Chapter - 1: Bond Durability of FRP Bars Embedded in Fiber-Reinforced Concrete
Chapter - 1: Bond Durability of FRP Bars Embedded in Fiber-Reinforced Concrete
Chapter - 1: Bond Durability of FRP Bars Embedded in Fiber-Reinforced Concrete
CHAPTER 1
1. INTRODUCTION
Composite
materials
offer
many
advantages
when
compared
with
conventional steel reinforcement, such as high tensile strength, light weight, and
corrosion free. Their application in civil engineering has become more technically
attractive and economically viable. The last few decades have been marked by
degradation of numerous concrete structures due to the corrosion of steel
reinforcements that required costly repairs or replacements. To mitigate the corrosion
problem, several methods, such as epoxy coated rebars, synthetic membranes, or
cathodic protection, have been developed. To date, all of them have showed limited
success . In recent years, research has been carried out on fibre reinforced plastic
(FRP) bars as an alternative to steel reinforcement. These FRP rebars have already
shown a promising future to overcome the corrosion problem in many projects,
especially in bridge deck and parking garage design.
Short polypropylene fibres provide resistance to plastic and drying shrinkage,
and improve resistance to crack growth, impact loading, fatigue loading and freezethaw durability. It was proven to have notable benefits to structures, especially under
service conditions. The combination of FRP reinforcement and short polypropylene
fibres may eliminate problems related to corrosion of steel reinforcement while
providing requisite strength, stiffness and desired ductility, which are shortcomings of
the plain concrete and FRP reinforcement system.
Considerable research efforts have been conducted on the bond behavior of
glass fiber reinforced plastic (GFRP) rebar in plain concrete. Different types of FRP
rebars have quite different bond characteristics, which are strongly dependent on
mechanical and physical properties of external layer of FRP rods.
Meanwhile, research has indicated that FRP materials are not immune to
Department Of Civil Engineering
ii)
iii)
iv)
Regardless of the fiber used in the FRP reinforcing bar, it is the resin matrix that plays
the major role in transferring forces from the surrounding concrete into the reinforcing
bar, and this resin matix is potentially susceptible to degradation at the concrete
interface . Once this degradation occurs , bond capacity can be reduced or lost
completely and in addition the reinforcing fibers themselves can be exposed to attack
leading to loss of the longitudinal strength and stiffness of the reinforcement.
Department Of Civil Engineering
Reduced Permeability
CHAPTER 2
2. EXPERIMENTAL SET UP AND PROCEDURES
2.1. EXPERIMENTAL PROGRAM
In this study, a total of 36 bond specimens were fabricated to study the effect
of various environmental agents on the durability of the FRP/FRC system. To
simulate the seasonal weather changes, specimens were subjected to combined
environmental cycles, consisting of freeze-thaw cycles and high temperature cycles,
while submerged in the salt solution. Then, bond behaviours were compared with
unweathered specimens to investigate the durability of this new hybrid system.
2.2 MATERIALS
2.2.1 Fiber-Reinforced Polymer Rods
Three types of commonly used FRP rods were adopted in this experimental study:
namely #8 (25-mm) GFRP, #4 (13-mm) GFRP, and #4 (13-mm) CFRP, as shown in
figure 2.1. The surface of the GFRP rods is tightly wrapped with a helical fibre strand
to create indentations along the bar, and sand particles are added to the surface to
enhance its bonding strength. The surface of the CFRP is very smooth, as shown in
figure 2.1. The resin used was epoxy-modified vinyl ester. The mechanical properties
of FRP rods as are shown in table 2.1.
