Construction and Building Materials 114 (2016) 364–373

Contents lists available at ScienceDirect

Construction and Building Materials

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

Models for predicting elastic modulus and tensile strength of carbon,

basalt and hybrid carbon-basalt FRP laminates at elevated temperatures
Rami A. Hawileh a,⇑, Jamal A. Abdalla a, Sara S. Hasan a, Meena B. Ziyada a, Adi Abu-Obeidah b
a
Department of Civil Engineering, American University of Sharjah, Sharjah, United Arab Emirates
b
Department of Civil Engineering, Rutgers University, Piscataway, NJ, United States

h i g h l i g h t s

Mechanical properties of basalt and carbon FRP laminates with temperature.

Carbon, basalt and carbon-basalt sheets were used.

The elastic modulus and tensile strength were significantly reduced.

Analytical models were developed to predict the mechanical properties as a function of temperature.

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

Article history: Basalt fiber-reinforced polymers (B) composite laminates have been extensively used over the last decade
Received 7 February 2016 in externally strengthening of reinforced concrete (RC) slabs and beams in flexure and shear. Basalt fibers
Received in revised form 24 March 2016 have higher thermal resistance, corrosion resistance, and ductility than the commonly used carbon (C)
Accepted 29 March 2016
laminates. However, there is a lack of knowledge about the mechanical properties of such laminates
and their hybrid combinations when exposed to elevated temperatures. This paper presents the results
of an experimental program that studies the mechanical properties of carbon (C), basalt (B), and their
Keywords:
hybrid combinations (BC, CBC, CCB, BBC, and BCB) of multiple layers at elevated temperatures. The exper-
Reinforced concrete
Strengthening
imental program consists of 140 coupon specimens were prepared and tested after being exposed to dif-
FRP ferent temperatures ranging from 25 to 250 °C. The results showed that both the elastic modulus and the
Carbon, basalt tensile strength of the C and B laminates degraded with the increase in temperature. However, the degra-
Hybrid FRP systems dation was greater in the C composite sheets. Based on the experimental results, it was also observed that
Temperature dependent properties the mechanical degradation was the highest in C laminates, which reached to almost 90% at 250 °C. In
Mechanical properties addition, the elastic modulus and tensile strength values had shown that the BBC and B laminates had
the highest mechanical performance when exposed to elevated temperatures. In addition, analytical
models are proposed from the generated test data to predict the variation in the elastic modulus and ten-
sile strength with temperature. The obtained results and proposed models can be used as input param-
eters in the analysis and design of externally strengthened members with such FRP laminates. This study
strongly endorses the use of B and hybrid combination of B and C laminates in strengthening RC slabs and
beams.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction tures [1–3]. Strengthening with FRP laminates have been recog-
nized to be one of the most practical, effective, and efficient
Externally bonded fiber reinforced polymer (FRP) laminates are systems to increase the load-carrying capacities of concrete struc-
widely used in strengthening of reinforced concrete (RC) members. tural members. Recently, many studies showed a great improve-
These laminates have been used to improve flexural and shear ment in the load-carrying capacity of concrete members when
capacity of slabs and beams to adapt with the increase of the live strengthened with carbon, basalt, or glass fiber sheets [4–13]. On
load and to maintain a longer service life time for concrete struc- the other hand, there are few studies that investigated the effect
of using different combinations of carbon laminates with other
types of composites on the load-carrying capacity and ductility of
⇑ Corresponding author. structural concrete members [14–16]. Although FRP have been
E-mail address: [email protected] (R.A. Hawileh).

http://dx.doi.org/10.1016/j.conbuildmat.2016.03.175
0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373 365

