BRIDGE STRENGTHENING USING ADVANCED

COMPOSITE SYSTEM

CHAPTER 1
INTRODUCTION AND OVERVIEW
INTRODUCTION
Any technology or material has its limitations. To overcome
these limitations new materials and technology make there way.
In civil industry to meet up the requirements of advance infra
structure new innovative materials/ technologies has started
making their way. Use of FRP composites in
Rehabilitation and Strengthening of structures is becoming
increasingly popular and is opening new possibilities in
construction and rehabilitation of structures. In the present paper
we present use of FRP for strengthening and rehabilitation for
bridges. Bridges are important and critical structures and present
many challenges for replacement or repair.
Many of the
advantages in these materials have proven to be revolutionary in
terms of time constraints and durability of these structures. This
paper presents three bridges which were either rehabilitated or
strengthened by using FRP composites. The resulting structure
was then tested for the effect after using FRP composites for
Rehabilitation and strengthening.
It is becoming
preferable, both environmentally and economically to upgrade
bridges rather than to demolish and rebuild them. The

deterioration of bridges from environmental influences and from

traffic loads require rehabilitation and renewal programs to
maintain even current service levels on the bridge infrastructure
network. There are increased demands for high durability
,longer service life , reduced maintenance cost and cost /
performance optimization. Advanced Composite Systems have
now become a viable method of strengthening existing bridges
worldwide.
This paper presents the evolution of carbon
fiber systems since 1991. In todays world, construction
engineers are faced more and more frequently with the task of
strengthening existing bridges in order to secure or even
increase their load bearing capacity. Many different
strengthening techniques are available, such as installing
additional steel, external posttensioning,bonded reinforcement,
increasing the concrete cross-section, etc.
In todays world, construction engineers are faced
more and more frequently with the task of strengthening existing
Bridges in order to secure or even increase their load bearing
capacity. Many different strengthening techniques are available,
such as installing additional steel, external posttensioning,
bonded reinforcement, increasing the concrete cross-section, etc.
The use of bonded steel plates has been used successfully since
the late 1960s.
External plate bonding is a
method of strengthening which involves adhering additional
reinforcement to the external faces of a structural member. As a

result of extensive research and development projects at the

Swiss Federal Laboratories for Materials Testing and Research
(EMPA), Dbendorf, Switzerland it was proven that the use of
steel plates can be replaced with composite materials. The highstrength Carbon Fiber Reinforced Polymer (CFRP) plate system
was applied for the first time outside the laboratory in 1991 for
strengthening the Ibach Bridge in Lucerne, Switzerland. In
virtually all applications for bridge strengthening, advanced
composite systems have been proven to be structurally efficient,
easy to handle and install on job site, and cost competitive when
compared to other conventional strengthening methods.
Why do bridges need strengthening?
There are many reasons why it may be necessary to strengthen
bridges. These include:
Corrosion of reinforcement
Corrosion of pre-stressing cables
Modified Codes and Standards
Increased permanent and traffic loads
Inadequate structural design
Seismic retrofitting

Advanced composite systems are comprised of fibers (typically

carbon, glass, or aramid) and resins (typically epoxy resins).
Only long term tested (fatigue resistant) and approved systems

should be recommended for strengthening purposes in Bridge

Engineering. Installation that involved minimum labor and
traffic interruption.This paper reports on the strengthening
techniques, design approach, and testing of the bridges.
What do we achieve by adopting this Technique?
In adopting the technique of advanced composites systems it is
possible to:
Increase the flexural strength
Increase the shear strength
Increase the seismic resistance
Increase the confining strength
OVERVIEW
The service life of bridges is often reduced due to the corrosion
of steel reinforcing bars in bridge decks and to the cracking
caused by loading in excess to the original design values due to
increased traffic volumes. In Indiana, numerous bridges are in
need of upgrading or rehabilitation. Current upgrading practices
include replacing the part of deteriorated portion of the deck
structure by patching damaged areas or replacing the whole deck
structure. Both of these practices have drawbacks. The first is
time-consuming and provides only a short-term solution, while
the latter is expensive and causes severe traffic disruption.
Therefore, alternative solutions should be devised for the

rehabilitation and upgrading of deteriorated bridge decks in

Indiana. Many industries, such as the aerospace and the
automotive industries have successfully used Fiber Reinforced
Polymer Composites (FRPC). These types of composite
materials offer significant advantages over conventional civil
engineering materials, such as concrete and steel. This is due to
their chemical and corrosion resistance, lightweight, and high
strength, which make them attractive for the rehabilitation of
civil infrastructures. Strengthening of Reinforced Concrete (RC)
structures by bonding external steel plates and composite plates
or sheets is an effective method for improving structural
performance under both service and ultimate load conditions. A
main disadvantage of using steel plates is the potential for
corrosion at the epoxy/steel interface with consequent reduction
in bond strength when exposed to harsh environments.
Composite plates or sheets, on the other hand, offer several
advantages over their steel counterparts, such as ease bondage to
irregular surfaces, lightweight, etc. FRPC have been used in the
replacement of deficient bridge decks. Studies of the feasibility
and long-term performance of this type of application have been
conducted. These studies have concluded that not only FRPC
decks should be considered as an alternative to conventional
reinforced concrete decks; they have a number of advantages
over the latter. In particular, their ease of construction should be
highlighted: instead of weeks only a few days are required for
their successful installation and consequently, traffic disruptions
are minimized. The objective of this research project is to study

the feasibility of using of FRP as a retrofit or construction

material for bridge decks. This has been accomplished by means
of a comprehensive literature review of externally bonded FRPC
strengthening systems and of the current state of knowledge on
technologies involved in the design and construction of FRPC
bridge decks. In addition, valuable information has been
obtained through a web-based survey of other state Departments
of Transportations (DOTs) on their experience with FRPC
materials for bridge decks.
FINDINGS:
The results from the literature review indicate that by externally
bonding FRP plates (or sheets) and/or rods provide excellent
retrofitting mechanisms to increase deck strength as well as
stiffness of aging or deteriorated structures. The advantages of
this retrofitting method include reduced labor costs, minimum
shutdown time/cost and traffic disruption, and minimal
maintenance requirements. From the literature review, it was
found that the values of such the increase in stiffness and
strength varied for the different field applications. However, in
all cases such an increase was observed. Furthermore, it was
also found that the benefits of such a retrofitting system do not
change with time. A number of demonstration projects that
studied FRP bridge deck panels have been conducted
countrywide. These projects range from small-scale pedestrian
bridges to large-scale highway bridges as well as from deck

replacement to bridges made entirely of composite materials.