2.2.2 Polypropylene Fiber
Currently, many fiber types are commercially available, including steel, glass,
synthetic, and natural fibers. To fulfill the completely nonferrous concept,
polypropylene fibre was used in this study. Fibres are fibrillated and commercially
available in 57-mm length. Some properties of interest as reported by the
manufacturer are listed as follows: specific gravity = 0:91;Youngs modulus = 3:5
GPa. Polypropylene fibers can
Department Of Civil Engineering
# 8 GFRP
# 4 GRRP
# 4 CFRP
The environmental agents that have potential effects on the long term structural
behaviors of this FRP/FRC hybrid system are discussed as follows:
Table 2.1: Mechanical properties of Fiber-Reinforced Polymer Rods
Items
Tensile strength
(MPa)
Elastic modulus
(GPa)
Fiber type
Resin type
#4 Carbon
fiber-reinforced
polymer
2,069
#4 Glass
fiber-reinforced
polymer
690
#8 Glass
fiber-reinforced
polymer
551
124
41
41
Polyacrylonitrile
(PAN)-based
carbon fiber
Epoxy-modified
vinyl ester
E-glass fiber
E-glass fiber
Epoxy-modified
vinyl ester
Epoxy-modified
vinyl ester
Portland
cement
type I
(kgm3)
Water
(kgm3)
Fly ash
(Class
C)
(kgm3)
Sand
(kgm3)
Coarse
aggregate
(limestone)
(kgm3)
Volume fraction
of
polypropylene
fiber
Airentraining
agent
(mLm3)
Plain concrete
371
119
119
771
1,020
0%
296
Fiber-reinforced
concrete
368
124
65
676
1,080
0.5%
266
strength (MPa)
Air Content
Plain concrete
48
4.9%
37
6%
structure. However, as shown in table 2.4, the CTEs for fibers and concrete are
different. Furthermore, the resin materials used to bind the fibers have much larger
Department Of Civil Engineering
CTEs in comparison with concrete. Significant stresses at the interface of the two
Longitudinal direction
Transverse direction
material
(1061/K)
(1061/K)
Aramid fiber
-6.0 to -2.0
55 to 60
Carbon fiber
Glass fiber
-0.9 to 0.7
5 to 15
8 to 18
5 to 15
Resin
60 to 140
60 to 140
Steel
12
6 to 13
12
6 to 13
Concrete
for the application of FRP materials in this project, FRP bars were protected by a
concrete cover. The degradation caused by UV radiation was expected to be negligible
and was not investigated in this study.
2.3.4 De-icing Salt Solution
Chloride penetration caused by the de-icing salt is a major cause of corrosion in steelreinforced highway structures. It may also affect the strength of the FRP materials. As
discussed previously, damage caused by the freeze-thaw cycles will be aggravated by
the use of salt solution. In this study, a solution of sodium chloride, 5% by weight,
was selected to simulate the de-icing salt solution. Specimens were submerged into
the salt solution in two large tanks.
2.3.5 Humidity Effect
The FRP rods are not waterproof. Moisture can diffuse into resin, leading to changes
in mechanical characteristics and in physical appearance (increase of volume). As a
consequence, the overall performance of the FRP/FRC hybrid system may be altered.
Because specimens in this study were submerged in salt water, the humidity effect on
the FRP/FRC system was covered automatically.
10
at the first 5db length to minimize the bottom plates restraint effect on the FRP bar
and to eliminate any undesirable confinement that may affect the bond characteristics.
Details are shown in figure 2.5.
Chemical agents will attack the reinforcing materials and the bond between the
concrete and the reinforcing materials. In typical RC structures, cracks exist under
service conditions. The degradation effect is expected to be more pronounced at
locations at which cracks exist. Because the portion away from the cracks is not
directly in contact with solution, to better simulate the real situations, the portion at
the loaded end of the bond specimens was directly exposed to salt water, where as the
portion at the free end was coated with waterproof epoxy to protect it from direct
attack from
anchorages and change the bond behaviour when the bar was pulled out, all the epoxy
that stuck to the bar was removed after environmental conditioning. The notation
for specimens is as follows: the first character, V or D, indicates the unweathered
specimen or weathered specimen; the second character, P or F, indicates the plain
concrete or FRC; the third character (#4 versus #8) is the bar size in U.S.designation;
and the last character, C or G, indicates the reinforcement type, CFRP or GFRP. The
test matrix of the bond specimen is shown in table 2.5.