successfully used in strengthening and repair of damaged RC slabs which is considered to be acceptable compared to the tensile
and beams there is a knowledge gap in the mechanical behavior of strength of dry fibers which is only 45%. Whereas, basalt bars
the different FRP laminates under elevated temperatures. Investi- showed a higher tensile strength at elevated temperatures due to
gating the performance of these sheets under a range of elevated their high glass transition temperature.
temperatures is a major issue that requires more attention by Foster and Bisby [20] investigated the effect of temperature on
researchers and engineers in order to provide FRP-strengthening the mechanical properties of carbon (C) and glass (G) sheets. The
systems that perform well under high environmental specimens were tested at temperatures of 20, 100, 200, 300, and
temperatures. 400 °C, respectively. The sheets were covered with epoxy on both
In this paper, the authors investigated the effect of elevated sides and then placed in the oven for 3 h. Then, tensile tests were
temperatures on the mechanical properties of carbon laminates conducted on the cured samples. The test results showed that
(C), basalt laminates (B) and their hybrid combinations (BC, BCB, the reduction in the tensile strength of the C sample was not signif-
CCB, CBC, and BBC) up to 3 layers. The coupon specimens were icant up to 300 °C. Whereas, G laminates showed a major loss in
heated in an oven at the specified temperature for 45 min and then the tensile strength at 200 °C.
left to cure at room temperature for 24 h prior to testing. The Sim et al. [21] conducted an experimental study to shed light on
obtained temperature-dependent material properties are the elas- the behavior of basalt fibers as a strengthening material for RC
tic modulus, the tensile strength, and the stress-strain response structures, compared to carbon and laminates. The three types of
curves. The mechanical properties could be used by engineers FRP sheets were heated at different temperatures that ranged from
and researchers as an input in the design and analysis of RC slabs 100 to 1200 °C for two hours and then tested after one day of cool-
and beams externally strengthened with such composite materials ing at room temperature. The results showed that the change in
when exposed to elevated temperatures. strength was not significant for the three material types when
the applied temperature was less than 200 °C. However, as the
temperature exceeded 200 °C, the reduction in strength for carbon
2. Literature review and glass was significant. On the other hand, the basalt fibers
retained 90% of its strength at 600 °C. Moreover, the volumetric
Limited studies investigated the performance of structural stability of carbon fibers was lost and the fibers were completely
members (slabs, beams, columns, etc.) externally strengthened molten, while the glass fibers were partially lost during heating.
with FRP composite sheets when exposed to elevated tempera- In addition to Sim et al. [21], other researchers investigated the
tures. An experimental program was conducted by Blontrock use of basalt fiber polymer (B) sheets in strengthening of RC mem-
et al. [17] to test the effect of fire on concrete slabs strengthened bers as an alternative to carbon and glass fibers [11–13]. Campione
with FRP sheets and protected with resistant plates with different et al. [12] studied the compressive strength of concrete cylinders
arrangements. The slabs were placed in the oven and subjected to a externally wrapped with basalt fiber sheets. A total of 26 mono-
sustained constant load applied at mid-span. The results showed tonic and cyclic compressive tests were performed. It was con-
that when the temperature of the adhesive ranged between 47 °C cluded that the specimens confined with the basalt sheets
and 69 °C, an excessive increase in the deflection occurs prior to exhibited a strain-softening behavior with a negligible increase
failure. in resistance but a significant increase in ultimate strain (up to 5
Similarly, Al-Salloum et al. [18] reported the effect of high tem- times the peak strain of unconfined concrete) corresponding to
peratures on 42 concrete cylinders wrapped with two types of FRP basalt failure and concrete crushing.
sheets. Fourteen cylinders were left unwrapped and the remaining Among the limited studies [22], Cao et al. [23] carried out an
28 specimens were wrapped with one layer of carbon and glass experimental study to examine the mechanical properties of CFRP
sheets, respectively. The specimens were then cured for 7 days at and hybrid CFRP-GFRP sheets at elevated temperatures up to
room temperature. Before performing the compression testing, 200 °C. It was observed that the elastic modulus and tensile
the specimens were placed in an oven at a temperature of 100 °C strength experienced a significant reduction at elevated tempera-
and 200 °C for time periods of 1–3 h. Then, a uniaxial compression tures. In addition, the authors [24,25] investigated the variation
test was conducted on every specimen till failure. The results indi- of the mechanical properties of carbon (C) laminates, glass (G) lam-
cated that the unwrapped samples presented significant strength inates, and their hybrid combination (CG) under temperatures
degradation at high temperatures with long exposure time which ranging from 25 to 300 °C. The results showed that the mechanical
caused a rapid loss of concrete compressive strength. However, degradation of the composite C and G sheets were more noticeable
the remaining specimens, wrapped with carbon and glass sheets, than that of the hybrid CG laminates. In particular, the modulus of
showed lower reduction in their strength since the heat required elasticity of the C, G, and CG was dropped by about 28%, 26% and
a longer time to be transferred to the concrete through the epoxy 9%, respectively, at an elevated temperature of 250 °C. They recom-
and the sheets. The results showed considerable loss in the com- mended that the measured values of the mechanical properties of
pressive strength of the unwrapped cylinder when exposed to ele- C, G, and CG at elevated temperatures can be used as a valuable
vated temperatures, compared to that of the wrapped ones. In input in computer modeling of concrete members externally
addition, the test results indicated that FRP composite materials strengthened with such composites.
are sensitive to fire and experience changes in strength and bond
performance when the ambient temperature exceeds the glass 3. Experimental program
transition temperature (Tg) of the polymer matrix used as an adhe-
sive/bonding agent. 3.1. Material properties
Few experimental studies investigated the behavior of compos-
MAPEWRAP-C-300-UNI-AX [26] carbon fiber fabric and FIDBASALT-UNIDIR-
ite laminates under high temperatures. Zhishen [19] investigated 400-C95 [27] basalt sheets were used in this study. It is noteworthy that these fab-
the FRP composite behavior under elevated temperatures and con- ric sheets are commonly used in FRP-strengthening applications of RC members.
cluded that the properties of FRP laminates are significantly influ- The adhesive used was a MapeWrap-31SP two-part epoxy adhesive provided by
enced when exposed to high temperatures. The results of the MAPEI [28]. The average measured glass transition temperature (Tg) of the epoxy
adhesive was 85 °C, as reported by the manufacturer. It was obtained through
experimental tests showed that the residual strength of the hybrid dynamic mechanical analysis (DMA). The material properties of these materials
basalt-carbon sheets which was 200 mm in length can maintain including the elastic modulus, tensile strength, elongation at rupture, and design
59% of the tensile strength at 200 °C after two hours of heating, thickness of the FRP sheets as reported by the manufacture are provided in Table 1.
366 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373