Most of the studies report that their FRP applications are
performing very well. In fact, some of these applications are
now 3 or 4 years old and continue to show excellent
performance. In all cases, it is reported that the installation time
is significantly reduced when compared to conventional
reinforced concrete decks. The experience of other state DOTs
in the use of FRP as a retrofit and as a construction material for
bridge decks was investigated by means of a web-based survey.
All 50 state DOTs were contacted and 34 responded the survey.
Of the responding DOTs, 23 responded that they have used FRP
for bridge desk rehabilitation and/or installed FRP bridge decks.
The major reasons provided by these states for adopting FRP
materials were their excellent strength, lightweight, and
durability. Most of the states using FRP as a material for bridge
deck rehabilitation reported that its main use was to strengthen
and upgrade damaged bridge decks. Eight states responded that
they had replaced a reinforced concrete bridge deck by a FRP
bridge deck. Based on their experience, these DOTs have not
observed any problems with their FRP application. Twenty state
DOTs have responded that they are considering using FRP in the
future. Most of them plan to utilize FRP as a
strengthening/upgrading system. The results from the literature
review and DOT survey indicate that FRP materials have been
successfully used in civil infrastructure applications, and in
particular for bridge deck strengthening and replacement. It also
appears, from the results of this study that the use of FRP in

bridges is likely to continue and potentially become a

mainstream material in the near future.
IMPLIMENTATION
The current state of knowledge of FRP materials as a
construction material for civil infrastructure indicates that it can
be successfully used in many types of applications. The present
study focuses in their use for bridge decks. In order to further
benefit from this technology, Indiana must become part of the
increasing research efforts in this area. Therefore, it is strongly
recommended that a demonstration project be developed in this
state. With this in mind, a proposal has been developed and
submitted to the FHWA Innovative Bridge Research and
Construction (IBRC) program. In the proposed project, the three
main spans of a bridge deck in Tippecanoe County will be
replaced by 8 FRP deck panels. The scope of this project
includes the evaluation and design of FRP bridge deck panels to
meet current code requirements. It also involves the
reconstruction of an existing bridge deck using the innovative
FRP deck panels. The monitoring of the performance of the
developed application will also be part of the proposed IBRC
project.

CHAPTER 2
BACKGROUND
2 BACKGROUND
The service life of bridges is often reduced due to the corrosion
of steel reinforcing bars in bridge decks and to the cracking
caused by loading in excess to the original design values due to
increased traffic volumes. In Indiana, numerous bridges are in
need of upgrading or rehabilitation. Current upgrading practices
include replacing the part of deteriorated portion of the deck
structure by patching damaged areas or replacing the whole
deck structure. Both of these practices have drawbacks. The
first is time-consuming and provides only a short-term solution,
while the latter is expensive and causes severe traffic disruption.
Therefore, alternative
solutions should be devised for the rehabilitation and upgrading
of deteriorated bridge decks in Indiana. Many industries, such
as the aerospace and the automotive industries have
successfully used Fiber Reinforced Polymer Composites
(FRPC). These types of composite materials offer significant
advantages over conventional civil engineering materials, such
as concrete and steel. This is due to their chemical and
corrosion resistance, lightweight, and high strength, which
make them attractive for the rehabilitation of civil
infrastructures.

Strengthening of Reinforced Concrete (RC)

structures by bonding external steel plates and composite plates
or sheets is an effective method for improving structural
performance under both service and ultimate load conditions. A
main disadvantage of using steel plates is the potential for
corrosion at the epoxy/steel interface with consequent reduction
in bond strength when exposed to harsh environments. Other
disadvantages are transportation, storage, installation difficulties
as well as increase to the structure self-weight.
Composite plates or sheets,
on the other hand, offer several advantages over their steel
counterparts, such as ease bondage to irregular surfaces,
lightweight, etc. Figure 1.1 shows a comparative sketch of the
procedures usually required for the installation of these two
types of retrofits. Another exciting application involves the use
of FRPC in the replacement of deficient bridge decks. Some
investigative studies have been conducted to date to study the
feasibility and longterm performance of this type of application.
These studies have concluded that not only FRPC decks should
be considered as an alternative to conventional reinforced
concrete decks; they have a number of advantages over the
latter.
In particular, their ease of construction should be
highlighted: instead of weeks only a few days are required for
their successful installation and consequently, traffic disruptions
are minimized. While it may be too soon to tell, it is expected
that FRP applications will have a much longer life span than

applications that use traditional civil engineering materials, since

FRP is corrosion resistant. However, more research is needed to
determine the long-term behavior of these materials under
various environmental and loading conditions. While composite
materials have been widely used in other industries, their
application to Civil Infrastructures is relatively new. However,
both researchers and practicing engineering have recognized that
these materials will eventually become part of the civil industry
mainstream.
FRP plates or sheets provide an
effective solution for strengthening bridge decks that have
become deficient due to deterioration, additional service loads or
excessive deflections created by change in use, construction or
design defects, or code changes. Furthermore, FRP deck panels
are a promising alternative as a replacement of conventional
reinforced concrete bridge decks. This report focuses on these
two applications of FRP to bridge decks.
2.1 OBJECTIVE
The objective of this research project is to study the feasibility
of using of FRP as a retrofit or construction material for bridge
decks. This has been accomplished by means of a
comprehensive literature review of externally bonded FRPC
strengthening systems and of the current state of knowledge on
technologies involved in the design and construction of FRPC
bridge decks. In addition, valuable information has been

obtained through a survey of other state Departments of

Transportations (DOTs) on their use of FRPC materials.

2.2 ORGANISATION OF THE REPORT

The organization of this report is provided next. In Chapter 2, a
literature review on the usage of FRPC strengthening systems
for bridge decks is carried out. Chapter 3 presents the current
state of knowledge of FRP bridge decks. On both of these
chapters, lists of relevant manufacturers are provided. In
Chapter 4, the results from the survey of all state DOTs are
summarized. Finally, in Chapter 5 recommendations are
provided to INDOT for the implementation of FRP decks in
Indiana.

Figure 1.1 Installation of bridge deck retrofits (Emmons et al.,

1998)

Figure 1.2 Components of the bonding material in FRPC sheets

(Emmonset al., 1998)

CHAPTER 3

FRP AS EXTERNAL/INTERNAL RETROFITS FOR

BRIDGE DECKS
3.1 INTRODUCTION
Advanced composite materials usually have two components: a
reinforcing element and a supporting matrix. The reinforcing
element is, in general, much stiffer and stronger than the matrix
and as such, it is the load-carrying element. The matrix, on the
other hand, provides lateral support for the reinforcing element
(Teng et al. 2000). The matrix in Fiber Reinforced Polymer
Composites (FRPC) consists of a polymer/resin used as a binder
material. It supports and separates the fibers, and it protects the
fibers against severe environmental conditions. Thermosetting
polymer resins are the most common types of matrix element.
In particular, polyesters,
epoxies and phenolics are the most frequently used resins in
civil engineering applications. The FRPC reinforcing elements
are used to provide the stiffness and strength to composite
materials. These reinforcing element materials, which are
typically used in civil engineering applications, are usually made
of carbon (graphite), glass, and aramid (Kevlar ) fibers. They
are imbedded in a resin matrix (e.g. epoxy resins) and they
provide most of the tensile strength of the composite just as steel
does in reinforced concrete. FRPC is usually manufactured in a
continuously woven form with different lengths or directions in
order to provide the best performance for different applications.

Using externally bonded FRPC plates

or rods to retrofit structures has been shown to be a practical
method for strengthening aging or deteriorated structures. The
advantages of this method include reduced labor costs,
minimum shutdown time/cost and traffic disruption, and
minimal maintenance requirements. This chapter focuses on the
application of this technology to bridge decks. Section 2.2
provides a literature review of the published research in which
FRP has been used as a retrofit for deficient reinforced concrete
bridge decks. In Section 2.3, the different manufacturers of these
types of FRP retrofits are provided.

