Composites Part B 106 (2016) 262e272

Contents lists available at ScienceDirect

Composites Part B
journal homepage: www.elsevier.com/locate/compositesb

Bond durability of basalt-fiber-reinforced-polymer (BFRP) bars

embedded in concrete in aggressive environments
Mohamed Hassan a, b, Brahim Benmokrane a, *, Adel ElSafty c, Amir Fam b
a
Dept. of Civil Engineering, Univ. of Sherbrooke, Sherbrooke, Quebec, J1K 2R1, Canada
b
Dept. of Civil Engineering, Queen's Univ., Kingston, Ontario, K7L3N6, Canada
c
Civil Engineering, College of Computing, Engineering, and Construction, Univ. of North Florida, UNF, Jacksonville, FL, 32224, USA

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

Article history: Recently, basalt-fiber-reinforced-polymer (BFRP) bars have emerged as a promising alternative to glass-
Received 7 August 2016 fiber-reinforced-polymer (GFRP) bars. So far, however, BFRP bars have not been incorporated into design
Received in revised form standards and specifications. This is due to limited studies and lack of knowledge on the performance of the
13 September 2016
bars in concrete, in particular, their bond durability when exposed to aggressive environments. This paper
Accepted 13 September 2016
Available online 15 September 2016
presents some results of an extensive research program investigating the bond durability behaviors of BFRP
bars in concrete structures and the long-term bond-strength-retention predications of the BFRP bars on the
basis of short-term tests results. This research included testing deformed BFRP bars measuring 12 mm in
Keywords:
Bond strength
diameter. Pullout specimens were tested with direct tensile loading after being exposed to an alkaline
Composite solution (pH 12.9) for 1.5, 3, and 6 months at temperatures of 40
C, 50
C, and 60
C. This paper investigated
Basalt-fibers the effects of alkaline environment, exposure periods, and elevated temperatures on bond strength as well
Concrete as the degradation mechanism and mode of failure of the BFRP-reinforced specimens. In addition, optical
Durability microscopy and scanning electronic microscopy were used to investigate the degradation of BFRP bars
Pullout tested. The test results indicate an initial increase in the bond strength of the conditioned specimens as the
Degradation temperature increased compared to their unconditioned specimens. After 1.5 months of exposure, the
Bars
specimens conditioned at 50
C and 60
C, respectively, had bond-strength increases of 25% and 26%, while
the specimens conditioned at 40
C exhibited no noticeable changes (a minor decrease of 4.3%). Never-
theless, the bond strength of the conditioned specimens deteriorated during immersion. The highest bond-
strength reductions occurred in the conditioned specimens after 6 months of exposure at 40
C (a 16% loss),
followed by specimens conditioned at 50
C (7% loss) and 60
C (5% loss) compared to their counterparts at
1.5 months. Lastly, the long-term bond-strength-retention predications of the BFRP bars after 50 years of
service life in dry, moist, and moisture-saturated environments with mean annual temperatures between
5
C and 35
C ranged from 71% to 92%.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction
Currently, glass and carbon are the fibers used most widely in
manufacturing fiber-reinforced-polymer (FRP) composite bars. Basalt
fibers, the latest generation of FRP composites, are currently attract-
ing the interest of the research community and construction industry
[1]. Basalt fibers are processed from volcanic rock with a melting
process similar to that used for glass fibers. Basalt fibers, which are
* Corresponding author.
E-mail addresses: Mohamed.Hassan@USherbrooke.ca (M. Hassan), Brahim.
Benmokrane@USherbrooke.ca (B. Benmokrane), Adel.el-safty@unf.edu (A. ElSafty),
Amir.Fam@queensu.ca (A. Fam).
continuous, offer higher tensile strength than E-glass fibers, greater
strain at failure than carbon fibers, as well as good resistance to
chemical attack, high-corrosion resistance, thermal stability, low
water absorption, and high impact and fatigue resistance [2e5].
Recent advances in polymer technology have led to the devel-
opment of a new generation of FRP reinforcing bars, namely basalt-
FRP (BFRP) bars. The use of BFRP bars in construction applications is
relatively new. This type of structural material is expected to pro-
vide performance comparable or superior to that of glass-FRP
(GFRP), while being significantly cost-effective [6e9]. In order for
this new material to gain wide acceptance for use in the construc-
tion industry and inclusion in FRP standards and guides, its safety,
sustainability, and several aspects of its structural and mechanical
behaviors require investigation. One of these fundamental aspects is

http://dx.doi.org/10.1016/j.compositesb.2016.09.039
1359-8368/© 2016 Elsevier Ltd. All rights reserved.
 M. Hassan et al. / Composites Part B 106 (2016) 262e272 263

