An Investigation on Fire Performance of FRP-

strengthened R/C Beams

BREA WILLIAMS1, LUKE BISBY2, VENKATESH KODUR3, JOSEPH SU3, and
MARK GREEN2
1
Technical Team Member
Halsall Associates Ltd.
3001-210 Gladstone Avenue
Ottawa, Canada
2
Faculty Member
Dept. of Civil Engineering
Queen’s University
Kingston, Canada
3
Senior Research Officers
Fire Risk Management Group
Institute for Research in Construction
National Research Council of Canada
Ottawa, Canada
ABSTRACT
As the use of Fibre-Reinforced Polymer (FRP) materials continues to expand into the
structural repair market, concerns over the performance of these materials in fire
conditions are now at the forefront of research. While externally bonded FRP sheets have
been shown to successfully enhance the flexural and shear capacity of bridges and other
structures, their application in interior spaces, where fire is a significant concern, remains
questionable in light of these materials’ comparatively poor resistance to elevated
temperatures. This paper presents the results of an ongoing study to document the
performance of FRP-strengthened reinforced concrete beam-slabs exposed to fire. With
such experimental results, fireproofing materials have the potential to earn standard
performance ratings, which are essential to the design engineer and necessary for the
continued increase of FRP applications worldwide.
A brief synopsis of existing literature related to FRP behaviour at high temperature is
provided, in addition to a review of the current fire endurance criteria for structural
members. Two large-scale concrete beam-slab assemblies were strengthened with FRP
sheets and protected with a two-part patented fire insulation system. The results of fire
tests performed on these specimens are presented herein, with emphasis placed on the
temperatures measured in the specimens during fire exposure. The data obtained from the
tests served to validate a numerical heat transfer model, which predicts the temperature
within a strengthened and insulated reinforced concrete beam-slab assembly. Test results
and model data indicate that appropriately designed and insulated FRP-strengthened
beam-slab assemblies can achieve fire endurances of four hours or more.

KEYWORDS: fibre-reinforced polymer (FRP), structural fire endurance, reinforced

concrete, repair, strengthening

FIRE SAFETY SCIENCE–PROCEEDINGS OF THE EIGHTH INTERNATIONAL SYMPOSIUM, pp. 247-258

COPYRIGHT © 2005 INTERNATIONAL ASSOCIATION FOR FIRE SAFETY SCIENCE 247
NOMENCLATURE
c specific heat (J/kg-K) φ moisture content of element
k thermal conductivity (W/m-K) Subscripts
∆l length of element (mm) c concrete
∆t time increment (hours) f fire
Tg glass transition temperature (ºC) m refers to adjacent node “m”
Greek r refers to adjacent node “r”
εc emissivity of concrete = 0.9 w water
εf emissivity of fire = 1.0 Superscripts
ρ density j
previous time step
σ Stefan-Boltzmann constant j+1
current time step

INTRODUCTION
The rapidly deteriorating state of the world’s infrastructure has promoted recent
innovations in materials science. In an effort to rehabilitate deteriorating structures, repair
under-strength members and offer durable new construction materials, FRP materials
have been adapted from their more conventional applications in the aerospace industry,
and are now being applied in civil engineering. FRP products, which consist of fibres
(carbon, glass, aramid) in a polymer matrix (epoxy, vinyl ester), offer lightweight, high
strength, corrosion resistant alternatives to conventional building materials. FRP
materials are also highly versatile; available as bars, tendons, sheets, tubes and many
other custom forms. Over the past 15 years, research initiatives around the world have
demonstrated the viability of FRP sheets for use in repair and rehabilitation schemes for
concrete structures. Externally-bonded FRP sheets have been shown to successfully
enhance the strength and ductility of reinforced concrete columns [1,2,3,4], and to
achieve increased flexural and/or shear strength in reinforced concrete beams [5,6]. In
beam applications, FRP sheets are bonded to the base of the beam web or to the face of
the web to gain strength in bending or shear, respectively