Table 1
Mechanical properties of the FRP sheets and epoxy adhesive at room temperature [24–26].

Elastic modulus (GPa) Tensile strength (MPa) Elongation at rupture (%) Sheet thickness (mm)
Carbon 230 4830 2.00 0.164
Basalt 95 3080 3.15 0.14
Epoxy adhesive 1 28 3.40 –

Table 2
Details and designation of the tested specimens.

Layers Laminate Designation Temperature (°C) Laminate thickness (avg.)

25 100 200 250
1 Carbon C 25/C 100/C 200/C 250/C 0.99
Basalt B 25/B 100/B 200/B 250/B 0.95
2 Basalt-carbon BC 25/BC 100/BC 200/BC 250/BC 1.49
3 Basalt-carbon-basalt BCB 25/BCB 100/BCB 200/BCB 250/BCB 2.14
Basalt-basalt-carbon BBC 25/BBC 100/BBC 200/BBC 250/BBC 2.65
Carbon-basalt-carbon CBC 25/CBC 100/CBC 200/CBC 250/CBC 1.38
Carbon-carbon-basalt CCB 25/CCB 100/CCB 200/CCB 250/CCB 1.46

3.2. Sample preparation laminates showed a reduction with the increase in temperature.
These findings are consistent with the outcomes of many previous
A total of 140 coupon specimens (7 for each set) were prepared and tested to
investigate the mechanical degradation of C and B composite sheets and their
studies conducted by several researchers [4,7,18,20,22–24]. It
hybrid combinations (BC, BCB, CCB, CBC, and BBC) under elevated temperatures should be noted that the tested samples failed by rupture of the
ranging from 25 °C to 250 °C. The width and total length of each coupon specimen composite laminates within the specimens gauge length, away
was 40 mm and 250 mm, respectively. Table 2 present the designation and details from the grips, as shown in Fig. 3. It was also observed that at room
for each group of specimens.
Each specimen was prepared by embedding every layer of FRP fibers between
two layers of epoxy adhesives. This simulates real strengthening applications,
where FRP sheets are externally bonded to concrete surfaces of RC members to
strengthen them in flexure and/or shear. The detailed procedure of the preparation
of these composite sheets is reported in a previous study conducted by the authors
[24,25]. As shown in Table. 2, the specimens in the first group consisted of either
one layer of carbon (C) or basalt (B) composite sheets. In the second group, the spec-
imens were composed of a hybrid combination of carbon-basalt (BC) composite
sheets. The coupon specimens of this group were prepared by applying a layer of
epoxy adhesive on a nylon sheet, followed by a layer of C sheet, epoxy adhesive,
B sheet, and finally covered with an additional layer of epoxy adhesive. The
specimens in the third group were composed of a hybrid combination of three lay-
ers of C and B sheets and were prepared in a similar fashion to group two, and were
assembled in four different ways. This group included a hybrid combination of
basalt-carbon–basalt (BCB), basalt-basalt-carbon (BBC), carbon-basalt-carbon
(CBC), and carbon-carbon-basalt (CCB) composite sheets. It should be noted that
the laminates were cured for one week before being cut into coupon specimens.
The coupon laminates were prepared and tested according to the ASTM D3039/
D 3039M-08 [29] guidelines. The specimens were heated for 45 min in a standard
Carbolite furnace at different specified temperatures of 100, 200 and 250 °C. The
temperature inside the furnace can reach up to 1100 °C. The heating time of
45 min is a standard practice, as it is sufficient to uniformly distribute the temper-
ature within the specimen. However, the effect of heating time on the mechanical
properties of the laminates should be examined in a future research study as an
additional parameter. The furnace is also equipped with an automatic digital sys-
tem to set and control the temperature inside the furnace. The specimens were then
left to cure at room temperature for 24 h prior to testing. Control specimens were
also prepared for each sample series by curing at room temperature for comparison
purposes.

3.3. Testing procedure

The tensile tests were conducted using an Instron universal testing machine at a
rate of 2 mm/min. Fig. 1 shows the prepared coupon specimens and the testing
set-up. The specimens were gripped for a length of 50 mm from both ends. Thus,
the gauge length of the tested specimens is 150 mm.