2.6 TEST SETUP AND PROCEDURES
The pullout tests were conducted in an MTS 880 machine. The test is run through
closed-loop displacement control using an external LVDT2 as a feedback device
(shown in figure 2.5).Because the FRP rods were weak in the transverse direction,
special anchorages were used to protect the reinforcement from crushing. The free end
of the FRP bar was embedded in a steel pipe using an expansive grout as an interface.
The pullout was then performed by pulling the steel pipe at one end, with the concrete
block encased in the steel reaction frame. The bars slip relative to the concrete was
computed from measurements of two LVDTs placed at both ends of the bar, as shown
in figure 2.5. To minimize the eccentricity effect, lead sheets were placed between the
concrete block and the reaction frame. The pullout tests were monotonic by increasing
Department Of Civil Engineering
11
12
13
Number of specimens
f c (MPa)
VP4C
48
VP4G
48
VP8G
VF4C
3
3
48
37
0
0.5
VF4G
37
0.5
VF8G
37
0.5
DP4C
48
DP4G
48
DP8G
DF4C
3
3
48
37
0
0.5
DF4G
37
0.5
DF8G
37
0.5
Vf (%)
14
CHAPTER 3
3. TEST RESULTS AND DISCUSSIONS
3.1 TEST RESULTS
In the following discussion, the environmental conditionings effect on the specimen
appearances and the bond behaviors for both plain concrete specimens and FRC
specimens are presented.
3.1.1 Appearance of Specimen after Environmental Conditioning
i) Plain Concrete Specimens
In addition to concrete scaling on the surface, most specimens also showed some
damage on the concrete, especially at the corner areas. One DP4C specimen and one
DP4G specimen were severely damaged, and large portions of concrete were broken
apart, as shown in figure 3.1.
ii) Fiber-Reinforced Concrete Specimens
Damages were limited to the surfaces of the specimens. With the scaling of concrete
at the surfaces, fibers could clearly be observed. However, all of the FRC specimens
remained integrated. In comparison with the plain concrete specimens, the FRC
specimens were more immune to the environmental attack.
3.1.2 Environmental Conditioning Effect on Bond Behaviors
The test results are shown in table 3.1. A typical bond slip is shown in figure 3.2 .The
typical bond-slip responses at the loaded end and the free end are shown in figures 3.3
and 3.4, respectively. In this study, the average bond strength was calculated as the
pullout force over the embedded area of the bar. The slip on the side of loading was
calculated as the value of LVDT2 minus the elastic deformation of the FRP bar
between the bond zone and the location of LVDT2.
i) Plain Concrete Specimens
Bond slip response : Unweathered specimens showed fairly consistent test results
Department Of Civil Engineering
15
(a)
(b)
Fig 3.1 : Difference in appearance of plain and fiber-reinforced concrete specimens
after environmental conditioning: (a)#4 glass fiber-reinforced polymer;(b)#4 carbon
fiber-reinforced polymer
with the same testing parameters .However , test results for specimens subjected
to environmental conditioning were volatile. Different levels of damage on the
specimens were observed visually. In general, specimens with concrete more severely
Department Of Civil Engineering
16
untolerated crack width. Bond stress corresponding to 0.2-mm slippage of loaded end
or 0.05-mm slippage of free end for steel-reinforced structures can be defined as the
critical bond stress. The criterion of 0.2-mm slippage at the loaded end was decided
on the basis of half of the crack width limitation. As mentioned previously, after being
subjected to environmental conditioning, surface scaling occurred on most of the
specimens. During the pullout bond tests, the loose concrete at the loaded end would
be compacted under compression. Thus, the measured loaded-end slip could be
exaggerated. However, slips measured at the free end would not be affected.