FRP at Elevated Temperature

While structural steel members require protection to preserve their strength in a fire
event, FRP-strengthened concrete beams require protection to maintain their strength, to
prevent combustion of the material, and, to a lesser extent, to preserve their bond with the
substrate. The introduction of FRP into a new field of application brings with it a host of
technical obstacles; one concern at the forefront of FRP research is related to the
performance of FRP-strengthened members in fire. The organic polymer matrices of FRP
make them susceptible to combustion when exposed to fire. The critical temperature is
the glass transition temperature (Tg); defined as the threshold at which an FRP’s
properties begin to change significantly. When Tg is exceeded, the polymer matrix
becomes soft and rubbery [7], with reduced strength [8]. Many commonly used thermoset
matrices exhibit a Tg in the range of 65-82ºC [9]. At even higher temperatures polymer
matrices may ignite, supporting flame spread and toxic smoke evolution [10].
The structural integrity of FRP materials after exposure to fire is not well known.
However, a study [11] was performed recently to define the temperature at which only
50% of the original room temperature strength of an FRP material remains. Based on
57 tension tests conducted at various increasing temperatures, it was determined that

248
carbon and glass FRPs lost 50% of their strength at 250ºC and 325ºC, respectively.
Stiffness appeared to suffer negligible losses up to about 400º, above which point it
decreased rapidly. A recent literature survey conducted by Bisby [12], and updated by
Williams [13], compiled temperature-dependent strength and stiffness data for FRP from
a number of studies and proposed predictive equations, based on a least-squares
regression analysis, to describe the observed trends. The equations were assumed to be
sigmoidal in nature. Figure 1 shows the individual data points used in the regression
analysis along with the predictive sigmoid curve for FRP strength. While the curve
reveals the general trends in the data, it is a rough approximation since many different
specific types of FRP materials are represented in the data.
120
PAN Carbon / Epoxy 1 (1)
PAN Carbon / Epoxy 2 (1)
PAN Carbon / Epoxy 3 (1)
100 Pitch Carbon / Epoxy 1 (1)
Pitch Carbon / Epoxy 2 (1)
Braided Carbon / Epoxy (4)
80 Stranded Carbon / Epoxy (4)
% Retained

Braided Carbon / Epoxy (5)

Spiral Carbon /Epoxy (5)
60 Carbon / Epoxy (6)
Carbon / Epoxy (7)
Carbon / Epoxy-Phenolic (7)
Carbon / Epoxy (8)
40
Pitch Carbon / Cement (1)
Carbon / Inorganic (5)
Carbon / Glass / Vinyl Ester (2)
20 2
R =0.524 Carbon / Glass / Vinyl Ester (3)
Carbon / Polyimide (7)
Idealized Carbon / Epoxy
0 (1) Kumahara et al., 1993 (5) Tanano et al., 1995
-50 50 150 250 350 450 550 (2) Rahman et al., 1993 (6) Sumida et al., 2001
(3) Fujisaka et al., 1993 (7) Dimitrienko, 1999
Temperature (deg. C) (4) Tanano et al., 1997 (8) Wang et al., 2003

Fig. 1. Variation in strength with temperature for various

carbon Fibre Reinforced Polymers [12,13].

Standard Fire Testing

Due to the random nature of fire, standard fire conditions have been prescribed in fire
testing codes to represent typical severe fires in the lab setting. In a fire test in North
America, a structural member must be subjected to standard temperature-time conditions
as described by ASTM E119 [14], which reaches temperatures in excess of 1000ºC after
two hours. The main parameter of interest measured in a structural fire test is a
specimen’s fire endurance, which is defined as the length of time that a structure or
member withstands exposure to the fire without losing its load-bearing or fire separating
capacities. For the flexural specimens evaluated in this paper, ASTM E119 defines fire
endurance as the time during which the following criteria are satisfied:
1. The member is capable of withstanding its applied load;
2. The reinforcing steel in the concrete maintains a temperature less than 593ºC; and
3. The average unexposed surface temperature does not rise more than 140ºC and no
individual point on the unexposed face rises more than 180ºC.