the bond-development phenomenon, which is a primary parameter
for BFRP's successful application as internal reinforcement in con-
crete structures [10]. Bar-bond characteristics affect bar anchoring,
lapesplice strength, concrete-cover requirement, serviceability, and
ultimate states. The long-term bond-strength durability also plays a
significant role in the long-term performance of concrete structures
incorporating internal FRP reinforcement [11,12].
Numerous research efforts have been put into investigating the
bond durability of GFRP and CFRP bars in concrete [11e13]. Very few
studies, however, have investigated the bond durability of BFRP bars
embedded in concrete and subjected to harsh environmental con-
ditions. El Refai et al. [14] investigated the bond durability of sand-
coated and helically grooved BFRP bars embedded in concrete.
Their work included various accelerated environments, including tap
water, seawater, elevated temperature, elevated temperature fol-
lowed by tap water, and elevated temperature followed by seawater.
The authors reported that moister environments caused enhanced
adhesion for all specimens at early loading stages. Moreover, expo-
sure to elevated temperatures of up to 80
C had a minor effect on the
bond strength of the tested bars. That notwithstanding, such envi-
ronments had a harmful effect on the bond strength at later stages
depending on the bar material's moisture absorption and its
manufacturing quality, regardless of the fiber material. Altalmas et al.
[15] conducted a supplemented study on the bond durability of the
same sand-coated BFRP bars in concrete exposed to acid, saline, and
alkaline solutions for duration of up to 90 days at 60
C. The study
revealed that BFRP specimens immersed in ocean water and alkaline
solution for 90 days exhibited a 25% reduction in bond strength
compared to 14% for BFRP specimens immersed in an acid solution. In
addition, the interlaminar shear between the FRP layers governed
the failure of both unconditioned and conditioned pullout speci-
mens. Dong et al. [16] investigated the effect of seawater environ-
mental conditions on bond strength of basalt-vinyl ester (BV), basalt-
epoxy (BE), and glass-vinyl ester (GV) FRP bars in concrete for ex-
posures periods of up to 60 days at 40
C. The test results indicated
that, after 60 days in 40
C seawater, the bond strength of the BV bars
and GV bars decreased by 9.1% and 7.1%, respectively, while the bond
strength of the BE bars remained essentially unchanged.
The lack of understanding of the bond-durability performance
of BFRP bars is a crucial hindrance to their wide acceptance in field
applications. It is worth mentioning that BFRP bars have not yet
been incorporated into design standards and specifications. An
extensive research project is being conducted at the University of
Sherbrooke to investigate the short- and long-term performance of
BFRP bars under real and simulated harsh environments as a pre-
liminary step in introducing these new bars into FRP codes and
materials specifications [7,8,17,18]. This paper presents an experi-
mental investigation aimed at assessing the bond durability of BFRP
bars embedded in concrete through accelerated tests in an alkaline
solution at different temperatures. This study provides insight into
how the bond behaves after long-term environmental condition-
ing. In addition, the long-term bond-strength retention after 50
years of service in dry, moist, and moisture-saturated environ-
ments (based on fib Bulletin 40 [19]) are predicted. The findings of
this work will contribute to integrating BFRP bars into North
American FRP codes and guides [10,20e23].
2. Experimental work
2.1. Materials properties
2.1.1. BFRP reinforcing bars
BFRP bars with a diameter of 12 mm (nominal cross-sectional
area of 113 mm2) and deformed external surfaces were used in
this study, as shown in Fig. 1. These bars are made of continuous
longitudinal basalt fibers bound together with a vinyl-ester resin
using a pultrusion process. The used basalt fibers are manufactured
by ASA.TEC GMBH (Austria) using a mixture of different natural
volcanic rocks. The color of the fibers is gold brown, the mono-
filament diameter is 13e20 mm, and a density of 2.6 g/cm3 [7]. The
physical and mechanical properties of the BFRP bars were deter-
mined in accordance with the ACI 440.6 M [20], CSA S807 [21] test
methods and the relevant ASTM standards. In addition, the physical
and mechanical properties of representative BFRP bars were
compared to the minimum requirements for FRP bars, as
mentioned in ACI 440.6 M [20] and CSA S807 [21]. Table 1

Fig. 1. BFRP bar 12 mm in diameter used in this investigation.

Table 1
Physical and mechanical properties of BFRP bars 12 mm in diameter.

Test Property 12-mm Specified limit

ACI [20] CSA [21]