3.2. LITERATURE REVIEW

Nanni (1995)
In this work, several applications of externally bonded FRP
reinforcement of concrete structures developed in Japan are
discussed. According to the author, the function of these retrofits
depends on the type of application, i.e., it may be any
combination of strengthening, stiffening, crack arrest, or
corrosion protection. In particular, two examples of bridge deck
retrofitting are highlighted. They are the Hata and Hiyoshikura
bridges.
The Hata Bridge (Figure 2.1) is located in Kyushu Highway in
Southern Japan. In this application, FRP sheets were installed on
the soffit of the cantilevered wing slab to provide the needed
additional capacity caused by the installation of a larger
windbreak wall. This project took was conducted in the spring
of 1994.

Figure 2.1. Hata Bridge Japan (Nanni, 1995)

Two layers of carbon FRP (CFRP) were
applied both parallel and perpendicular to the traffic direction. In
some critical locations three plies were used. The sheets were
applied by roller brushing the adhesive to the underside
followed by the application of the FRP sheet, as shown in Figure
2.2. The fiber were always oriented in the direction parallel to
the long dimension of the sheets, which were 50 cm wide and
the length was cut to size. On-site loading test were conducted to
test the effectiveness of the strengthening method. More

specifically, these tests showed the strains were reduced

considerably on the steel reinforcement.

Figure 2.2. Installation of FRP sheets on the soffit of the

cantilevered slab of Hata Bridge (Nanni, 1995)
In the spring of 1994, the deck of the
Hiyoshikura Bridge located on the Tokando Highway was in
need of upgrading due to the increased traffic load and the
presence of mapping cracks. The bridge consisted of a
reinforced concrete deck supported by steel girders. Instead of

replacing the deck, the cracks were sealed and FRP wraps were
applied to the underside of the 8 deck for strengthening. More
specifically, the area of soffit of the deck (164 m2 or 1760 ft2 )
was covered with two layers of CFRP placed parallel and
perpendicular to the traffic direction (Figure 2.2).
In order to evaluate the developed application,
strain gages were installed on steel reinforcing bars on the
underside of the deck. Running vehicle tests were conducted that
showed that the tensile strain in the steel reinforcement reduced
by 30 to 40%.
Hoa et al. (1996)
In this work, the effect of environmental conditions, in particular
temperature and moisture effects, on structures repaired by
externally bonding carbon/epoxy composite sheets is
investigated. Portland cement was used to cast concrete
specimens. The proportion of cement: sand: coarse aggregate
was 1:2:3 in volume. The formwork was removed 24 hours after
casting. The curing time was 28 days at room temperature. The
average cylinder strength of the concrete after 28 days was 18
MPa. Unidirectional graphite/epoxy composite sheets were
used. The thickness of the composite plates varied from 0.33
mm (3 layers) to 6 mm (45 layers).

Before bonding the FRP sheets to the concrete surfaces, these

surfaces were prepared by:
(a) sandblasting until the aggregates were exposed;
(b) washing with water and blasting it with air for drying; and
(c) cleaning with acetone.
The preparation of the surfaces of the
composite sheets consisted of sanding with sand paper and then
cleaning with acetone. Both accelerated tests and long-term
environmental tests were conducted on the developed
specimens. Two types of accelerated tests were performed. In
one of them, the specimens were immersed at room temperature
for 60 days. In the other, hot-cold cycles were applied, i.e., 9
samples were placed in an oven at 40 C for one week and then
in a refrigerator at -23C for another week. The total process
lasted 60 days. In addition, four samples were left outdoors,
under Montreal weather conditions, for the long-term
environmental testing. Strength tests were conducted on two of
these specimens after 200 days, while the remaining samples
were tested after 28 months exposure. Three-point bending tests
using an MTS machine were carried out to investigate the effect
of externally bonding composite sheets to concrete with
composite sheets. The findings from the exposure and strength
tests can be summarized as follows:

1. The use of externally bonded FRPC sheets to structural

members can increase the flexural loading bearing capacity by
up to 49%.
2. Increasing the thickness of the composite sheet did not seem
to lead to an improvement in strength. Instead, a decrease in
strength was observed when composite sheet became too thick.
The length of the composite sheet had a noticeable effect on the
strength, i.e., the longer the composite sheet, greater the
strength.
3. The exposure to water for 60 days at room temperature of
samples retrofitted with FRPC sheets had no significant effect
on their load bearing capacity.
4. The specimens subjected to 200-day and 28-month long-term
outdoor exposure showed a reduced load bearing capacity. In
both cases, this reduction was less than 7%, even though
samples subjected to 28-month exposure exhibited traces of
debonding between the concrete and composite sheet. It is
interesting to note that the results obtained by the accelerated
tests using hot-cold cycles are quite close to those of long-term
exposure and in the conservative side. The results also suggest
that the effect of temperature is more important than humidity in
term of the reduction of bonding strength. Humidity alone seems
to only have a hardening effect on the samples.

In conclusion, the authors have

found that using hot-cold cycles is an effective method for
accelerated testing of the long-term performance of FRPC sheet
retrofitted specimens. Finally, they concluded that using
externally bonded FRPC sheet can restore the load bearing
capacity of deficient specimens.
Arockiasamy et al. (1996)
In this study, two solid slabs 1219 mm x 305 mm x 4420 mm
(48 in. x 12 in. x 14 ft. 6 in.) and two voided slabs 1194 mm x
203 mm x 6553 mm (47 in. x 8 in. x 21 ft. 6 in.) were studied.
Both of these slabs were pre-cracked and then one of each type
was reinforced with externally bonded CFRP plates to evaluate
the contribution of the retrofit to the strength and stiffness of the
slabs. The specimens were loaded to failure after complete cure
of the adhesives.
From the tests, it was observed that failure mode of the
retrofitted solid slab occurred by crushing of concrete at
midspan, while the control precracked slab failed at point of
application of the load. The results show that by retrofitting
severely damaged solid slab with CFRP laminates, improve
significantly its flexural capacity (approximately 90% of the
flexural capacity of the uncracked slab). The retrofitted voided
slab experienced a sudden and catastrophic failure. This
suggests that prior damage to the slab may have existed leading

to local concrete crushing failure. The retrofitted solid and

voided slabs exhibit larger deflection than the control precracked
slabs at both service and ultimate loads. Crack patterns of the
retrofitted slabs were identical to those of the control slabs.
Alkhrdaji et al. (1999)
In this study, a full-scale application was tested to investigate the
effectiveness of using FRPC to strengthen actual bridge decks.
More specifically Bridge J-857, located on Route 72 in Phelps
County, Missouri, was chosen for testing and demolition. Fieldtests using CFRP sheets and rods as strengthening systems were
conducted. In addition, a test of a non-strengthened bridge deck
was also conducted for comparison. Figure 2.3 depict the
strengthening schemes used in the three bridge decks. Figures
2.4 (a) and (b) show the strengthened bridge deck with external
CFRP sheets and rods, respectively.

