imparted noticeable improvement to their duc-
tility: the strength factor (SF) was reduced by 40% upon heating
whereas the ductility factor (DF) was increased by 57%.
2. Strengthening/repairing reinforced concrete intact/heat-
damaged beams with NSM CFRP strips, respectively, con-
tributed to improving their load capacity, yet reducing their
ductility; depending upon the embedment length. The residual
ultimate load capacity and rotational ductility achieved its
highest and lowest at an embedment length of 450 mm.
3. Intact beams, strengthened using three NSM CFRP strips, showed
noticeable improvement in their load capacity and stiffness. In
contract, repairing heat-damaged beams with the latter configu-
ration caused degradation in their mechanical performance.
4. The overall performance factor, computed for intact and
strengthened beams ranged from 1.34 to 1.45, whereas that of
heat-damaged and repaired beams ranged from as low as
0.80–1.05; suggesting that the benefit from repairing the pre-
sent heat-damaged was feasible only when relatively long
embedment lengths for NSM CFRP strips beyond the high
moment region were used.
5. The significant degradation in bond strength between concrete
and NSM CFRP strips in heat-damaged and repaired beams jus-
tified the limited contribution of the present repair schemes in
restoring their original mechanical performance. The strains
induced in NSM CFRP strips at failure of heat-damaged and
repaired led to similar conclusion.
6. The analytical predictions for ultimate load capacity of repaired
and strengthened beams confirmed experimentally obtained
results with an average prediction error of 9%.
Acknowledgement
ð4Þ
The authors acknowledge the technical and financial support
provided by the research deanship at Jordan University of Science
and Technology (project number 153/2015) and the assistant by
the technicians at the structural and materials laboratory via the
ð5Þ Department of Civil Engineering.
References
concrete after high temperature exposure and thermo-damage constitutive
model, Constr. Build. Mater. 68 (2014) 777–782.
[2] Wu. Yaopeng, Wu. Bo, Residual compressive strength and freeze–thaw
resistance of ordinary concrete after high temperature, Constr. Build. Mater.
54 (2014) 596–604.
[3] Xu. Yuye, Wu. Bo, Ming Jiang, Xin Huang, Experimental study on residual
flexural behavior of reinforced concrete beams after exposure to fire, Adv.
Mater. Res. 457 (2012) 183–187.
ð6Þ
concrete beams exposed to high temperatures, J. Fire. Prot. Eng. 32 (2) (2013)
135–151.
[5] V.K.R. Kodur, A. Agrawal, An approach for evaluating residual capacity of
reinforced concrete beams exposed to fire, Eng. Struct. 110 (2016) 293–306.
[6] R.H. Haddad, R.H. Al-Saleh, N.M. Al-Akhras, Effect of elevated temperature on
bond between steel reinforcement and fiber reinforced concrete, Fire Saf. J. 43
(5) (2008) 334–343.
[7] R.H. Haddad, M.J. Shannag, R. Hamad, Repair of heat-damaged reinforced
concrete T-beams using FRC jackets, Mag. Concr. Res. 59 (2007) 223–231.
[8] R.H. Haddad, N. AL-Mekhlafy, A.M. Ashteyat, Repair of heat-damaged
reinforced concrete slabs using fibrous composites materials, Constr. Build.
Mater. 25 (2011) 1213–1221.
[9] R. Jones, R.N. Swamy, A. Charif, Plate separation and anchorage of reinforced
concrete beams strengthened by epoxy-bonded steel plates, Struct. Eng. 66 (5)
(1988) 85–94.
[10] M. Hussain, A. Sharif, I.A. Basunbul, M.H. Baluch, G.J. Al-Sulaimani, Flexural
behavior of pre-cracked reinforced concrete beams strengthened externally by
steel plates, Struct. J. 92 (1) (1995) 14–22.
[11] H. Al-Nimry, R.H. Haddad, S. Afram, M. Abdel-Halim, Effect of advanced
composites in repairing heat-damaged RC columns, Mater. Struct. 46 (2013)
1843–1860.
[12] M. Yaqub, C.G. Bailey, P. Nedwell, Axial capacity of post-heated square
columns wrapped with FRP composites, Cem. Concr. Compos. 33 (2011) 694–
701.
[13] M.G. Naghibdehi, M.K. Sharbatder, M. Mastali, Repairing reinforced concrete
slabs using composite layers, Mater. Des. 58 (2014) 136–144.