249
Related Research
While a significant amount of testing has been conducted on FRP materials in isolation at
high temperatures, these results cannot be directly extrapolated to predict the fire
performance of concrete strengthened with FRP sheets. Moisture content, temperature
and stress in the concrete member affect the fire performance and should be taken into
account. To date, two research groups [15,16] have performed tests similar to those
presented in this paper. Deuring [15] and Blontrock et al. [16] exposed reinforced
concrete beams with a number of strengthening and insulation configurations to fire. In
the first group of tests [15], it was found that a specimen without strengthening withstood
fire for 118 minutes. An FRP-strengthened concrete beam without protection displayed
81 minutes of fire resistance, whereas the same beam with a 40 mm-thick protection of
calcium silicate board applied to the exterior of the FRP improved its fire endurance by
1.8 times over the unprotected specimen. In the second set of tests [16], the effects of
varying protection thickness, location, method of bonding, and length were studied. A
beam with calcium silicate board insulation anchored by adhesive lost its bond after only
7 minutes of fire exposure, while mechanically anchored insulation maintained its
protective capacity for 45 minutes. It was determined that U-shaped fire protection
provided more effective insulation for the FRP-strengthened beams. Finally, insulation
applied only within the anchorage zone of the FRP sheets, preserved the bond sufficiently
to allow the CFRP sheet to maintain its contribution as tensile reinforcement well into the
test (38 minutes), in a manner similar to the fully protected beams.

EXPERIMENTAL PROGRAM
Researchers at Queen’s University at Kingston, Canada, the National Research Council
of Canada (NRC), and Fyfe Co. LLC collaborated in an attempt to quantify the thermal
and structural behaviour of FRP-strengthened reinforced concrete beams subjected to fire
conditions. This project involved the design and fabrication of two large-scale beam-slab
assemblies, both of which were strengthened with FRP, protected with a patented
insulation system, loaded to service load levels, and exposed to the ASTM standard fire.

Specimens, Strengthening and Protection

Figure 2 shows a cross section of the beam-slab assembly. Two beams were fabricated,
spanning 3.8 m, with cross sectional dimensions selected to satisfy ASTM requirements
[14], and to be representative of field conditions. The beams were reinforced to resist
self-weight and a 2.4 kPa live load. The same concrete mix design was used for both
beams, and included 16 mm limestone aggregate and a 28-day design compressive
strength of 41 MPa. The measured relative humidity of the specimens at the test date was
73%.
The FRP strengthening scheme was devised assuming that an increase in live load to
4.8 kPa was required. Both beams were strengthened with a 100 mm wide strip of
Fyfe Co. Tyfo® SCH composite strengthening system, applied to the soffit of the web
over the entire beam span. The composite consisted of a carbon reinforcing fabric and a
two-part epoxy saturant/adhesive, applied using a hand lay-up procedure. A 1.3 mm-thick
SCH laminate possesses a design tensile strength of 460 MPa in the direction of the
fibres, and an ultimate elongation of 2.2% at failure. To enhance the bond of the flexural
sheet in fire, a U-wrap of Fyfe Co. Tyfo® SEH glass/epoxy FRP sheet was applied to the

250
anchorage zones (600 mm in length) of the beam. This strengthening resulted in an
approximate 15% increase in the flexural capacity of the beams.

1220

Carbon FRP sheet

150
Insulation (VG)

Sealant (EI-R)

250
125
All reinforcement
20mm dia. 10mm dia. unless
steel rebar otherwise specified
100 All dimensions in mm
40mm clear
cover 300

Fig. 2. Cross section schematic of beam-slab assembly.