4. Results and discussions

Temperature-dependent stress-strain response curves of the

tested specimens were plotted using the experimental results.
Fig. 2 shows results of selected samples from every set of tested
specimens. According to the stress-strain curves presented, both
the elastic modulus and tensile strength of the composite Fig. 1. Coupon specimens and testing set-up.
R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373 367

Fig. 2. Stress-strain curves.

368 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373

Fig. 4. Normalized elastic modulus versus temperature.

Fig. 3. Typical failure of tested specimens.

Table 3
Test results for the elastic modulus and tensile strength.

Layers Specimen E RSDa ET/ fu RSDa fu,T/

designation (MPa) (%) E25 (MPa) (%) fu,25
1 25/C 37258 22 1.00 578 5.1 1.00
100/C 11836 12 0.32 413 13 0.71
200/C 4112 25 0.11 136 29 0.24
250/C 3382 19 0.09 112 20 0.19
25/B 9023 11 1.00 242 14 1.00
100/B 8998 4.2 0.99 258 2.4 1.07
200/B 6675 3.6 0.74 162 10 0.67
250/B 6559 12 0.73 143 13 0.59 Fig. 5. Normalized tensile strength versus temperature.

2 25/BC 11254 14 1.00 405 15 1.00

100/BC 12681 21 1.13 389 13 0.96
200/BC 6863 3.8 0.61 194 2.2 0.48 Table 4
250/BC 4602 17 0.41 137 8.9 0.34 Percent degradation in the mechanical properties of the FRP laminates at 250 °C.
 
3 25/BCB 10604 25 1.00 431 23 1.00 X 250 X 25
Degradation % = X 25  100
100/BCB 12415 19 1.17 303 23 0.70
200/BCB 8613 25 0.81 199 21 0.46 Elastic modulus Tensile strength
250/BCB 5827 17 0.55 195 25 0.45
C 90.92 80.53
25/BBC 8964 20 1.00 291 19 1.00
CBC 84.25 79.87
100/BBC 7135 21 0.79 248 26 0.85
BC 59.11 66.25
200/BBC 6965 25 0.78 266 26 0.91
CCB 56.58 52.58
250/BBC 6262 19 0.69 218 22 0.75
BCB 45.04 54.63
25/CBC 23904 4.5 1.00 650 17 1.00
B 27.31 40.95
100/CBC 7859 12 0.33 294 24 0.45
BBC 30.14 24.90
200/CBC 4660 16 0.19 187 9.3 0.29
250/CBC 3764 0.4 0.16 131 25 0.20
25/CCB 26093 4.5 1.00 853 19 1.00
100/CCB 23972 12 0.92 749 17 0.88
200/CCB 11723 16 0.45 391 24 0.46 in Table 3. Furthermore, the normalized elastic modulus (ET/E25)
250/CCB 11329 0.4 0.43 404 22 0.47 and tensile strength (fu,T/fu,25) as a function of increasing tempera-
a
RSD: relative standard deviation. ture for all specimen series are shown in Figs. 4 and 5, respectively.
Table 4 also presents the average percent degradation of the elastic
modulus and tensile strength at 250 °C for all types of tested FRP
temperature, the specimen failed in a brittle fashion by cracking of laminates.
the epoxy adhesive. However, at higher temperatures, the failure According to Figs. 3–5 and Tables 3 and 4, the degradation in the
was more ductile due to the softening of the epoxy adhesive that mechanical properties at elevated temperatures was smaller for the
lead to debonding of the fibers from the epoxy matrix. laminate specimens with one layer of basalt fibers or more. In par-
Table 3 illustrates the results of the average elastic modulus (E) ticular, it is clearly indicated in Table 3 and Fig. 4 that the percent
and tensile strength (fu) along with their relative standard devia- degradation of the elastic modulus for the C laminates at 100,
tions (RSD) for every series of specimens at 25, 100, 200, and 200, and 250 °C was 68, 89, and 91%, respectively. However for
250 °C. The range of RSD of the elastic modulus and tensile the B laminates, the percent degradation of the elastic modulus at
strength of the tested specimens is 0.4%–26% and 5.1–29%, respec- 100, 200, and 250 °C was 29, 76, and 81%, respectively. It can be also
tively. The relatively higher computed RSD values for both the elas- noted from Table 4 that at 250 °C, the average percent degradation
tic modulus and tensile strength is attributed to the sensitivity in of the elastic modulus for the C, B, and BC specimens was 91, 27, and
measuring the thickness of the composite specimens. The ratio of 59%, respectively. Among the laminates with three layers of FRP
the elastic modulus (ET/E25) and tensile strength (fu,T/fu,25) at each fibers, the BBC specimens had the highest stiffness with an average
specified temperature to that at room temperature is also reported degradation of 30% in the elastic modulus, compared to 84% for CBC
 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373 369