Figure 2.3 Retrofitting scheme used in the three bridge decks

(Alkhrdaji et al., 1999)
Examination of the test results
indicates that both strengthening systems were successful. The
specific findings from the field-testing data are given below:
1. The increase in the moment capacity was 17% and 27% for
CFRP sheets and rods, respectively.
2. The strengthened decks had smaller deflections (therefore
higher stiffness) and higher load capacity at the ultimate loading
conditions.
3. The CFRP rod system provides slightly better benefits than
those of externally bond CFRP sheets.
Additional advantages observed included
minimal surface preparation, 12 rapid installation time, and
ability of anchoring the reinforcement into adjacent RC
members.

(a)

(b)
Figure 2.4 Strengthening schemes: (a) FRP sheets, (b) FRP rods
(Alkhrdaji et al., 1999)

Rizkalla and Labossire (1999)

This article describes some projects in Canada that use FRP
materials to strengthen bridge. One of such projects consisted of
the application of CFRP to internally strengthen a bridge deck
underneath the overlay. The developed application is shown in
Figure 2.5. The structure is referred to as the Country Hills
Boulevard Bridge in Calgary, Alberta, Canada. The main reason
for the bridge strengthening was that it was found that its thin
deck would overload under full truck loading. The main
considerations that lead to the decision to use such a retrofit,
included the fact that they did not wish to replace the whole
deck (nondestructive alternative) and that they wished to
minimize traffic disruption. The procedures used in the
development of this application were:

1. CFRP strips were installed at 20 inches center to center.

2. The deck surface was rough. A layer of Sikadur 30 with sand
aggregate was applied for leveling purposes.
3. The CFRP strips were applied with epoxy after one day.
4. The excess epoxy in each strip was removed through rolling.
5. The surface of each strip was cleaned and sanded after one
day.
6. A bonding agent was applied on the back surface of each strip
four hours prior to the installation of the overlay.

Figure 2.5 FRP strips applied on deck of the Country Hills

Boulevard Bridge (Rizkalla and Labossire, 1999)
Another project described in this article is
the strengthening of the Ste-mlie-de-lnergie Bridge in
Qubec, Canada using FRP materials. The site preparation
included a curing time of the concrete used in the repair of four
weeks. The composite strips were installed in eight days over a
period of three weeks. The CFRP strips were 50 mm (2 in) wide.
The behavior was monitored using strain gages, thermocouples,
and optic fibers with Bragg sensors or Fabry-Perot 14 sensors.
The Ministre de Transports performed loading test both prior

and after the repair. The goals of increasing the bending strength
by 35% and the shear strength by 20% were achieved.
Taerwe and Mathys (1999)
In this article, the strengthening of damaged concrete structures
using FRP is discussed. In particular, the strengthening of the
Tannberg Bridge in Austria is mentioned, in which CFRP fabric
strips were applied to the underside of the bridge deck as shown
in Figure 2.6. Freyssinet manufactured these sheets, which are
referred to as TFC sheets. Details on
the strengthening scheme and on the performance of the
developed scheme are not provided in this article.

Figure 2.6 Strengthening of the Tannberg Bridge, Austria

(Taerwe and Mathys, 1999)
Mayo et al., 2000
Another application of reinforcing bridge decks with FRPC is
that of Missouri Bridge G-270. The damaged bridge is shown on
Figures 2.7. The strengthening method used in this application
consisted of CFRP sheets externally bonded to the underside of
the bridge deck. The Figures 2.8 shows the application of the
adhesive prior to the application of the CFRP sheets and Figure
2.9 shows the installation of the sheets themselves. The goal of
this strengthening project was to increase the flexural capacity
of the bridge.

Figure 2.7. Bridge G-270, Missouri (Mayo et al., 2000)

Figure 2.8. Application of the adhesive to the underside of the

deck (Mayo et al., 2000)

Figure 2.9. Installation of the CFRP sheets (Mayo et al., 2000)

Both full-scale laboratory and in-situ field tests were conducted
before/after strengthening to evaluate the effectiveness of the
developed strengthening system. Furthermore, the long-term
performance of the bridge has also been monitored.
In-situ field tests were conducted before
and after strengthening to evaluate the effectiveness of the
developed strengthening system in May 1998. These load tests
measured the deflection due to a load truck driving over the
bridge. Six passes were made by the truck on the North and
South sides and on the centerline. It is found that in average, the
deflections after strengthening were 94% of the original
deflections. However, it was observed that in the more severely
deteriorated areas the reduction was more significant (at most
77%).

In August 1999, a second load test

was performed on August 19, 1999. This foal of this test was to
investigate the effects of time on the performance of the system.
Once again, it was found that the deflections were not uniform
throughout the bridge. It was concluded from this second loaddeflection tests that the FRP sheets continue to carry tensile
stresses. In fact, they found that the deflections are almost the
same as those measured in May 1998.
2.3 Manufacturers of External FRP Reinforcement Systems for
Bridge Decks
A number of FRP manufacturers, which were originally
dedicated to other industries such as the automotive and
aerospace industries, have been alternatively re-focusing their
scope to the civil engineering industry. Among these
manufacturers, the ones that have participated in most of the
developed field applications are members of the Market
Development Alliance of the FRP Composites Industry (MDA).
This self-funded, non-profit trade alliance is a consortium of
organizations with interest in FRP composites. In addition to
manufacturers and material suppliers, it also includes owners,
constructors, consultants and designers. The mission of MDA is
to identify and market new applications for FRP products. The
manufacturer members for FRP reinforcing systems for bridge
decks are provided in this section.

The use of FRP materials to strengthen

concrete structures can be traced to the 1950s, however their use
as an external reinforcement of concrete bridge structures began
in 1980s (MDA 2000). According to MDAs report (MDA
2000), more than 1000 bridges (concrete slab/steel girders) in
Japan have been strengthened by bonding FRP sheets to the slab.
In the U.S., this technology has been widely used to retrofit
columns for seismic upgrade. Of the companies that specialize
in the use of FRP sheets to retrofit bridge structures, the ones
that have used this technology to upgrade bridge decks are listed
below.

COMPTEK Structural Composites, Inc.

This manufacturer is based in New York City, while its
manufacturing facility (National Composites Center NCC) is
located in Dayton, Ohio. Their product, referred to as the
ATLAS System, has been developed to strengthen structural
components (beams, columns and slabs). Both sheets and rods
are manufactured using this system. Of their products, those that
have been used to reinforce bridge decks are the Atlas Carbon
Laminates (ACL) and the Atlas Carbon Rod (ACR). The ACL
can be installed on the underside of the bridge for strengthening
(Figure 2.10), while the ACR can be embedded in the concrete
slab (Figure 2.11).

Figure 2.10. Atlas carbon laminates installed on underside of a

concrete slab

Figure 2.11. FRP Rods embedded into a concrete slab

Fyfe Co.,LLC :
While this companys headquarters is located in San Diego,
California, it has representatives throughout the U.S. and the
world. Their product, the Tyfo Fiberwrap System uses wet
layup and preformed composites (unidirectional or bi-directional
glass or carbon fibers) for strengthening of structural
components. It has been mostly used for seismic retrofit of
columns, but it has also been used to strengthen beams and slabs
both in the positive and negative moment regions. Most of their
completed projects have been on building structures, however, it

has the potential to be successfully used to upgrade bridge

decks.