Fire protection for both beams consisted of a two-part system consisting of Fyfe Co.
Tyfo® VG insulation and Fyfe Co. Tyfo® EI-R coating. VG is a lightweight, fire
resistant cementitious plaster that was spray-applied to the underside of the beams.
Materials similar to VG have been used in the past to fireproof structural steel members.
As shown in Fig. 2, the VG was applied to three sides of the beams’ webs, and extended
to the underside of the slabs, along the entire beams’ length. EI-R is an impermeable
surface-hardening layer that was spray-applied over the installed VG layer. Beam 1 was
protected with 25 mm of VG insulation, while Beam 2 had a VG thickness of 38 mm.
Both beams had an EI-R thickness of approximately 0.13 mm. Table 1 summarizes the
properties of the VG. The EI-R coating has been ignored in the thermal analysis since it
has been observed to burn off rapidly in fire, and its overall effects on heat transfer are
thus considered negligible.
Table 1. VG insulation properties.
Property Value ASTM Test
Density, kg/m3 240-272 -
Compressive strength, kPa 112.0 E 761
Bond strength, kPa 18.6 E 736
Combustibility Non-combustible E 136
Surface flame spread 0.0 E 84
Effect of impact Passed E 760
Effect of deflection Passed E 759
Thermal conductivitya, W/m-ºC 0.00815 -
Specific heata, J/kg-K 1047 -
a
Approximate room temperature data based on previous research [17]
Test Setup
The two beams were tested simultaneously in the full-scale floor furnace at NRC. They
were secured in a steel ring frame that was subsequently placed on top of the furnace
chamber; exposing the web and lower slab face of the beams to fire from below while the
top of the slabs were exposed to ambient conditions. The ends of the beams were axially
restrained, though this fixity did not translate into full moment fixity at the supports. It

251
was anticipated that the axial restraint provided by the steel ring beam would enhance the
fire endurance of the beams during fire testing, as is suggested in the fire testing literature
[18]. The required superimposed load was determined according to standard methods
[19], and applied to the unexposed surface using a series of 12 hydraulic jacks, to a load
level representing 56% of the ultimate capacity of the concrete beams alone, and 48% of
the ultimate capacity of the strengthened beams. Finally, each beam-slab assembly was
instrumented with 36 thermocouples located within the concrete, at the unexposed
surface, and within the FRP and insulation layers, to monitor the transfer of heat during
the test.

EXPERIMENTAL RESULTS
Through small viewports in the furnace walls, it was observed that limited surface
flaming (combustion) of the EI-R layer occurred in the first four minutes of the test.
Localized cracking of the VG occurred around 60 minutes of fire exposure, likely related
to accidental unloading and reloading of the beams resulting from a minor problem with
the hydraulic system. At four hours of fire exposure, the applied load on the beams was
gradually increased to twice the original load level in an effort to induce structural
failure. As failure did not appear imminent while applying the maximum possible load,
the test was stopped at 267 minutes.

Temperatures
The temperatures measured in the insulation and FRP strengthening layers of Beams 1
and 2 are shown in Fig. 3.
1200

1000
Temperature [ºC]

Avg. Furnace Temp.

800 EI-R/VG
VG/FRP
600 FRP/Concrete
ASTM E119

400
Beam 1
200 Beam 2

0
0 60 120 180 240
Time [minutes]

Fig. 3. Temperatures in Beams 1 and 2.

The temperature at each interface was measured at three points along the span of each
beam, and has been averaged for comparison purposes. The furnace temperature strayed
slightly from the standard temperature-time curve due to low (sub-zero) ambient
temperatures and the large mass of the assemblies. Initially, the temperature at the EI-
R/VG interface rose rapidly due to burning-off of the EI-R layer, but then followed the
trend of the furnace temperature for the remainder of the test. The temperature at this
interface was nearly identical for the two beams, which is to be expected since both were
protected with the same thickness of EI-R coating. At the VG/FRP interface, the
temperatures rose at a slower rate, as expected, and again essentially followed the