Physical Cross-sectional area (mm2) 125 N/A N/A

Fiber content by weight (%) 81.5 55% (by vol.) 70 (by wt.)
Transverse CET (  10-6
C-1) 22.1 N/A 40
Density (gm/cm3) 2.09 ± 0.02 N/A N/A
Moisture uptake (%) 0.13 ± 0.006 <1 1.0 (D2)a; 0.75 (D1)a
Cure ratio (%) 98.0 ± 0.001 N/A 93 (D2)a; 95 (D1)a
Tg (
C) 117.0 ± 2.65 100 80 (D2)a; 100 (D1)a
Mechanicalb Ultimate tensile strength (ffu) (MPa) 1706 ± 40 e e
Modulus of elasticity (Ef) (GPa) 62.1 ± 5.6 39.3 GPa 40.0 GPa
Ultimate tensile strain (εu) (%) 2.52 ± 0.2 >1.2% >1.2%
Ultimate transverse shear strength (tu) (MPa) 272 ± 18 >124 MPa >160 MPa
Ultimate interlaminar shear strength (Su) (MPa) 68 ± 4.9 e e
Bond strength (t max) (MPa) 15.48 ± 0.4 >9.6 MPa >8 MPa
a
D1 and D2 classifications can be found in CSA [21].
b
The mechanical properties were calculated using the nominal cross-sectional area.
264 M. Hassan et al. / Composites Part B 106 (2016) 262e272
summarizes the physical and mechanical properties of the BFRP
bars tested. As depicted in Table 1, the test results show that the
newly developed 12 mm BFRP bar satisfies the minimum re-
quirements of ACI 440.6 M [20] and CSA S807 [21]. Further
research, however, is needed to assess the long-term durability of
the BFRP bars in an alkaline environment.
2.1.2. Concrete mix
The pullout specimens were cast using a ready-mixed, normal-
weight concrete with a target 28-day concrete compressive
strength of 30 MPa and 5%e8% of entrained air. The concrete
mixture consisted of 350 kg of Type 10 cement (corresponding to
ASTM I cement), 813 kg of fine aggregate, 1032 kg of coarse
aggregate, and 154 kg of water per cubic meter of concrete. Pullout
specimens and concrete cylinders were removed from the molds
24 h after casting, sealed with wet jute sheets, and cured with water
for one week. After curing, all of the specimens and cylinders were
left outside the laboratory for 4 weeks before being conditioned.
The concrete compressive strength (f’c) was determined on six
100  200 mm concrete cylinders for each batch. The 28-day
concrete compressive strengths ranged from 31 to 34 MPa.
2.2. Environmental conditioning
The pullout blocks were fully immersed in an alkaline solution
inside large steel tanks, while the free length of the BFRP bars
remained freely unconditioned inside the environmental chambers.
The alkaline solution consisted of 118.5 g Ca(OH)2 (calcium hy-
droxide), 4.2 g KOH (potassium hydroxide) and 0.9 g NaOH (sodium
hydroxide) in 1 L of deionized water, according to ACI 440.3R [23].
The alkaline solution's pH value was 12.9. The pullout specimens
were separated from each other, and the tank bottom allowed free
circulation of the solution between and around the blocks. The top of
the steel tank was hermetically sealed with a polyethylene film to
prevent excessive water evaporation during conditioning. In addi-
tion, the water level was kept constant throughout the study to avoid
the pH from increasing due to water evaporation. Fig. 2 shows the
test specimens inside the conditioning tank in the environmental
chambers. In order to accelerate the BFRP bareconcrete bond
degradation, the test specimens were subjected to elevated tem-
peratures of up to 60
C for exposure periods of up to 6.0 months.
2.3. Test specimens
The pullout specimens consisted of a 1200 mm long BFRP bar
embedded centrally in a concrete prism (200  200  200 mm).
The embedded length was kept constant at 5db, where db is the
Fig. 2. Test specimens in the conditioning tank inside the environmental chambers.
nominal diameter of the BFRP bar. Plastic tubes were used at the
loading end of the bar to minimize the stress concentration near
the loading plate. Steel pipes were used as anchors and were cast
with cement grout before testing. Fig. 3 depicts the typical details
and schematic for specimens inside the conditioning container. A
total of 50 specimens were tested under direct pullout tests. Forty-
five specimens divided into 3 groups (15 specimens in each group)
were fully immersed in an alkaline solution at three elevated
temperatures (40
C, 50
C, and 60
C) for 1.5, 3, and 6 months of
exposure. In addition, five unconditioned specimens were tested at
room temperature 28 days after casting, which served as a refer-
ence for comparison. At the end of each period, five similar speci-
mens from each set were removed from each environmental
chamber and tested at ambient laboratory temperature using direct
pullout tests. The specimens were denoted as follows: B for BFRP
bar; 1 to 5 for the specimen number in each set; and conditioning
exposure time (t) in months and environmental temperature (T) in
degrees Celsius. For example, specimen B5-(3e40) is reinforced
with a BFRP bar (B), is the fifth specimen in its set, and was exposed
for 3 months at a temperature of 40
C.
2.4. Test setup and test procedures
All specimens were subjected to pullout testing according to
ASTM [24]. The tests were carried out with a Baldwin testing ma-
chine in displacement control mode at a maximum rate of 20 kN/
min. The slip at the free end of the BFRP bar was measured with a
linear variable displacement transducer (LVDT). Fig. 4 shows the
test setup and instrumentation of the test specimens. The applied
load and bar slippage were recorded automatically throughout the
test with a data-acquisition system. The maximum bond stress at
the peak load tmax was calculated assuming a uniform bond-stress
distribution along the embedded length of the bar in concrete ld
using Eq. (1):
Pmax
tmax ¼ (1)
pdb ld
where tmax is the maximum bond stress (MPa), Pmax is the peak
load during pullout (kN), db is bar nominal diameter (12 mm), and ld
is the embedded length in the concrete (¼60 mm).
3. Test results and discussion
3.1. Failure mechanism
All of the conditioned BFRP-reinforced specimens failed in a
typical pullout mode of failure by exhibiting slip through the free
 M. Hassan et al. / Composites Part B 106 (2016) 262e272 265

Fig. 3. a) Test specimen geometry; b) schematic for specimens inside the conditioning tank.