[14] F. Ceroni, M. Pecce, A. Bilotta, E. Nigro, Bond behavior of FRP NSM systems in
concrete elements, Composites Part B 43 (2012) 99–109.
[15] S. Gopinath, A.R. Murthy, H. Patrawala, Near surface mounted strengthening of
RC beams using basalt fiber reinforced polymer bars, Constr. Build. Mater. 111
(2016) 1–8.
[16] A.M. Khalifa, Flexural performance of RC beams strengthened with near
surface mounted CFRP strips, Alexandria Eng. J. (2016) 1–9.
[17] R.H. Haddad, H.H. Almasaeid, Recovering shear capacity of heat-damaged
beams using NSM CFRP strips, Constr. Build. Mater. 105 (2016) 448–458.
[18] C.N. Thi, W. Pansuk, L. Torres, Flexural behavior of fire-damaged reinforced
concrete slabs repaired with near-surface mounted (NSM) carbon fiber
reinforced polymer (CFRP) rods, J. Adv. Concr. Technol. 13 (2015) 15–29.
[19] V. Bianco, G. Monti, J.A. Barros, Design formula to evaluate the NSM FRP strips
shear strength contribution to a RC beam, Composites Part B 56 (2014) 960–
971.
[20] I.A. Sharaky, L. Torres, J. Comas, C. Barris, Flexural response of reinforced
concrete (RC) beams strengthened with near surface mounted (NSM) fiber
reinforced polymer (FRP) bars, Compos. Struct. 109 (2014) 8–22.
[21] H. Peng, J. Zhang, C.A. Cai, Y. Liu, An experimental study on reinforced concrete
beams strengthened with prestressed near surface mounted CFRP strips, Eng.
Struct. 79 (2014) 222–233.
[22] S. Seo, L. Feo, D. Hui, Bond strength of near surface mounted FRP plates for
retrofit of concrete structures, Compos. Struct. 95 (2013) 719–727.
[23] T.H. Almusallam, H.M. Elsanadedy, Y.A. Al-Salloum, S.H. Alsayed, Experimental
and numerical investigation for the flexural strengthening of RC beams using
near-surface mounted steel or GFRP bars, Constr. Build. Mater. 40 (2013) 145–
161.
[24] D. Lee, L. Cheng, Bond of NSM systems in concrete strengthening – examining
design issues of strength, groove detailing and bond-dependent coefficient,
Constr. Build. Mater. 47 (2013) 1512–1522.
 R.H. Haddad, O.A. Almomani / Construction and Building Materials 154 (2017) 632–643
[25] S.S. Zhang, J.G. Teng, Interaction forces in RC beams strengthened with near-
surface mounted rectangular bars and strips, Composites Part B 45 (2013)
697–709.
[26] P. Sabol, S. Priganc, Shear strengthening of concrete members using NSM
method, Procedia Eng. 65 (2013) 364–369.
[27] S.J. Dias, J.A. Barros, NSM shear strengthening technique with CFRP laminates
applied in high-strength concrete beams with or without pre-cracking,
Composites Part B 43 (2012) 290–301.
[28] K.N. Hong, J.W. Han, D.W. Seo, S.H. Han, Flexural response of reinforced
concrete members strengthened with near-surface-mounted CFRP strips, Int. J.
Phys. Sci. 6 (5) (2011) 948–961.
[29] T. Hassan, S. Rizkallah, Investigation of bond in concrete structures
strengthened with near surface mounted carbon fibers reinforced polymer
strips, J. Compos. Constr. ASCE 7 (3) (2003) 248–257.
643
[30] American Concrete Institute, Building Code Requirements for Structural
Concrete and Commentary (ACI-318), Farmington Hills, Michigan, 2008.
[31] American Society for Testing and Materials International (ASTM), West
Conshohocken, PA, 2004.
[32] ACI Manual of Concrete Practice, Standard practice for selecting proportions
for normal, heavyweight, and mass concrete (ACI 211.1), Part I: Materials and
General Properties of Concrete, Detroit, Michigan, 2008.
[33] E. Hognestad, N.W. Hanson, D. McHenry, Concrete stress distribution in
ultimate strength design, ACI J. Proc. 52 (4) (1955) 455–480.
[34] Eurocode 2, Design of concrete structures-Part 1-2. Europian Committee for
Standardization, Avenue Marnix 17 – B-1000, Brussels, Belgium, 1995.