252
temperature in the furnace. Unexpected rises in temperature were observed at this
location and were likely due to localized cracking of the VG, which would allow more
rapid heating of the underlying FRP layer. Due to a greater thickness of VG insulation,
Beam 2 recorded slightly lower temperatures at this interface than Beam 1. The
temperature at the FRP/concrete interface (the FRP/concrete bondline) is presumed to be
an important indicator of the ability to maintain bond between the FRP strengthening
system and concrete substrate during fire. Figure 3 shows that temperature at this location
increased at a slower rate than at any other location, and evaporation of moisture in the
VG layer, as evidenced by the temperature plateau around 100ºC, assisted in maintaining
a lower temperature for a longer period of time. After four hours of fire exposure, the
average FRP bondline temperature was 361ºC and 220ºC in Beams 1 and 2, respectively.

Fire Endurance
Both beams resisted their applied load for over four hours of fire exposure. However,
equally important in defining fire endurance is the application of the thermal criteria of
ASTM E119 [14]. The maximum recorded steel reinforcement temperatures of 561ºC
(Beam 1) and 540ºC (Beam 2) were less than the maximum allowable temperature of
593ºC for the full test duration. Note that these maximum temperatures were measured in
the slab portion of the section, while the steel rebar in the web portion reached only a
maximum temperature of approximately 250ºC and 200ºC in Beams 1 and 2,
respectively. In addition, the individual point and average unexposed temperatures at the
unexposed face of the beams were less than the maximum allowable temperature
increases of 180ºC and 140ºC, respectively. Since load and temperature criteria were
satisfied for the full test duration, both beams earned four hour fire ratings. At this time,
fire testing provisions do not exist to limit the temperature of an FRP strengthening
system during exposure to fire. However, if it were required to maintain the FRP layer at
or below the Tg of the FRP’s epoxy saturant/adhesive (93ºC for the system used herein),
Beam 1 (25 mm VG) and Beam 2 (38 mm VG) would exceed this criterion within 16 to
36 minutes, and 55 to 57 minutes, respectively. The range in time occurs because the
temperature at the FRP was measured at three different points along each beam, and each
recorded slightly different temperatures. Structurally speaking, a Tg limiting criterion is
over-conservative in that it assumes failure of an FRP-strengthened structural member
occurs at the time the Tg of the FRP is exceeded. Based on the performance of the
specimens herein, this is clearly an unrealistic criterion for failure. The Tg was exceeded
within one hour of fire exposure for both beams, yet both assemblies continued to resist
their full strengthened service loads for more than four hours.

NUMERICAL MODEL

Theory
Validated numerical methods of analysis are extremely useful tools in fire risk
management. Many software programs are available worldwide that are capable of
predicting temperature distributions in structural members consisting of steel or
reinforced concrete. One of the more prominent currently available models is SAFIR, a
special-purpose finite element program developed to perform thermal and structural
analysis of building elements and frames exposed to fire [20]. The program accounts for
the presence of insulating materials, moisture content and temperature-dependent thermal
properties, and has been validated and used in several case studies to date. Unfortunately

253
this model was not available for modification for this program’s purposes. Thus, a model
was created to allow modeling of the temperature distribution within an FRP-
strengthened and insulated concrete beam-slab assembly. The two-dimensional heat
transfer model developed during this research program is capable of predicting the
temperature distribution within rectangular and tee-shaped reinforced concrete beams
during exposure to a standard fire. The model employs the finite difference method, with
the bulk of its theory based on equations outlined previously by Lie [21]. The model is
capable of analyzing an unstrengthened reinforced concrete beam, as well as simulating
the thermal effects of FRP sheets and/or insulation applied to the base and sides of such
beams. The model is also capable of simulating the conditions of the test specimens
described in this paper; that is, tee-sections with an FRP layer on the soffit and insulation
applied to the soffit and sides of the beam, including a portion of the underside of the slab
overhang. Enhancing the model’s flexibility is its ability to account for delamination of
FRP and insulation; this was useful in comparing the model’s ability to predict data from
previous fire tests where FRP and/or insulation delamination was observed during fire.
The analysis begins by discretizing a beam’s cross section into nodal points, as shown in
Fig. 4, for half of the cross section of a beam-slab assembly strengthened with FRP and
protected with a layer of insulation.