Fig. 4. Pullout-test setup.

end with no signs of splitting cracks appearing on the cubes, which
is similar to that observed with the unconditioned specimens. Fig. 5
shows the typical mode of failure of the tested specimens. Some
prisms were split after testing for closer inspection of the concrete
surface along the embedded length. The visual examination of
these locations revealed some tiny chopped basalt fibers from the
bar coating surface attached to the concrete surface (see Fig. 5 a).
On the other hand, examination of the BFRP bar surface showed
concrete residue between the deformations, after pullout, and
minor scratches at some rib locations (Fig. 5 b). Similar behavior
was observed for both the unconditioned and conditioned tested
specimens. The final mode of failure of the tested specimens was
dominated by concrete shearing located at the BFRP bareconcrete
interface. The reason for the increased bond strength might be a
result of increased concrete strength produced by immersion [25].
That notwithstanding, an abrasive type of fracture in either the
concrete or the composite bar itself could be involved. This
behavior will be discussed further in the sections below.
266 M. Hassan et al. / Composites Part B 106 (2016) 262e272
Fig. 5. a) pullout-block cross section after cutting and b) typical BFRP bar surfaces after
testing.
3.2. Bond StresseSlip responses
Fig. 6 shows the bond-stress and free-end slip relationships for
all the specimens tested. The free-end slip was recorded directly
from the LVDT at the unloaded end. While the bond stress vs. free-
end slip curves showed similar trends for the unconditioned and
conditioned pullout specimens, the bond strengths were different
after environmental conditioning. The free-end slip remained zero
until the bond stress reached quite high levels, compared to the
ultimate bond stress. Thereafter, a pre-peak response was observed
up to the peak bond stress, followed by a post-peak response of up
to 87% (on average) of their peak stress. Finally, a sudden loss of
bond strength accompanied by a sudden free slip was observed. It
should be noted that not all of the specimens exhibited noticeable
measured slip changes at their unloaded ends (average of 1.3 mm)
until the peak pullout load was achieved, regardless of the envi-
ronmental conditions and exposure periods. Table 2 provides the
maximum and average free-end slips for all of the specimens.
3.3. Bond strength
Table 2 presents the peak bond load and corresponding
maximum bond stress for all unconditioned and conditioned
specimens. The average bond strength (based on an average of five
test results) was calculated as the pullout force over the embedded
area of the bar (Eq. (1)). As shown in Table 2, the average bond
strengths of the specimens conditioned for 1.5 months in an alka-
line environment at 40
C, 50
C, and 60
C were 14.8, 19.3, and
19.4 MPa, respectively. This indicates that the bond strength of the
BFRP bareconcrete interface remained unchanged or increased as
the temperature increased, compared to the bond strength of the
unconditioned specimens (15.5 MPa). After 1.5 months of exposure,
the specimens conditioned at 50
C and 60
C, respectively, had
bond-strength increases of 25% and 26%, while the specimens
conditioned at 40
C exhibited no noticeable changes (a minor
decrease of 4.3%). After 6 months of exposure, however, all the
conditioned specimens exhibited drops in bond strength with time
of exposure to the alkaline environment in comparison to condi-
tioned specimens after 1.5 months of exposure. The highest
reduction occurred in specimens conditioned at 40
C, which
experienced a 16% loss of bond strength, followed by the specimens
conditioned at 50
C (7% loss) and 60
C (5% loss). It should be noted
that the lowest bond strength after 6 months of exposure at 40
C
was 12.49 MPa, which is higher than the limits for bond strength
specified in ACI 440.6 M [20] (>9.6 MPa) and CSA S807 [21]
(>8 MPa).
3.4. Effect of temperature level on bond degradation
Fig. 7 (a and b) shows the effect of the temperature level on the
bond strength between the BFRP bars and concrete, and bond-
strength retention after exposure. As shown in Fig. 7 (a), the bond
strength tended to generally increase as the temperature increased.
Nevertheless, no significant differences were observed in the bond
strength of the specimens at 50
C and 60
C. Since the final mode
of specimen failure was governed by concrete shearing, the con-
crete's compressive strength increased during immersion, likely
resulting in increased bond strength. This can explain the increased
bond strength of specimens conditioned at elevated temperature.
The change in concrete strength was noted from testing six con-
crete cylinders immersed in alkaline solution for 6 months at 60
C.
The compressive strength of these cylinders was as high as 53 MPa,
compared to 33 MPa for the control cylinders. This observation is in
agreement with past studies on bond durability for pullout tests
involving BFRP and GFRP reinforcement [14,15,25e27]. The bond-
strength retentions of the conditioned specimens after 6 months
of exposure were 81%, 116%, and 119% at 40
C, 50
C, and 60
C,
respectively (see Fig. 7 (b)).
3.5. Effect of exposure time on bond degradation
Fig. 8 (a and b) shows the effect of exposure time on the strength
of the bond between the BFRP bars and concrete, and the bond-
strength retention at different elevated temperatures. As
mentioned above, the bond strength of the conditioned specimens
increased or remained unchanged up to 1.5 months compared to
their unconditioned counterparts due to the increase in concrete
strength. After 1.5 months of exposure at different elevated tem-
peratures, the specimens exhibited decreasing bond strength as the
immersion time increased. The mechanism underlying the bond-
deterioration mechanism might be as follows. The environmental
conditioning and elevated temperature creates gaps along the
bareconcrete interface due to variations in the coefficient of ther-
mal expansion (CTE) of both materials, resulting in a radial bursting
force on the concrete surface at the interface. If the concrete stress
is greater than the concrete's tensile strength, cracks will develop,
loosening the contact between the FRP bars and concrete, and
decreasing the bond strength [27]. Immersing the specimens in the
alkaline solution leads to chemical ions filling these voids, facili-
tating the attack on the FRP bar's polymeric matrixdespecially on
the outer surfacedand degrading the reinforcement itself, which
may play a role in bond-strength degradation [14,25,28,29]. The
microcracks and chemical reactions seem to counteract the effect of
the increased concrete strength, reducing specimen bond strength
[14].
 M. Hassan et al. / Composites Part B 106 (2016) 262e272 267