and 57% for CCB. According to Fig. 5 and Table, the degradation of kE ¼ 1 for 0 6 T 6 100
ð6Þ
the tensile strength for the C laminates was significantly greater kE ¼ 1:175  0:00175T for 100 6 T 6 300
than that of the B laminates. It can be noted from Table 3 that the
percent degradation of the tensile strength for the C laminates at
kf ¼ 1 for 0 6 T 6 100
100, 200, and 250 °C was 29, 76, and 81%, respectively. However ð7Þ
for the B laminates, the percent degradation of the tensile strength kf ¼ 1:267  0:00267T for 100 6 T 6 475
at 100, 200, and 250 °C was 7, 33, and 41%, respectively. Moreover Bisby [33] also developed an empirical model, presented in
it is clearly indicated in Table 4 that the B and BBC laminates Eq. (8), to predict the degradation of the elastic modulus and ten-
showed the highest tensile strength with an average degradation sile strength for FRP materials with temperature.
of 41 and 25%, respectively at 250 °C. The basalt fibers enhanced
   
the fire resistance of the carbon sheets for all hybrid combinations. PðTÞ 1a 1þa
¼ tanh½bðT  cÞ þ ð8Þ
The degradation in the elastic modulus was reduced up to about PU 2 2
59% for the BC laminates. In addition, a second layer of basalt will
reduce the degradation of the elastic modulus up to about 45% where, a is an assumed constant and b and c are parameters
and 30% for the BCB and BBC laminates, respectively. It should also obtained by experimental data fitting. The values of these three
be noted from Table 3 that the percent level of degradation in the parameters for the elastic modulus and tensile strength of CFRP
mechanical properties of the BBC laminate is lower than that of and GFRP materials were also reported by Bisby [33].
the BCB, since in the BBC laminate, the C sheet is directly exposed Figs. 6 and 7 show the measured data and predicted using the
to temperature, while in the BCB laminate, the C sheet is embedded available models of the elastic modulus and tensile strength for
between two layers of B sheets. Based on the experimental results, the C and B composite sheets, respectively. It is clearly indicated
the fire resistance of RC members strengthened externally with from Figs 6 and 7 that the available models cannot predict the
basalt laminates is significantly higher than that of RC members degradation in the elastic modulus and tensile strength of the C
strengthened with carbon laminates. and B composite sheets that were prepared using the wet-layup
technique. As discussed previously, the published models were
based on test data for CFRP and GFRP bars, and CFRP pultruded
5. Comparison of test results with prediction models
laminates. This is the main reason for the deviation between the
experimental and predicted reason. The main reason of such
Firmo et al. [22] summarized the available prediction models
deviation between the obtained data and analytical results for
that could predict the degradation in the elastic modulus and ten-
the elastic modulus and tensile strength with temperature of the
sile of FRP materials at elevated temperatures. It should be noted
tested laminates is attributed to different composition of the fibers
that the published models are based on coupon tests conducted
used within the matrix of pultruded CFRP plates and strips, FRP
for carbon (C) and glass (G) FRP materials. However, the test data
of this study was for C and B composite sheets and their hybrid
combinations (BC, BCB, CCB, CBC, and BBC).
Most of the proposed relations for the variation of the mechan-
ical properties of FRP materials with temperature were based
on the general model proposed by Gibson et al. [30], as given in
Eq. (1).
 
PU þ PR PU  PR 
PðTÞ ¼ Rn  tanh km ðT  T g;mech Þ ð1Þ
2 2
where, P(T) is either the elastic modulus or tensile strength at a
specified temperature T, PU is the material’s elastic modulus or ten-
sile strength at room temperature, PR is the material’s relaxed mod-
ulus or tensile strength before decomposition, Rn is a power law
that accounts for the residual resin content and suggested to be
Rn = 1 since the strength in the FRP laminates is dependent on the
strength of the fibers after decomposition of the resin adhesive,
km and Tg,mech are variables determined by fitting the experimental
data of the laminate using regression analyses.
Yu and Kodur [31] used Eq. (1) to predict the degradation in the
elastic modulus and tensile strength of CFRP pultruded strips and
bars as follows:
CFRP pultruded strip:
Elastic modulus E(T) and tensile strength f(T):

EðTÞ ¼ 0:51  0:49 tanh½0:0035ðT  340Þ ð2Þ

f ðTÞ ¼ 0:56  0:44 tanh½0:0052ðT  305Þ ð3Þ

CFRP bars:

EðTÞ ¼ 0:51  0:49 tanh½0:0033ðT  320Þ ð4Þ

f ðTÞ ¼ 0:54  0:46 tanh½0:0064ðT  330Þ ð5Þ

Saafi [32] proposed the following degradation factors of the elas- Fig. 6. Comparison of measured and predicted results using published models for
tic modulus (kE) and tensile strength (kf) relations for CFRP bars. the C laminates.
370 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373

tive to achieve a minimum square error between the predicted and

obtained test data for the elastic modulus and tensile strength with
temperature. The calculated coefficients of the elastic modulus and
tensile strength for the tested FRP laminates are provided in Tables
5 and 6, respectively. Based on these parameters the following
models, Eqs. (9)–(11), which are based on Gibson et al. [30] equa-
tion, are proposed for predicting the elastic modulus of carbon (C),
basalt (B) and carbon-basalt (BC) at elevated temperatures.