CL

Concrete

FRP
Insulation

Fig. 4. Discretization of beam for heat transfer modelling.

The temperature of the standard fire at the current time step is calculated using an
equation provided by ASTM E119 [14]. However, the model can easily be modified to
replicate the effects of any type of fire, so long as the temperature-time data is known. A
thermal equilibrium equation is applied to each node in the section, and the current nodal
temperature is determined based on the temperature and thermal properties of the
surrounding nodes during the previous time step. For example, the thermal equilibrium
equation for any element, m, along the exterior of the concrete beam at time step j+1 is:

Tmj +1 =Tmj +
∆t ( ⋅⎢
)(
⎡ kqj + kmj
) ( )(
k j +k j
⋅ Tqj −Tmj + r m ⋅ Trj −Tmj)
[ ( )]
(ρc ⋅ cc )mj + ρw ⋅ cw ⋅φmj ⋅ ∆l ⎣
⎢ 2
2 2
2
( ) ( ⎧
) ( 4
) 4⎫
( )( )
+ σ ⋅ ε f ⋅ εc ⋅ 2 ⋅ ∆l ⋅ ⎨ Tfj + 273 − Tmj + 273 ⎬ + 9035.1⋅ 2 ⋅ ∆l Tmj −Tambient ⎥
⎩ ⎭
1.25⎤

⎦
(1)

254
The second last term in the above equation describes radiative heat transfer with the fire,
and is only applied to those elements which are exposed directly to fire. The last term of
the equation describes the change in temperature due to convection heat transfer, and is
only used for those elements along unexposed surfaces. The effects of moisture
evaporation are included in the analysis by assuming that when the temperature in a node
reaches 100ºC, all of the heat transferred to that node is used to evaporate water. During
evaporation, the temperature in the node is assumed to be constant at 100ºC, and the
temperature rises again only when the moisture has completely evaporated. The effects of
moisture migration away from the heat source, which are thought to be significant in
accurately predicting temperatures near 100°C, are not currently included. In the
derivation of the heat transfer equations, the analysis is performed on a unit length of the
beam cross section. This implies that the entire length of the beam experiences identical
fire exposure conditions; end effects are ignored.

Validation
Figure 5 provides a comparison between recorded temperature data from the fire tests
discussed previously and model predictions. Figure 5a shows that the model
underestimated the temperature at the FRP/concrete location of Beam 1 throughout the
fire test, but that this difference reduced as the test progressed. Experimental data show
that the Tg of the epoxy was exceeded at about 35 minutes, while the model predicts this
occurrence at 43 minutes. While the model prediction is unconservative, the time at
which the epoxy Tg is exceeded is likely not a valid indicator of FRP-strengthened
members’ performance, as discussed previously. Figure 5b shows that the model
predicted the temperature of the steel reinforcement in the web of Beam 2 satisfactorily,
with a maximum difference of only 25ºC.
It is clear that the model provides only approximate temperature predictions around the
FRP layer. This is likely due in part to the sensitivity of the FRP temperature to localized
cracking in the VG insulation layer, which is a phenomenon that should somehow be
accounted for in future modelling scenarios, perhaps with an artificial increase of the
thermal conductivity of the material, or by consideration of thermal effects and moisture
migration in three dimensions as opposed to two. Overall, the model predicts
temperatures at various locations throughout the beam-slab assembly satisfactorily. This
is notable considering the probable slight variation in VG thickness along the beam span,
the potential for movement of installed thermocouples from their design locations during
fabrication, the likelihood that the concrete’s moisture content varies throughout the
section, and the exclusion of moisture movement effects from the modelling process.
Finally, it is important to note that since only two full-scale tests were completed in this
programme, model validation is based on limited data. Further model validation will be
performed in future phases of this research study as relevant specimens data become
available.