Master Builders, Inc.

This manufacturer, located in Cleveland, Ohio, has developed
the MBrace Composite Strengthening System, which is
externally bonded to concrete or masonry structures to increase
their strength. MBrace uses unidirectional aerospace grade
carbon, E-glass, or aramid fiber 20 fabrics embedded in
engineered materials that include epoxy surface primers, putty
fillers, and high solids resins. One of their completed projects is
the upgrade of the MoDOT Bridge G270 in Iron County, MO
(Figure 2.12) in May 1998. The MBrace system was applied to
the underside of the bridges deck to allow for a larger load
rating.

Figure 2.12. MoDOT Bridge G270 (Iron County, MO)

CHAPTER 4
FRPC BRIDGE DECKS
4.1. INTRODUCTION
Due to aging, environmentally induced degradation, poor initial
construction, overloading, and lack of maintenance, bridge
components such as decks, superstructures, and columns may

become deficient. Nearly 40% of all highway bridges in the

USA are classified as either functionally or structurally
deficient, and in approximately one-half of these bridges this
deficiency can be attributed to their decks (Hayes et al. 2000). It
is estimated that a traditional bridge deck lasts 35 years on
average; however, in cold regions, such as the Midwest of the
U.S., the life of a deck averages 10 years. This is because of the
extensive use of de-icing salts during winter months (Karbhari et
al. 1997).
During the past decade, the use of
fiber-reinforced polymer composites (FRPC) in civil
infrastructures has begun to receive significant attention by the
civil engineering community. This is because these materials
offer significant advantages over conventional materials due to
their chemical and corrosion resistance, lightweight, and high
strength. However, much of the research carried out in this area
has focused mainly in the use of these materials to retrofit
existing deficient structural components such as columns,
beams, and slabs. However, an exciting applicationn involves
the use of FRPC in the replacement of deficient bridge decks.
This section provides a summary of the research and
manufacturing information available in this area.
Section 4.2 provides a literature
review of the published research projects in which FRP deck
panels were developed. This literature review is organized

chronologically, i.e., from older to more recent publications. In

Section 3.3, the different manufacturers of FRP bridge deck
panels are provided.

4.2. LITERATURE REVIEW

Johansen et al. (1997)
In the work by Johansen et al. (1997), two fiberglass reinforced
plastic (GFRP) truss bridges in the Golden Gate National
Recreation Area, CA were investigated. GFRP was chosen for
durability and maintainability reasons. The original bridges had
maintenance problems, since they were made with conventional
materials, such as wood (which experienced rotting), and steel
and concrete (which experienced severe corrosion). The lengths
of the two developed bridges are 35 ft and 70 ft. The region
where the bridges were installed is prone to seismic attacks and
extreme wind conditions. Therefore, these possibilities were
considered in the design of the bridges. In the final design, the
improvement of the overall strength and stiffness was achieved
by means of camber, X-bracing, and steel bolts connections
(Figure 3.1). The installation of each bridge took approximately
1.5 hours. Both bridges were airlifted and placed along the
cliffs. The total time taken to design, fabricate, ship and install
these bridges was approximately 60 days. The total cost of the
project (including design, fabrication, and shipping) was

$45,000. No testing or long-term monitoring results are reported

in this paper. However, the authors point out that developed
applications illustrate that the use of FRP is feasible for
longspan bridges because of its unique strength/stiffness
characteristics combined with its lightweight. This is particularly
important for applications with difficult site constraints, such as
those where the developed applications were installed. The
authors also report that the two bridges were easy to assemble
and install, and that they are practically maintenance-free when
23 compared to bridges made of conventional civil engineering
materials, such as concrete, wood, and steel.

Figure 3.1. 70-ft long FRP pedestrian bridge (Johansen et al.

1997)

Karbhari et al. (1997)

In Karbhari et al. (1997) an experimental program was
conducted to investigate different configurations of bridge deck
panels, from various manufacturers. The FRP deck panels were
developed using the following three criteria: stiffness
requirements, displacement limits, and cost. The tested
specimens ranged from subcomponent, component, and fieldsize levels. Figure 3.2 shows the different tested panels. The
main goal of their tests was to study the effectiveness of the
various deck panel configurations. To achieve this, quasi-static
testing of a number of FRP deck specimens were conducted. In
all cases, it was found that all the FRP deck specimens have
much higher failure loads and comparable initial stiffness than
that of the reinforced concrete specimen. They have also found
that the box and trapezoid configurations have significantly
better energy absorption capacity. Notably, one of their main
findings was that the FRP deck components 24 continued to
carry load even when substantial cracking and fracture had
occurred, i.e., no catastrophic failure was observed.

Figure 3.2. Tested deck panel specimens (Karbhari et al. 1997)

Overall, they concluded that FRP decks are a
suitable alternative to conventional civil reinforced bridge decks.
Furthermore, they found that these decks could be fabricated
using many different processes. Some related topics that were
not addressed in this work include: the response under dynamic
loads, the behavior of the connections between deck and girders,

and deck and barrier and side rails, the effect of the different
material properties between FRP and existing substructure, and
the long-term durability.
Chajes, M. et al. (1998)
The paper by Chajes, M. et al. (1998) discusses the evolution
and status of three bridges made of advanced composites in
Delaware. In this research extensive monitoring through both
initial load testing and long-term monitoring programs were
developed 25 The three bridges were selected such that they
were incrementally more complex and had more restrictive
service requirements. These bridges were designed using the
AASHTO LRFD Bridge Design Specifications and factors
taking into account deterioration of material properties over time
were used (for a life span of 75 years). Both strength and service
limit states were considered, including the effects of fatigue
loading. The first bridge, the Magazine Ditch Bridge, is a 22 m
long, single-span, simply supported bridge (Figure 3.3). It was
installed on a private service road and it was completed on June
23, 1997. This bridge carries a small traffic volume, even though
it is also traveled by heavily loaded maintenance vehicles. The
developed bridge is made of glass fiber reinforced polymer
(GFRP). A 45-mm wearing surface made of latex modified
concrete was installed on the deck surface. The installation of
the bridge superstructure, including the edge girders and the
GFRP composite deck, was completed in a one day.

Figure 3.3. The Magazine Ditch Bridge (Chajes, M. et al. 1998)

Laboratory tests on sub-components and on
a full-scale portion of the deck (1.2 m long by 6 m wide) were
performed at the University of Delaware. The test program
included the application of AASHTO service and strength loads,
and fatigue tests of up to 2,000,000 cycles. The second bridge,
Bridge 1-351, replaced an existing bridge in the state of
Delaware (Figure 3.4). The original bridge was a 9 m long by 12
m wide simply supported slab bridge. The developed GFRP
bridge is 9 m long by 8 m wide, with an all-composite deck.