EC ðTÞ ¼ 0:545  0:455 tanhf0:014ðT  60:27Þg ð9Þ

EB ðTÞ ¼ 0:86  0:140 tanhf0:035ðT  163:24Þg ð10Þ

EBC ðTÞ ¼ 0:705  0:295 tanhf0:091ðT  196:34Þg ð11Þ

Similar models, Eqs. (12)–(14), are also proposed for predicting
the tensile strength of carbon (C), basalt (B) and carbon-basalt (BC).

f C ðTÞ ¼ 0:594  0:405 tanhf0:017ðT  117:74Þg ð12Þ

f B ðTÞ ¼ 0:795  0:205 tanhf0:075ðT  190:58Þg ð13Þ

f BC ðTÞ ¼ 0:67  0:33 tanhf0:087ðT  192:41Þg ð14Þ

Figs. 8 and 9 show a comparison between the predicted and
measured results for the elastic modulus and tensile strength with
temperature for all the tested FRP laminates. It is clearly indicated
from Figs. 8 and 9 that there is a close correlation between the pre-
dicted and measured results within the tested temperature range.
As shown in Table 5, the predicted elastic modulus values using the
models developed in this study had very small Normalized Mean
Square Error (NMSE) with a maximum value of 0.0479, Mean
Fig. 7. Comparison of measured and predicted results using published models for Absolute Percent Error (MAPE) ranging between 0.17% and 3.67%,
the B laminates.
and correlation coefficient (R) almost equal to one for most cases.
Table 6 shows the NMSE, MAPE and correlation coefficients for
laminates prepared using the wet layup process, NSM strips, and predicting the elastic tensile strength. It is observed that the max-
FRP bars. In addition, it is clearly shown in Fig. 6 that the elastic imum NMSE is 0.336, MAPE ranges between 0.1% and 5.63%, and R
modulus and tensile strength of the C laminates were noticeably is almost equal to one for almost all cases. This demonstrated the
lower than the predicted results using the existing models, which accuracy of the developed models.
indicates that C laminates when prepared using the wet layup Thus, the proposed models of the elastic modulus and tensile
technique possess lower resistance to elevated temperatures. This strength could be safely used as input parameters to evaluate the
is due to the embedment of the fibers between two layers of epoxy temperature response in the analysis and design (especially in
adhesive. Accordingly, empirical relations for the degradation of computer modeling) of externally strengthened concrete members
the elastic modulus and tensile strength with temperature of the with such laminates.
C, B, BC, BCB, CCB, CBC, and BBC laminates are proposed in the sub-
sequent section, based on the general model of Gibson et al. [30]. 7. Summary and conclusion

6. Proposed models for elastic strength and tensile strength This paper investigated the mechanical degradation of carbon
with temperature (C), basalt (B) laminates, and their hybrid combinations up to three
layers (CB, BCB, BBC, CCB, CBC) at elevated temperatures. Tensile
A least square regression analysis was performed to obtain the tests were conducted and the elastic modulus and tensile strength
km and Tg,mech coefficients of Eq. (1) for both the elastic modulus were obtained from the resulting stress-strain curves. Based on the
and tensile strength [30]. These coefficients were determined by experimental results, the following observations and conclusions
calibrating the test results, presented in Section 4, with the objec- were drawn:

Table 5
Derived coefficients based on Gibson et al. [30] model for the elastic modulus of FRP laminates.

Coefficient C B BC BCB BBC CBC CCB

PU 1.00 1.00 1.00 1.00 1.00 1.00 1.00
PR 0.09 0.72 0.41 0.55 0.60 0.15 0.43
km 0.014 0.035 0.091 0.126 0.003 0.009 0.026
Tg,mech 60.27 163.24 196.34 201.33 108.83 23.45 134.97
NSME 0.00003 0.00056 0.04797 0.02378 0.02965 0.00027 0.00003
MAPE (%) 1.468 0.242 2.881 3.669 2.044 2.323 0.170
R 0.99998 0.99985 0.98750 0.99273 0.98507 0.99987 0.99999
 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373 371

Table 6
Derived coefficients based on Gibson et al. [30] model for the tensile strength of FRP laminates.