255
 800 600
Model Prediction Measured
Measured Model Prediction
500
600

Temperature [ºC]

Temperature [ºC]
400

400 300

200
200

100

0 0
0 60 120 180 240 300 0 60 120 180 240 300
(a) Time [minutes] (b)
Time [minutes]

Fig. 5 (a) Predicted versus measured temperatures at FRP/concrete interface – Beam 1.

(b) Predicted versus measured temperatures of steel rebar – Beam 2.

CONCLUSIONS
The following conclusions can be made based on the experimental studies and numerical
analyses described herein:
• FRP-strengthened and insulated beam-slab assemblies with 25 mm and 38 mm of
Tyfo® VG insulation can achieve four-hour fire endurance ratings, as defined by ASTM
E119;
• The insulation system evaluated in this research program appears to have effectively
protected the FRP strengthening systems from heat penetration. A thickness of 38 mm of
insulation maintained the average FRP temperature below the matrix Tg for 54 minutes.
In addition, the insulation system maintained low temperatures in the reinforced concrete
beams such that the structural elements did not fail under service load levels during fire.
• The two-dimensional thermal model described herein is capable of predicting
temperatures within the concrete of FRP-strengthened concrete beams with sufficient
accuracy to allow for fire endurance predictions. Further extension of the model will be
required to precisely model the thermal behaviour at the FRP layer, and
• To enhance the fire endurance of FRP-strengthened concrete members, thermal
conductivity, specific heat and application configuration of the insulation layer are key.

ACKNOWLEDGEMENT
The authors are members of the Intelligent Sensing for Innovative Structures Network
(ISIS Canada) and wish to acknowledge the support of the Networks of Centres of
Excellence Program of the Government of Canada and the Natural Sciences and
Engineering Research Council. The authors would also like to acknowledge the National
Research Council of Canada, Fyfe Company LLC, and Watson Bowman Acme
Corporation.

256
REFERENCES
[1] Ballinger, C., Maeda, T., and Hoshijima, T., “Strengthening of Reinforced
Concrete Chimneys, Columns and Beams with Carbon Fiber Reinforced
Plastics,” Fiber-Reinforced-Plastic (FRP) Reinforcement for Concrete
Structures: Properties and Applications, Nanni A. (ed.), Elsevier Science
Publishers B.V., 1993, pp. 55-86.
[2] Challal, O., and Shahawy, M., “Performance of Fiber-Reinofrced Polymer-
Wrapped Reinforced Concrete Column under Combined Axial-Flexural
Loading,” ACI Structural Journal, 97(4), 2000, pp. 659-668.
[3] Fukuyama, H., “FRP Composites in Japan,” Concrete International, American
Concrete Institute, 21(10), 1999, pp. 29-32.
[4] Saadatmanesh, H., Ehsani, M., and Jin, L., “Repair of Earthquake-Damaged RC
Columns with FRP Wraps,” ACI Structural Journal, 94(2), 1997, pp. 206-215.
[5] Grace, N.F., Soliman, A.K., Abdel-Sayed, G., and Saleh, K.R., “Strengthening
of Continuous Beams Using Fiber Reinforced Polymer Laminates,” Fourth
International Symposium on Fiber Reinforced Polymer Reinforcement for
Reinforced Concrete Structures, Dolan, C.W., Rizkalla, S.H., and Nanni, A.,
(eds.), American Concrete Institute, Detroit, Michigan, SP188-57, 1999, pp.
647-658.
[6] Tadros, G., McWhinnie, K, and Kroman, J., “Deck Strengthening for Country
Hills Bridge in Canada,” Fourth International Symposium on Fiber Reinforced
Polymer Reinforcement for Reinforced Concrete Structures, Dolan, C.W.,
Rizkalla, S.H., and Nanni, A., (eds.), SP188-41, 1999, pp. 455-466.
[7] Bank, L.C., “Properties of FRP Reinforcements for Concrete,” Fibre-
Reinforced-Plastic (FRP) Reinforcements for Concrete Structures: Properties
and Applications, Elsevier Science Publishers B.V., 1993, pp. 59-86.
[8] Blontrock, H., Taerwe, L., and Matthys, S., “Properties of Fiber Reinforced
Plastics at Elevated Temperatures with Regard to Fire Resistance of Reinforced
Concrete Members,” Fourth International Symposium on Fiber Reinforced
Polymer Reinforcement for Reinforced Concrete Structures, Dolan, C.W.,
Rizkalla, S.H., Nanni, A., eds., American Concrete Institute, Detroit, Michigan,
SP188-5, 1999, pp. 43-54.
[9] ACI, ACI 440.2R-02: Guide for the Design and Construction of Externally
Bonded FRP Systems for Strengthening Concrete Structures, American
Concrete Institute, Farmington Hills, Michigan, 2002, pp. 113.
[10] Apicella, F., and Imbrogno, M., “Fire Performance of CFRP-Composites Used
for Repairing and Strengthening Concrete,” Proceedings of the Fifth ASCE
Materials Engineering Congress, Bank L.C., ed., ASCE, Cincinnati, Ohio,
1999, pp. 260-266.
[11] Wang, Y.C., Wong, P.M.H., and Kodur, V.K.R., “Mechanical Properties of
Fibre Reinforced Polymer Reinforcing Bars at Elevated Temperatures,”
SFPE/ASCE Specialty Conference: Designing Structures for Fire, Baltimore,
MD, 2003, pp. 183-192.