Fig. 6. Bond stresseslip curves: a) unconditioned specimens; b, c and d) specimens conditioned at 40
C, 50
C, and 60
C for 1.5 months; e, f, and g) specimens conditioned at 40
C,
50
C, and 60
C for 3 months; and h, i, and j) specimens conditioned at 40
C, 50
C, and 60
C for 6 months.
268 M. Hassan et al. / Composites Part B 106 (2016) 262e272

Table 2
Test matrix and bond-test results for basalt specimens after pullout testing.

Immersion Temperature (
C) Specimena Peak load (kN) t max (MPa) t aver. (MPa) COV (%) Retention (%) Smax (mm) Saver. (mm)
time (Months)

Unconditioned 23 B1- (0e23) 34.35 15.18 15.48 ± 0.44 2.9 100 1.28 1.39
B2- (0e23) 36.48 16.13 1.48
B3- (0e23) 35.59 15.73 1.35
B4- (0e23) 34.55 15.27 1.45
B5- (0e23) 34.10 15.07 e
1.5 40 B1- (1.5e40) 30.86 13.64 14.81 ± 1.51 10.2 96 1.32 1.22
B2- (1.5e40) 36.58 16.17 1.40
B3- (1.5e40) 37.28 16.48 1.08
B4- (1.5e40) 33.31 14.72 1.28
B5- (1.5e40) 29.50 13.04 1.10
50 B1- (1.5e50) 39.95 17.66 19.31 ± 1.60 8.1 125 1.24 1.32
B2- (1.5e50) 41.08 18.16 1.19
B3- (1.5e50) 42.68 18.87 1.45
B4- (1.5e50) 48.42 21.41 1.20
B5- (1.5e50) 46.22 20.43 1.50
60 B1- (1.5e60) 44.20 19.54 19.44 ± 0.8 4.1 126 1.10 1.14
B2- (1.5e60) 41.44 18.32 1.29
B3- (1.5e60) 46.08 20.37 1.55
B4- (1.5e60) 43.05 19.03 0.32
B5- (1.5e60) 45.13 19.95 1.44
3 40 B1- (3e40) 24.08 10.64 13.30 ± 3.10 23.1 86 1.38 1.25
B2- (3e40) 41.76 18.46 1.11
B3- (3e40) 27.86 12.32 1.01
B4- (3e40) 30.46 13.46 1.38
B5- (3e40) 26.25 11.60 1.36
50 B1- (3e50) 36.93 16.32 18.42 ± 1.95 10.6 119 1.76 1.53
B2- (3e50) 41.55 18.37 1.86
B3- (3e50) 37.71 16.67 1.32
B4- (3e50) 45.59 20.16 1.45
B5- (3e50) 46.58 20.59 1.24
60 B1- (3e60) 45.02 20.78 18.99 ± 0.97 5.13 123 0.67 1.19
B2- (3e60) 43.00 19.01 1.29
B3- (3e60) 39.45 17.44 1.32
B4- (3e60) 44.66 19.74 1.22
B5- (3e60) 42.64 18.85 1.47
6 40 B1- (6e40) 26.14 11.56 12.49 ± 1.88 15.0 81 1.48 1.49
B2- (6e40) 35.56 15.72 1.41
B3- (6e40) 26.42 11.68 1.60
B4- (6e40) 24.98 11.04 1.49
B5- (6e40) 28.13 12.43 1.46
50 B1- (6e50) 35.13 15.53 18.01 ± 1.60 8.9 116 1.88 1.56
B2- (6e50) 42.74 18.90 1.51
B3- (6e50) 43.35 19.16 1.35
B4- (6e50) 38.99 17.24 1.66
B5- (6e50) 43.51 19.24 1.38
60 B1- (6e60) 49.34 21.81 18.39 ± 2.97 16.2 119 1.12 1.10
B2- (6e60) 39.00 17.24 1.07
B3- (6e60) 48.22 21.32 1.06
B4- (6e60) 36.30 16.05 1.14
B5- (6e60) 35.11 15.52 1.08

Note: In all cases, the mode of failure was pullout.

t max ¼ Maximum bond stress at peak load (MPa).
Smax and Saver. ¼ Slip corresponding to maximum stress and average slip, respectively.
a
Bn- (t-T): B denotes BFRP bar; 1 to 5 for the specimen number in each set; and conditioning exposure time (t) in months and environmental temperature (T) in degrees
Celsius.