Coefficient C B BC BCB BBC CBC CCB

PU 1.00 1.00 1.00 1.00 1.00 1.00 1.00
PR 0.19 0.59 0.34 0.45 0.75 0.20 0.47
km 0.017 0.075 0.087 0.018 0.004 0.008 0.058
Tg,mech 117.74 190.58 192.41 95.70 103.23 53.78 110.41
NSME 0.00003 0.02534 0.00470 0.00001 0.33632 0.00342 0.00065
MAPE (%) 0.59614 1.54508 1.22674 0.10240 5.01116 5.63169 0.84704
R 0.99998 0.99356 0.99879 0.99999 0.81496 0.99835 0.99973

Fig. 8. Comparison of measured and predicted elastic modulus results using the proposed models.

1. The mechanical properties of FRP composite sheets degraded by members and fire resistance of the FRP strengthening system
increasing the temperature. The carbon laminates had the high- under elevated environmental temperatures.
est elastic modulus and tensile strength at room temperature. 3. The obtained test data and proposed empirical models for
However, the carbon laminates showed the greatest mechanical the elastic modulus and tensile strength as a function of
degradation at elevated temperature. Thus, the carbon lami- temperature could be used in the analysis and design of
nates should be well-insulated to maintain their mechanical externally strengthened reinforced concrete members with
properties at elevated temperatures. the investigated laminates. In particular, the obtained results
2. Basalt laminates showed by far a better mechanical perfor- and proposed models could be used as input parameters in
mance at elevated temperatures, compared to the carbon lami- numerical modeling of externally strengthened concrete
nates. It is highly recommended to use a hybrid combination of members with such FRP laminates, when subjected to elevated
basalt and carbon laminates to enhance the strength of RC temperatures.
372 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373

Fig. 9. Comparison of measured and predicted tensile strength results using the proposed models.

It is noteworthy that the main problem for the composite References

application is the adhesive, which is considered to be the weakest
[1] O.A. Rosenboom, T.K. Hassan, A. Mirmiran, S. Rizkalla, Static and fatigue
part of the composite sheet. Thus, the outcomes of the tests are performance of 40 year old prestressed concrete girders strengthened with
depended on the type of the used epoxy adhesive matrix. Thus, various CFRP systems, in: Proceedings of the 2nd International Conference on
the effect of other types of commercially available epoxy adhesives FRP Composite in Civil Engineering (CICE), 2004. Adelaide, Australia.
[2] J.P. Firmo, M.R.T. Arruda, J.R. Correia, C. Tiago, Flexural behaviour of partially
on the performance of hybrid laminates produced by other manu-
bonded carbon fibre reinforced polymers strengthened concrete beams:
facturers under elevated temperatures should be investigated in application to fire protection systems design, Mater. Des. 65 (2015) 1064–
future research studies. In addition, the effect of heating time on 1074.
the mechanical properties of such FRP laminates should be exam- [3] A. Abu-Obeidah, R.A. Hawileh, J.A. Abdalla, Finite element analysis of
strengthened RC beams in shear with aluminum plates, Comput. Struct. 147
ined in a future research study. (2015) 36–46.
[4] K.H. Tan, Z. Yuqian, Performance of FRP-strengthened beams subjected to
elevated temperatures, J. Compos. Constr. ASCE 15 (3) (2010) 304–311.
[5] Y. Zhou, M. Gou, F. Zhang, S. Zhang, D. Wang, Reinforced concrete beams
Acknowledgement strengthened with carbon fiber reinforced polymer by friction hybrid bond
technique: experimental investigation, Mater. Des. 50 (2015) 130–139.
This research project was supported by the American University [6] H. Akbarzadeh, A.A. Maghsoudi, Experimental and analytical investigation of
reinforced high strength concrete continuous beams strengthened with fiber
of Sharjah (AUS). Their support is highly acknowledged. The
reinforced polymer, Mater. Des. 31 (3) (2010) 1130–1147.
authors would like to thank and acknowledge the support of [7] Y. Kim, K. Harries, Statistical characterization of reinforced concrete beams
MAPEI for supplying the CFRP Wraps and KERAKOLL Spa for strengthened with FRP sheets, J. Compos. Constr. ASCE 17 (3) (2013) 357–370.
supplying the BFRP Wraps, along with their adhesives. Their [8] P. Fanning, O. Kelly, Ultimate response of RC beams strengthened with CFRP
plates, J. Compos. Constr. ASCE 5 (2) (2001) 122–127.
support is gratefully acknowledged. The authors would like to also [9] H. Akbarzadeh, A.A. Maghsoudi, Experimental and analytical investigation of
acknowledge the help of Eng. Ihab Karam, Laura Al Rifai and Farah reinforced high strength concrete continuous beams strengthened with fiber
Kazmouz for their help in conducting the tests. reinforced polymer, Mater. Des. 31 (3) (2010) 1130–1147.
 R.A. Hawileh et al. / Construction and Building Materials 114 (2016) 364–373 373