Figure 3.4. Bridge 1-351 (Chajes, M. et al. 1998)

Laboratory tests conducted at the
University of Delaware have shown that fatigue cycles (up to
2,000,000 fatigue cycles) do not cause significant losses in
strength and stiffness to the GFRP deck. The design of this
bridge wasimilar to that of an adjacent reinforced concrete
bridge design. The third bridge, Bridge 12, is located in Rout 13
in Delaware (Figure 3.5). It represents a typical highway bridge,
since it has multiple spans, is a heavily traveled road, and carries
large volume of truck traffic. The scope of this project included
the design, structural certification, sub component testing,
fabrication, construction, and monitoring and evaluation.

Figure 3.5. Bridge (Chajes, M. et al. 1998)

The University of Delaware is
currently monitoring the three bridges. The three important limit
states considered in this study were serviceability (deflection),
strength (stress and strain), and fatigue. The ultimate goal is to
correlate the measured responses to the laboratory test results.
The most important parameters that are being measured in the
monitoring program include: traffic statistics (including number
of trucks and classifications), strains (both longitudinal and
transverse), deflections, and daily weather conditions
(temperature and humidity). The collected data is being used in
the performance evaluation of the bridges with respect to the
following effects: live load, sustained load, environmental,
thermal, and fatigue.

Walker (1998)
The paper by Walker (1998) describes a bridge installed over the
No Name Creek west of Russell, KS, which was opened to
traffic in November 1996. The bridge was made in three
sections, each 2.74 m (9 ft) wide by 7.01 m (23 ft) long, which is
the length of the bridge. The sections were assembled at the
bridge site. Strain gages were installed in the core for field 28
monitoring. The bridge was designed to withstand standard
highway traffic loads as specified by the AASHTO standards. It
was made entirely of fiberglass and resin. Two fiberglass plates
sandwiching a fiberglass honeycomb core form the bridge deck.
A polymer concrete wearing surface was installed on the top
surface to improve traction. It took two working days to install
this composite bridge.

Figure 3.6. Plan view of core of the deck system (Walker 1998)
A plan view of the core of the deck system used in this bridge is
shown in Figure 3.6. The advantage of this core geometry is that
by changing the period or amplitude of the sine wave the
behavior can be easily modified. Furthermore, the sine waves
can be connected to a flat plate as shown in
Figure 3.7 (a), or alternatively they could be connected only to
the facings as shown in Figure 3.7 (b).

(a)

(b)
Figure 3.7. Different core geometries (Walker 1998)
Lopez-Anido et al. (1998), GangaRao et al. (1999), GangaRao
and Cairo (1999)
In these three papers, two demonstration projects are discussed
that involve two advanced composite bridges installed on
secondary roads in West Virginia. These bridges are the Laurel
Lick Bridge (short-span FRP bridge) and the Wickwire Run
Bridge (FRP deck on steel beams). The West Virginia
Department of Transportation Division of Highways (WVDOH)
bridge engineers were the lead participants in these efforts.

Both bridge decks were engineered using Eglass FRP. The composite deck cross-sectional shape and fiber
architecture was designed to withstand highway bridge loads
while minimizing the weight. The core of the decks consists of
full-depth hexagons and half-depth trapezoids as shown in
Figure 3.8. The decks were built with a depth of 203 mm (8 in),
since this is the typical depth of concrete decks for highway
bridges.
The authors point out that the Pultruded
FRPC deck modules fabricated for these field applications have
some of the advantages of the pultrusion process, namely: its
low labor and 30 operating costs, minimal production of
material waste, and high production rate. However, they also
mention that pultruted FRP decks may exhibit high stress
concentration at re-entrant angles, which may lead to horizontal
shear failure.

Figure 3.8. Components of the H-Deck (GangaRao and Craigo

1999)
The deck panels were formed by
connecting FRP deck modules (20-foot long by 16-foot width)
with shear keys (12.7 mm (0.5 in) blind fasteners) to provide the
necessary interlocking mechanism. In addition, a two-part
polyurethane was used to bond the FRP deck to the FRP beams,
to increase the composite action. This adhesive was chosen
because it has good elongation, high peel and energy absorbing
properties, fatigue resistance, environmental resistance, working
time of at least 30 minutes, minimum surface preparation,
acceptance of variable bond line thickness, 0.5-3 mm, good gap
filling capabilities, and ease of application for field conditions.
The developed FRP composite deck modules were
installed transverse to the traffic direction. The depth of the
decks was kept at 8 since they were used as replacement to the
conventional concrete decks. The connection between the FRP
deck modules and the steel girders was achieved by means of
0.5 in diameter blind fasteners and adhesive bonding. 31 A thin
polymer concrete overlay was applied on the FRP deck as the
wearing surface. This was achieved by first sandblasting and
cleaning the surface of the FRP deck followed by the application
of a urethane-based primer using a broom. The latter was done
to improve the adhesion between the overlay and the deck. The

total thickness of the polymer

approximately 1 cm (3/8 in).

concrete

overlay

was

The Laurel Lick Bridge is a short-span bridge

located off county route 26/6 in Lewis County, WV. The original
structure consisted of a of timber deck on steel stringers. At the
time of replacement, this structure was in critical condition.
305x205x12.7 mm (12x12x0.5 in) beams are used to support the
new FRP deck (Figure 3.9).

Figure 3.9. Laurel Lick Bridge (Lopez-Anido et al. 1998)

The Wickwire Run Bridge is located off US Route 119 in Taylor
County, WV. The bridge is 9.14 m (30-ft) long by 6.60 m (21.7ft) wide. Four longitudinal galvanized steel beams, spaced 1.83
m (6-ft) apart, support the modular FRP deck (Figure 3.10).

From the field tests, the authors have concluded that

the performance of developed decks is excellent, especially
when they are used as a replacement for concrete decks. This is
because FRPC decks are much lighter than decks built using
traditional materials (for example: FRPC 32 deck weighs 98
Kg/m2 while concrete/steel decks weigh 540 Kg/m2 ). They also
mention that they expect that the costs associated with FRPC
applications will decrease significantly as this technology
becomes more widely used.

Figure 3.10. Wickwire Run Bridge, WV (GangaRao et al. 1999)

In the work by Lopez-Anido et al. (1998) laboratory testing was

performed to establish the performance of the developed deck
modules. Two specimens were tested: one to failure and the
other to fatigue. The first specimen, tested to failure, was a
2.743 m (108 in) long by 0.914 m (36in) wide FRP deck
specimen. A patch load simulating a wheel load was applied to
the specimen. The load level at failure was 577 kN (129.7 kip).
The observed failure mode was
interlaminar shear in the pultruded material in the proximity of
the bonded connection. In particular, punching damage on the
deck was not observed in these experiments. The second deck
specimen, tested under fatigue loading, was subjected to 2
million cycle loads from 9 kN (2 kip) to 156 kN (35kip).
Inspection of the tested specimen did not reveal any 33 crack
propagation due to fatigue. After the application of the cyclic
loads the FRP deck was tested to failure. It was found that the
failure load decreased only by about 4% when compared to the
specimen with no load history. However, the midspan deflection
increased by about 10%. It was found that developed FRP decks
have a very high strength capacity than concrete decks with only
20% of the weight. However, they are more flexible when
compared to concrete decks.
Thus, in general, serviceability (deflection) requirements
control the design of FRP composite decks. This is because
excessive deformation can cause premature deterioration of the
wearing surface as well as it can affect the performance of the