3.6. OM and SEM analysis
Fig. 9 shows optical-microscopy (OM) photographs through the
cross section in the longitudinal and in-plane directions of some
representative BFRP samples cut in the free length (sample 1) and
debonded length (sample 2) after 3 months of conditioning at
40
C. Fig. 9 shows that the BFRP deformations in both samples
were similar with no noticeable major damage taking place after
pullout testing. Moreover, concrete residue was present between
the deformations of the BFRP bars (Fig. 9 b), which may confirm
that the final failure was controlled by local concrete fracturing
rather than fracturing at the surface of the BFRP bars.
Furthermore, Fig. 10 presents selected representative scanning
electronic microscopy (SEM) micrographs of the cross section of the
free and debonded lengths of the bars after pullout testing. As the
figure shows, the SEM micrographs of the unconditioned bars taken
along the free length showed some typical pores and voids in the
coating layer on the bars' external surface, which are mainly due to
the manufacturing process. No pores or defects such as debonding
at the fiberematrix interface were observed in the bar core, indi-
cating that the core was not affected by the elevated thermal
exposure. The SEM micrographs of the conditioned specimens
along the debonded length exhibit radial and circumferential
damage in the resin matrix, primarily induced close to or around
the existing defects in the coating surface or as a result of peeling
off during pullout testing. There was no fiber degradation or
damage to the core of the bars. Fig. 10 reveals concrete residue
attached to bar surfaces. Examination of the bareconcrete
 M. Hassan et al. / Composites Part B 106 (2016) 262e272 269

Fig. 7. Effect of temperature on bond strength: a) average bond strength and b) bond-strength retention.

Fig. 8. Effect of exposure on the bond strength: a) average bond strength and b) bond-strength retention.

Fig. 9. OM photographs of the cross-section of representative BFRP samples after pullout testing after 3 months of conditioning at 40
C: a) longitudinal section at the free length; b)
longitudinal and in-plane sections along debonded length.

interfaces reveals microcracks at the interfaces, which were likely
induced by the differing transverse CTEs of the BFRP bars and
concrete. The bond degradation might also come from reinforce-
ment degradation, especially along the outer surface, resulting
from alkaline attack when the specimens were submerged in the
solution, as shown in Fig. 10, causing bond degradation. Fig. 10 b
shows an abraded polymer layer along the debonded surface of the
rebar after pullout (conditioning at 40
C).
4. Prediction of long-term bond-strength retention
Predications of the long-term bond-strength retention of the
BFRP bars were performed according to the method in fib Bulletin
40 [19]. The method includes a safety factor for bond strength that
takes into account bond deterioration with time. The bond strength
should be reduced by henv,b determined according to the equation
below.
270 M. Hassan et al. / Composites Part B 106 (2016) 262e272

Fig. 10. Selected SEM micrographs at different magnifications of representative BFRP bars along free and debonded lengths after pullout testing.

According to fib Bulletin 40 [19], for instance, nmo ¼ 1 and

henv;b ¼ 1=½ð100  R10 Þ=100n
n ¼ nmo þ nT þ nSL
where nmo, nT, and nSL are the influence terms for moisture con-
dition, temperature, and desired service life, respectively. R10 is the
standard reduction in bond strength in percent per decade (loga-
rithmic decade) due to environmental effects, which can be extrap-
olated from each individual degradation line (see Fig. 11). The values
of the environmental influence parameter R10 of the BFRP specimens
conditioned at 40
C, 50
C, and 60
C were 12.3%, 5%, and 4%. From
the curve fitting, the bond-strength retentions after 50 years were
45.4%, 92.0%, and 98.9% at 40
C, 50
C, and 60
C, respectively.
(2) nSL ¼ 2.7 at a service life of 50 years, assuming a moisture-saturated
condition. As adopted by Serbescu [1], nT is equal to 1.5, 2, and 2.5 at
(3) 40
C, 50
C, and 60
C, respectively. The value R10 for all the en-
vironments tested can be determined by using the average slope of
the individual degradation lines, assuming that the degradation
rate is similar regardless of environment [1]. Thus, R10 is equal to
7.10%. The estimated bond-strength retention (1/henv,b) is equal
68%, 66%, and 63% at 40
C, 50
C, and 60
C, respectively. Noticeable
differences between the two methods were observed for each
environmental conditioning. The differences might be attributed to
increased concrete strength resulting from immersion. This is not
considered in the equation, nor are the effects of moisture diffusion
on the degradation mechanism. Table 3 reports the (1/henv,b)
 M. Hassan et al. / Composites Part B 106 (2016) 262e272 271