[10] R. Barnes, J. Fidell, Performance in fire of small-scale CFRP strengthened [22] J.P. Firmo, J.R. Correia, L.A. Bisby, Fire behaviour of FRP-strengthened
concrete beams, J. Compos. Constr. ASCE 10 (6) (2006) 503–508. reinforced concrete structural elements: a state-of-the-art-review, J.
[11] J. Sim, C. Park, Characteristics of basalt fiber as a strengthening material for Compos. B Eng. 80 (2015) 198–216.
concrete structures, Compos. B Eng. 36 (6) (2005) 504–512. [23] Shenghu Cao, Zhis Wu, Xin Wang, Tensile properties of CFRP and hybrid FRP
[12] G. Campione, L. La Mendola, A. Monaco, A. Valenza, V. Fiore, Behavior in composites at elevated temperatures, J. Compos. Mater. 43 (4) (2009) 315–
compression of concrete cylinders externally wrapped with basalt fibers, 330.
Compos. B Eng. 69 (2015) 576–586. [24] R.A. Hawileh, A. Abu-Obeidah, J.A. Abdalla, A. Al-Tamimi, Temperature effect
[13] D. Shen, H. Shi, Y. Ji, F. Yin, Strain rate effect on effective bond length of basalt on the mechanical properties of carbon, glass and carbon–glass FRP laminates,
FRP sheet bonded to concrete, Constr. Build. Mater. (2015) 206–218. Constr. Build. Mater. 75 (2015) 342–348.
[14] N. Attari, S. Amziane, M. Chemrouk, Flexural strengthening of concrete beams [25] R.A. Hawileh, A. Abu-Obeidah, J.A. Abdalla, A. Al-Tamimi, Residual mechanical
using CFRP, GFRP and hybrid FRP sheets, Constr. Build. Mater. 37 (2012) 746– properties of carbon and hybrid carbon-glass FRP sheets at elevated
757. temperatures, in: Proceedings: The 7th International Conference on FRP
[15] A. Hosny, H. Shaheen, A. Abdelrahman, T. Elafandy, Performance of reinforced Composites in Civil Enginerring, 2014.
concrete beams strengthened by hybrid FRP laminates, Cem. Concr. Compos. [26] MapeWrap technical sheet for MAPEWRAP-C-300-UNI-AX, 2013.
28 (10) (2006) 906–913. [27] FIDBASALT technical sheet for Fidbasalt-UNIDIR-400-C95. Technical global
[16] R.A. Hawileh, H.A. Rasheed, J.A. Abdalla, A.K. Al-Tamimi, Behavior of reinforced services, 2004.
concrete beams strengthened with externally bonded hybrid fiber reinforced [28] Mapei technical sheet for MapeWrap-31SP two-part epoxy adhesive, 2014.
polymer systems, Mater. Des. 53 (2014) 972–982. [29] ASTM D3039/D 3039M-08, Standard Test Method for Tensile Properties of
[17] H. Blontrock, L. Taerwe, P. Vandevelde, Fire testing of concrete slabs Polymer-Matrix Composite Materials, 2008.
strengthened with fiber composite laminates, in: Proceedings: The 5th [30] A.G. Gibson, Y.S. Wu, J.T. Evans, A.P. Mouritz, Laminate theory analysis of
International Confernce on FRP Reinforced Concrete Structures, Thomas composites under load in fire, J. Compos. Mater. 40 (7) (2006) 639–658.
Telford, London, 2001, pp. 547–556. [31] B. Yu, V.K.R. Kodur, Effect of temperature on strength and stiffness properties
[18] Y. Al-Salloum, H. Elsanadedy, A. Abadel, Behavior of FRP-confined concrete of near-surface mounted FRP reinforcement, Compos. B Eng. 58 (2014) 510–
after high temperature exposure, Constr. Build. Mater. 25 (2) (2011) 838–850. 517.
[19] Z. Wu, X. Wang, G. Wu, Basalt FRP composite reinforcements in infrastructure, [32] M. Saafi, Effect of fire on FRP reinforced concrete members, Compos. Struct. 58
in: Proceedings: The 17th Annual International Conference on Composites/ (1) (2002) 11–20.
Nano Engineering (ICCE-17), 2009. New Orleans. [33] L.A. Bisby, Fire behaviour of fibre-reinforced polymer (FRP) reinforced or
[20] S. Foster, L. Bisby, Fire survivability of externally bonded FRP strengthening confined concrete PhD thesis in Civil Engineering, QUEEN’s University,
systems, J. Compos. Constr. 12 (5) (2008) 553–561. Kingston, Ontario, Canada, 2003.
[21] J. Sim, C. Park, D. Moon, Characteristics of basalt fiber as a strengthening
material for concrete structures, J. Compos. B Eng. 36 (6–7) (2005) 504–512.