fasteners. In the work by GangaRao et al. (1999), both

components and deck modules were tested in the laboratory.
Three-point static bending tests were
conducted on both hexagonal and doubletrapezoid component
specimens with three different spans: 60, 84, and 108 inches.
Both a 20x10 patch load, intended to simulate a wheel load of
an AASHTO standard truck, and a strip load using a 6-inch wide
plate intended to cause the maximum bending strains, were
applied to the specimens. The deck module testing included
static and fatigue bending tests on 3-ft long simply supported
deck modules.
Only the patch load was used in
the fatigue tests. For the fatigue tests, a sinusoidal load ranging
from 2 to 35 kips at a rate of 3 cycles per second was applied at
a maximum of 2 million cycles. From the static bending tests, it
was found that the flexural rigidity of an FRP composite
component is about one half of the flexural rigidity of an
uncracked concrete component, and about 3.7 times of the
flexural rigidity of a cracked concrete component. For the
fatigue tests two FRP deck specimens were used. One was
subjected to a prior load history (two million fatigue cycles),
while the other had no load history.
From the results of the static
failure tests, it was found that both specimens experienced about
the same maximum 34 deflection and failure load. Thus, the
prior load history was found to have no significant effect on the
strength and stiffness of FRP deck. The authors conclude once

again that FRP has an excellent energy absorbing capability.

The ultimate load capacity of the tested FRP composite deck
specimens exceeded the AASHTO-HS25 load by an excess of
about 100 kips. The failure mode of the double-trapezoid
component was such that it failed at the junction of web and
flange at the applied load location. The failure mechanism
consisted initially of web buckling at the applied load location
and propagated on both sides of the load patch. For the doubletrapezoid component, failure occurred at the web-flange
junction.
This was attributed to the less than
satisfactory fiber wet-out and high stress concentration zones
near the re-entrant angles of these specimens. In the work by
GangaRao and Craigo (1999), a third demonstration bridge
located in Russell, KA is discussed. In this application, a
Cellular deck system using multi-cellular panels made of E-glass
and polyester resin connected by wide-flange H-sections was
used (Figure 3.11). This type of system was also successfully
used in the construction of a building in Weston, WV, in 1994.
In this research, it is found that this deck system is economical.
In particular, they have found that these cellular deck systems
are ideal for pedestrian bridges.

Figure 3.11. Layout of the Cellular deck panel (GangaRao and

Craigo 1999)
Lopez-Anido et al. (1999)
In the study by Lopez-Anido et
al. (1999) high-temperature fatigue tests were performed on an
FRP-concrete bridge deck. In this application, the deck was
made of FRP pultruded panels, which served as stay-in-place
formwork as well as reinforcement for concrete (Figure 3.12).

Figure 3.12. FRP-Concrete deck and test set-up (Lopez-Anido

et al. 1999)

The pultruded panels were 457 mm (18 in) wide. They were
stiffened using two tubular cells of 76 mm (3 in) in height. Eglass fibers in a polyester-vinyl ester resin blend were used to
form the composite material. Finishing of the material was

achieved by means of epoxy coating and sand spraying the top

surface. The total depth of the deck was 203 mm (8in), including
the FRP panels. E-glass bi-directional top reinforcement was
used to improve the bond with concrete. Since no specification
is currently in place for fatigue performance evaluation of FRP
concrete slabs, one million load cycles at a controlled hightemperature (49C) were used in this 36 work. More
specifically, the FRP panel specimens were subjected to a
maximum and minimum fatigue load cycles of 92.5 kN and 8.9
kN, respectively, at a frequency of 4 Hz. The test specimen
consisted of a two-span continuously supported FRP-concrete
deck panel with a girder spacing of 2.74 m (108 in). The load
was applied to simulate an AASHTO HS20-44 wheel load. The
main goal of the tests was to monitor the structural degradation
during the fatigue tests, since failure was not expected to occur
during the applied load cycles. This is because accumulation of
damage due to cyclic loading is usually reflected in loss of
stiffness of the FRPconcrete deck material. Therefore, this work
adopts the stiffness degradation as the fatigue performance
criterion for FRP-concrete decks. The main findings from the
performed tests are:
A 13% decrease in stiffness was observed for an increase in
temperature from 19C to 49C.
For high-temperatures, the stiffness decreased by
approximately 5 to 6 % within the first 100,000 load cycles
and remained almost unchanged after that and up to one
million cycles.

Foster et al. (2000)

In the work by Foster et al. (2000), a 10-m-long by 7.3-m-wide
(33x24 ft) GFRP composite highway bridge installed in Butler
County, Ohio is described. Both the support beams and the deck
were built using composite materials. In order to keep the cost of
the application down, the composite bridge components (deck
and the support beams) were made of E-glass fibers in an
isopolyester resin matrix. Glass fibers cost about 10% less than
carbon fibers (often used in the aerospace industry) and
isopolyester resins cost less than structural epoxy resin. This
bridge, referred to as Tech 21, was open to traffic in July
1997. Figure 3.13 shows the developed FRP composite beam
and Figure 3.14 (a) and (b) shows the assemblage of support
beams.

Figure 3.13. FRP support beam (Foster et al. 2000)

In this application, asphalt was used for the
wearing surface. Even though, the weight of the asphalt layer
was larger than that of the deck, the AASHTO HS-20 load
requirement was satisfied. It should be noted that most FRP
bridge decks developed in the U.S. have adopted a polymer
concrete surface, since it is lighter in weight than asphalt.
However, the authors justify their choice by the fact that
highway crews are more accustomed to using asphalt, especially
for resurfacing.
In this work, it is reported that the total
installation time of the FRP composite bridge was six weeks.
The authors claim that the erection of an equivalent reinforced
concrete bridge would take ten weeks. In addition, the weight of
the FRP bridge is 10.5 tons, while an equivalent RC bridge
would weigh 89 tons.

(a)

(b)
Figure 3.14. (a) Assembled FRP beams; (b) Underside view of
the assembled FRP beams (Foster et al. 2000)

The bridge was subjected to live loads

slightly lower than the required by AASHTO HS-20. Figure
3.15 illustrates this test, where the loading was applied by means
of two heavy-duty trucks fully loaded with sand. The
measurements were obtained with 28 steel strain transducers
externally installed for six different axle locations. The
maximum load in the test series was a static load of 64.6 metric
tons (142,600 lb), which produced a maximum stress of 13.8

MPa (2,000 lb/in2 ). In addition, the deflection curve obtained

was comparable to that of a comparable steel span.