2. The average bond strengths of the specimens conditioned for

1.5, 3, and 6 months of exposure at 40
C were 14.8, 13.3, and
12.5 MPa. At 50
C, the average bond strengths were 19.3, 18.4,
and 18.0 MPa and, at 60
C, they were 19.4, 19.0, and 18.4 MPa,
respectively. The bond strength of the unconditioned speci-
mens, however, was 15.5 MPa. The bond strength after 6 months
of exposure met the minimum bond-strength requirements for
use as non-prestressed reinforcement for concrete structures in
accordance with ACI and CSA specifications.
3. Generally, the bond strength increased as did the temperature
when the specimens were exposed to accelerated conditioning
environments as a result of an increase in concrete strength
during immersion. After 1.5 months of exposure, the bond
strengths increased by 25% and 26% in the specimens condi-
tioned at 50
C and 60
C, respectively, whereas no noticeable
changes occurred in the specimens conditioned at 40
C (a mi-
nor decrease of 4.3%).
4. The bond strength of the conditioned specimens exposed to the
alkaline solution for 1.5 months deteriorated during immersion
in comparison to their counterparts. The highest bond-strength
reductions occurred in the conditioned specimens after 6
Fig. 11. Experimental bond-strength-retention curves of BFRP bars in the test months of exposure at 40
C (a 16% loss), followed by specimens
environments. conditioned at 50
C (7% loss) and 60
C (5% loss)
5. The SEM micrographs of the BFRP bars along the debonded
length after pullout reveal radial and circumferential cracks in
Table 3 the resin matrix, some microcracks at the BFRP bareconcrete
Bond-strength-retention predications after service life of 50 years based on the
method in fib Bulletin 40.
interface, and chemical attack essentially in the coating surface,
yet there was no fiber degradation nor damage to the core of the
Material Moisture nmo MAT (
C) nT n henv,b 1/henv,b
bars.
condition
6. The bond-strength retention predications after 50 years of ser-
BFRP 12 mm Dry 1 <5 0.5 1.2 1.09 92% vice life in dry, moist, and moisture-saturated environments
in diameter 5e15 0 1.7 1.13 88%
with mean annual temperatures between 5
C and 35
C ranged
15e25 0.5 2.2 1.18 85%
25e35 1 2.7 1.22 82% from 71% to 92%.
Moist 0 <5 0.5 2.2 1.18 85%
5e15 0 2.7 1.22 82% Since the conclusions in this research study were developed
15e25 0.5 3.2 1.26 79%
based on investigating particular BFRP bars, additional research
25e35 1 3.7 1.31 76%
Moisture 1 <5 0.5 3.2 1.26 79%
targeting bond-durability issues is needed to investigate the effect
saturated 5e15 0 3.7 1.31 76% of different concrete strengths, different accelerated moist condi-
15e25 0.5 4.2 1.36 73% tions, and longer exposure times.
25e35 1 4.7 1.41 71%

Acknowledgments

predications at different moisture-saturated conditions and mean

annual temperatures (MATs) after 50 years of service life. The bond-
strength retentions after 50 years of service life in dry, moist, and
moisture-saturated environments and MATs varied from 71% to
92%. In order to validate the model's reliability, further work is
needed with different concrete strengths, different accelerated
moist environments, and longer exposure times.
5. Conclusions
The authors would like to acknowledge the support of the
Natural Sciences and Engineering Research Council of Canada, the
Canada Research Chair in Advanced Composite Materials for Civil
Structures, the Fonds de Recherche du Que bec - Nature et Tech-
nologies, and the Florida Department of Transportation. The au-
thors would like also to thank the technical staff of the structures
laboratory in the Department of Civil Engineering at the University
of Sherbrooke.