257
[12] Bisby, L.A., "Fire Behaviour of FRP Reinforced or Confined Concrete,"
Doctoral Thesis, Department of Civil Engineering, Queen's University,
Kingston, Ontario, Canada, 2003, pp. 372.
[13] Williams, B.K., Fire Performance of FRP-Strengthened Reinforced Concrete
Flexural Members, Doctoral thesis, Department of Civil Engineering, Queen’s
University at Kingston, Canada, 2004, pp. 389.
[14] ASTM, E 119: Standard Test Methods for Fire Tests of Building Construction
and Materials, 2002, West Conshohocken, PA.
[15] Deuring, M. “Brandversuche an Nachtraglich Verstarkten Tragern aus Beton,”
Research Report EMPA No. 148’795, Swiss Federal Laboratories for Materials
Testing and Research, Dubendorf, Switzerland, 1994.
[16] Blontrock, H., Taerwe, L., Vandevelde, P., “Fire Tests on Concrete Beams
Strengthened with Fibre Composite Laminates,” Third Ph.D. Symposium,
Vienna, Austria, 2000, pp. 10.
[17] Bisby, L.A., Green, M.F., and Kodur, V.K.R., “Fire Endurance of FRP-confiend
Concrete: Test Results and Model Validation,” accepted, ACI Structural
Journal, 2004.
[18] Gustaferro, A.H., “Temperature Criteria at Failure,” American Society for
Testing and Materials Symposium on Fire Test Performance, Denver, Colorado,
1969, pp. 68-84.
[19] CAN/ULC, “S101-89: Standard Methods of Fire Endurance Tests of Building
Construction and Materials,” Underwriters’ Laboratories of Canada,
Scarborough, Ontario, 1989, pp. 49.
[20] Alfawakhiri, F.; Kodur, V.K.R., and Frater, G., “Temperature Field Modeling of
the First Cardington Test,” Third International Workshop on Structures in Fire,
IRC-NRC, Ottawa, Canada, 2004, pp. 59-69.
[21] Lie, T.T., Structural Fire Protection, ASCE Committee on Fire Protection,
Structural Division, American Society of Civil Engineers, 1992, pp. 241.

258