Therefore to keep it comparable with limits imposed on the steel reinforcement, bond
strength corresponding to 0.05-mm slippage at the free end was adopted as the
designing bond strength. As shown in figure 3.6, 25%, 43%, and 11% reductions were
observed for DP4C, DP4G, and DP8G specimens, respectively.
ii) Fiber-Reinforced Specimens
Bond-slip response: In general, test results of the FRC specimens showed good
consistency. The behavior of the specimens in the same testing group was similar.
Similar to the plain concrete specimens, all bond-slip curves were softened after being
subjected to the environmental cycles.
Failure modes: Similar to the plain concrete specimens, most FRC specimens had
the same failure modes as the unweathered specimens. However, the failure mode of
one of the three DF8G specimens was changed from reinforcement pullout to concrete
splitting.
Ultimate bond strength: Reductions of the bond strength were observed for most of
the FRC specimens, as they were for the plain concrete specimens. As shown in figure
3.4, 5% and 15% reductions were observed in the ultimate bond strength for DF4C
and DF4G specimens, respectively. However, little change was observed for DF8G
specimens.
Design bond strength: Similar to the plain concrete specimens, reductions of the
design bond strength defined previously were observed for most of the FRC
specimens. As shown in figure 3.5, 12% and 16% reductions were observed in the
Department Of Civil Engineering
18
Significant bond reductions were observed for specimens whose concrete were badly
19
Fig 3.3: Typical loaded- end bond- slip curve (#4 GFRP): (a) Plain concrete
specimen; (b) Fiber-reinforced concrete specimen
Fig 3.4: Typical free- end bond- slip curve (#4 GFRP): (a) Plain concrete specimen;
(b) Fiber-reinforced concrete specimen
20
design bond strength for DF4C and DF4G specimens, respectively. However, little
changes were observed for DF8G specimens.
3.2 DISCUSSIONS
In the following discussion, mechanisms responsible for the bond degradation are
presented. Bond durability affected by different testing parameters, including
specimen size, plain concrete or FRC, GFRP specimens or CFRP specimens, is also
discussed.
Department Of Civil Engineering
21
22
23
24
reinforcement, especially the outer surface, was damaged. Thus, the contact area
would be reduced. Research has shown that the CFRP bar has superior durability
characteristics compared with the GFRP bar. Thus, less damage was expected in the
case of the CFRP bar, and hence, the CFRP specimens showed better durability of
bond. This phenomenon further supported the finding that bond strength degradation
is tightly related to the material degradation.
25
CHAPTER 4
4. CONCLUSIONS
Long-term bond performance between FRP bars and FRC was investigated
and compared with that of FRP bars in plain concrete. The accelerated aging test was
accomplished by submerging specimens in salt solutions and subjecting them to 10
combined environmental cycles, each of which consisted of 20 freeze-thaw cycles and
20 high temperature cycles. The following conclusions can be drawn from this study:
Three mechanisms cause bond degradation. The first is the expansion of solutions in
the microvoids at the interface. The second is the different CTE between
reinforcement and concrete. The first two mechanisms function together and degrade
the bond primarily by creating microcracks in concrete at the interface. The third
reason is the damage of the reinforcement, especially at its surface.
With the addition of polypropylene fibers, the bond durability significantly
improved owing to the restriction of the crack development at the interface. The loss
of ultimate bond strength of FRP bars in plain concrete owing to aging effects was
found to be 28% on average, whereas only a 6% reduction was observed in the FRC
specimens. Similarly, design bond strength exhibited a 26% average reduction in the
plain concrete specimens, whereas only a 10% reduction was observed in the FRC
specimens.
The larger specimens with thicker concrete cover and relatively smaller direct
exposed area to the solution of sodium chloride showed better bond durability.
Degradation of bond is tightly correlated with the degradation of FRP bar. As
observed in this study, CFRP specimens had superior long-term bond durability
compared with GFRP specimens. This is attributed to the more durable characteristics
of the CFRP bar.
26
REFERENCES
Journals
27