References
The main objective of this study was to investigate the bond
durability of BFRP bars embedded in concrete when exposed to
harsh environments. In order to achieve this objective, a bond-
durability study was performed on BFRP-reinforced pullout speci-
mens, which were fully immersed in an alkaline solution and
subjected to elevated temperatures. Based on the test results and
discussions presented in this study, the following conclusions can
be drawn:
1. All of the BFRP-reinforced specimens failed in a typical pullout
mode of failure by exhibiting slip through the free end. The final
mode of failure was dominated by shearing of the concrete
located at the BFRP bareconcrete interface.
[1] Serbescu A, Guadagnini M, Pilakoutas K. Mechanical characterization of basalt
FRP rebars and long-term strength predictive model. J Compos Constr
2014;19(2):04014037. http://dx.doi.org/10.1061/(ASCE)CC.1943-
5614.0000497.
[2] Sim K, Park C, Moon DY. Characteristics of basalt fiber as a strengthening
material for concrete structures. Compos Part B 2005;36(6e7):504e12.
[3] Palmieri A, Matthys S, Tierens M. Basalt fibers: mechanical properties and
applications for concrete structures. Concrete solutions: proceedings of the
international conference on Concrete Solutions. CRC Press/Balkema; 2009.
p. 165e9.
[4] Wu J, Li H, Xian G. Influence of elevated temperature on the mechanical and
thermal performance of BFRP rebar. CICE 2010. In: 5th int. Conf. FRP com-
posites in Civil engineering. Beijing: Tsinghua University Press; 2010.
p. 69e72.
[5] Lu Z, Xian G, Li H. Effects of exposure to elevated temperatures and
272 M. Hassan et al. / Composites Part B 106 (2016) 262e272
subsequent immersion in water or alkaline solution on the mechanical
properties of pultruded BFRP plates. Compos Part B 2015;77:421e30.
[6] Wang X, Wu G, Wu Z. Tensile property of prestressing basalt FRP and hybrid
FRP tendons under salt solution. In: Proc. 6th int. Conf. On FRP composites in
Civil engineering. Kingston, ON, Canada: International Institute for FRP in
Construction; 2012. 8p.
[7] ElSafty A, Benmokrane B, Rizkalla S, Mohamed H, Hassan M. Degradation
assessment of internal continuous fiber reinforcement in concrete environ-
ment. Materials Research Rep. Contract No. BDK82#977-05. Jacksonville. FL:
College of Computing. Engineering and Construction. Univ. of North Florida;
2014. p. 280.
[8] Benmokrane B, Elgabbas F, Ahmed E, Cousin P. Characterization and
comparative durability study of glass/vinylester, basalt/vinylester, and basalt/
epoxy FRP bars. J Compos Const 2015. ASCE ISSN 1090e0268/04015008(12).
[9] Elgabbas F, Vincent P, Ahmed E, Benmokrane B. Experimental testing of
basalt-fiber reinforced polymer bars in concrete beams. Compos Part B
2016;91:205e18.
[10] ACI Committee 440. Guide for the design and construction of concrete rein-
forced with FRP bars (ACI 440.1R-15). Farmington Hills. MI: American Con-
crete Institute; 2015. 88p.
[11] Davalos JF, Chen Y, Ray I. Effect of FRP bar degradation on interface bond with
high strength concrete. Cem Concr Compos 2008;30(8):722e30.
[12] Robert M, Benmokrane B. Effect of aging on bond of GFRP bars embedded in
concrete. Cem Concr Compos 2010;32(6):461e7.
[13] Bakis CE, Boothby TE, Jia J. Bond durability of glass fiber-reinforced polymer
bars embedded in concrete beams. J Compos Constr 2007:269e78. http://
dx.doi.org/10.1061/(ASCE)1090-0268(2007)11:3(269).
[14] El Refai A, Abed F, Altalmas A. Bond durability of basalt fiber-reinforced
polymer bars embedded in concrete under direct pullout conditions.
J Compos Constr 2014:04014078. http://dx.doi.org/10.1061/(ASCE) CC.1943-
5614.0000544. ASCE.
[15] Altalmas A, El Refai A, Abed F. Bond degradation of basalt fiber-reinforced
polymer (BFRP) bars exposed to accelerated aging conditions. J Constr Build
Mater 2015;81(15):162e71.
[16] Dong, Z., Wua, G, Xua, B., Wang, X, Taerwe, L. Bond durability of BFRP bars
embedded in concrete under seawater conditions and the long-term bond
strength prediction. Mater Des J; http://dx.doi.org/10.1016/
j.matdes.2015.12.066, 92(15): 552e562.
[17] Hassan M, Mohamed H, Benmokrane B. Development and characterization of
basalt FRP bars for concrete structures. In: Structural faults & repair-European
bridge conference concrete, materials & conservation. 8th e 10th July. South
Kensinton. London. UK: Imperial College; 2014. 8p.
[18] Hassan M, Benmokrane B. Physico-mechanical and durability characteristics
of basalt-fibre-reinforced polymer (BFRP) bars. CONMAT 2015, In: The fifth
international conference on construction materials: performance, innovations
and structural implications; 2015. 19e21 August. Whistler. Canada 8p.
[19] Fib Bulletin 40. FRP reinforcement in concrete structures. In: International
federation for structural concrete, Lausanne, Switzerland; 2007.
[20] ACI Committee 440. Specification for carbon and glass fiber-reinforced poly-
mer bar materials for concrete reinforcement (ACI 440.6M-08). Farmington
Hills, MI: American Concrete Institute; 2008. p. 6.
[21] Canadian Standards Association (CSA). Specification for fibre reinforced
polymers (CAN/CSA S807e10 (R2015). Rexdale. Ontario. Canada. 2015. p. 27.
[22] Canadian Standards Association (CSA). Design and construction of building
structures with fibre reinforced polymers (CAN/CSA S806e12). Rexdale. ON.
Canada. 2012. p. 208.
[23] ACI Committee 440. Guide test methods for fiber-reinforced polymers (FRPs)
for reinforcing or strengthening concrete structures (ACI 440.3R-04). Farm-
ington Hills. MI. USA: American Concrete Institute; 2004. p. 40.
[24] ASTM standard D7913/D7913M. Standard test method for bond strength of
fiber-reinforced polymer matrix composite bars to concrete by pullout
testing. West Conshohocken, PA: Copyright © ASTM International; 2014.
http://dx.doi.org/10.1520/D7913_D7913M-14. 19428e2959. United States.
[25] Abbasi, A., and Hogg, P. Temperature and environmental effects on glass fibre
rebar: modulus, strength, and interfacial bond strength with concrete, Com-
pos Part B; 36(5): 394e404, doi:10.1016/j.compositesb.2005.01.006.
[26] Tastani SP, Pantazopoulou SJ. Experimental Evaluation of the direct tension-
pullout bond test. In: Bond in concretedfrom research to standards. Buda-
pest; 2002. p. 1e8.
[27] Zhou J, Chen X, Chen S. Durability and service life prediction of GFRP bars
embedded in concrete under acid environment. J Nucl Eng Des 2011;241(10):
4095e102.
[28] Belarbi A, Wang H. Bond durability of FRP bars embedded in fiber-reinforced
concrete. J Compos Constr 2012:371e80. http://dx.doi.org/10.1061/(ASCE)CC
.1943-5614.0000270.
[29] Chen Y, Davalos J, Ray I, Kim H. Accelerated aging tests for evaluations of
durability performance of FRP reinforcing bars for concrete structures. Com-
pos Struct 2007;78(1):101e11.