Figure 3.15. Live load test of the Tech 21 bridge (Foster et al.
2000)

The long-term performance of the FRP bridge is being

monitored using the twenty fiber optic sensors and 102
mechanical sensors that were embedded in the bridge. The
authors of this work expect that the findings from this research

will be used in the development of the new AASHTO composite

bridge standards.
Hayes et al. (2000)
The work by Hayes et al. (2000) studies the feasibility of
utilizing a composite bridge deck as a replacement for
deteriorated bridge decks or for new construction. More
specifically, quasistatic and fatigue were performed on a
prototype composite bridge deck section. In these tests, 40 the
flexural strength and stiffness were measured under a simulated
wheel load. In addition, the fatigue behavior and residual
strength were assessed after fatigue loading, and failure modes
from fatigue and static loadings were determined.
Twelve pultruded 102x102x6.35-mm-thick square
tubes sandwiched between two pultruded 9.53-mm-thick plates
formed the studied deck section (Figure 3.16). The dimensions
of the specimen are 4.27-m in length, 1.22-m in width, and 121mm in depth. The material of the plates and tubes was formed
by unidirectional and continuous strand mat glass fibers in an
isophthalic polyester resin. The tubes were connected using
studs and nuts, and epoxy adhesive, while the top plates were
fastened using epoxy adhesive. The prototype deck panel did not
include a wearing surface, since it was assumed that such a
surface would not significantly affect its structural response.

Steel girders (W16x40) parallel to the short side of

the deck were used to support the deck (Figure 3.19 (a). The
adopted girder spacing was 1.22-m, and the orientation of the
square tubes was transversal to the steel girders. The connection
between the deck and the girders was achieved with steel bolts,
which passed through holes drilled through the deck and top
flanges of the steel beams. Flat steel washers were used to
prevent the bolt head from bearing directly on the top composite
plate. A bearing pad was placed between the top flange of each
W16x40 and the deck. In order to provide transverse integrity
under bearing load, wood block inserts were place inside the
fiberglass tubes at the hole locations.

(a)

(b)

Figure 3.16. The studied FRP deck: (a) Side view; (b) End view
(Hayes et al. 2000)
The prototype deck panel was subjected to
three types of tests. The first one was a static service load test in
the middle span of the deck; the second was a static loading to
failure on the left end; and the third was a fatigue performance
and residual strength test (fatigue up to 3,000,000 cycles,
followed by static loading to failure on the right end of the
deck). A 508x305- mm loading patch was used to simulate a
wheel load on the top surface of the deck. Figure 3.17 shows the
failure modes of the deck panel: (a) shear failure of tubes around
load patch, (b) shear failure of the fiber bolt, and (c) top surface
cracking of the deck panel.

Failure of the deck occurred at 369 kN for

deck in as-received condition and 369 kN for decks subjected
previously to fatigue strength test. These loads are about four
times the design wheel load, which is 92.6 kN. Therefore, the
authors conclude that strength should not control design. The
midspan deflections of the deck panel under design wheel load
were 3.81, 3.81, and 4.32 mm for the service load test, the as
received test, and the post-fatigue strength test, respectively.
It was found that even though the proposed
deck system used off-the-shelf pultruded sections, it met the
necessary strength performance criteria. However, the
deflections were found to control the design when using the
AASHTO criterion for limits of live load deflection for steel,
aluminum, and concrete construction. This criterion was used
because no criterion is available for FRP composite
construction.
At the ultimate failure mode, shear failure
of the top and bottom deck flanges were observed. Even after
3,000,000 cycles of a fatigue load in excess of the design wheel
load, no change in stiffness or strength of the deck was
observed. Finally, it was found that the connections between
deck and girder did not negatively impact the performance of the
deck under static or fatigue loading.

(A)

(b)

(c)
Figure 3.17. (a) Shear failure of tubes around load patch, (b)
Shear failure of the fiber bolt, and (c) Top surface cracking of
the deck panel (Hayes et al. 2000)
Ohio DOT (2000)
The Ohio Department of Transportation
spearheaded a study to evaluate different types of FRP deck
panels to replace a deteriorated reinforced concrete deck of a
five-span continuous haunched steel plate girder bridge. This
demonstration project is known as the Salem Avenue Bridge.
This bridge carries six lanes of traffic and consists of twin
structures with a longitudinal joint and a 4-ft raised concrete
median at the center. The girder spacing is approximately 8 feet
9 inches.
The deck of the north bridge structure was
replaced by four different types of FRP deck systems

manufactured by the following four manufacturers: Creative

Pultrusions (CP), Composite Deck Solutions (CDS), Hardcore
Composites (HC), and Infrastructure Composites International
(ICI) in collaboration with Kansas Structural Composites
(KSCI). The CDS system (Figure 3.18(a)) is the most similar to
conventional reinforced concrete decks. This system uses FRP
stay-in-place forms to support the concrete deck and serve as
bottom reinforcement, and GFRP bars for the top reinforcement.
The CP deck system is formed by bonding interlocking
pultruded FRP tubes that are installed in the direction
perpendicular to the girders (Figure 3.18(b)). The HC and the
ICI deck systems are similar. Both of these panels consist of a
lightweight FRP core sandwiched by high strength FRP skins. In
the HP system, the core consists of foam blocks wrapped with
fiber cloth (Figure 3.18(c)). The ICI panels core is made of
corrugated glass fiber reinforced sheets (Figure 3.18(d)). On all
three FRP deck panels (CP, HC, and ICI) a 3/8-inch-thick
polymer wearing surface manufactured and installed by PolyCarb, Inc., was applied. Prior to the application of this wearing
surface, the decks surfaces were lightly sandblasted.

(A)

(b)

(c)

(d)
Figure 3.18. Different types of FRP bridge decks (a) Composite
Deck Solutions, (b) Creative Putrusions, (c) Hardcore
Composites, and (d) Infrastructure Composites International

This demonstration application was evaluated

by a third party evaluation team. This team was charged with the
identification of potential maintenance and serviceability
problems in the application. A number of potential problems
were identified by the team:
In both the HC and ICI deck panels, both delamination and
debonding in panel skins were detected visually and via
nondestructive testing. The evaluation team recommended
that this issue be addressed by the manufacturers.
Some of the CP, HP and ICI deck panels lift off the haunch
as much as 1/16 in. Therefore, the connections between
girder and deck may be inappropriate. This problem was
not anticipated and therefore not used as a criterion by
ODOT or the manufacturers. The evaluation team
recommended that manufacturers together with ODOT to
devise uniform bearing.
The wearing surface cracked above the field joints of the
CP, HP and ICI deck panels, which indicates that these
joints are not working properly. This indicates that the
Poly-Carbs wearing surface was not flexible enough to
allow for this movement. During the evaluation teams
investigation, the cracks were repaired with FRP fabric
reinforcement, which seem to have solved the problem.
Hairline cracking was observed on the surface of the CDS
deck. The cover was 1/2 to 3/4 inch less than the
recommended 2 inches. The concrete deck was sealed with

highmolecular-weight methcrylate (HMWM). However,

the team recommended that future designs consider the
elastic modulus of the GFRP bars in the determination of
the amount of shrinkage.
Joint between different deck systems did not work
properly. This was caused because the different decks had
different stiffness. The displacement differentials measured
ranged from 47 1/64 to 1/8 inch. The evaluation team
recommended that diaphragms be developed to provide
support for these joints.
Water intrusion was detected in the HCI panel and water
retention was observed in the ICI panel. Potential water
entry points include: anchor holes, which were open for
more than a month, face plate removal from CPI panels, or
holes drilled for screw attachment of conduits within
concrete sidewalk. Drilling of drain holes in the underside
of the panels were recommended by the evaluation team.
While a fire occurred alongside the HCI deck panel, no
obvious structural damage seems to have occurred. The
evaluation team recommended periodic inspection and
monitoring.

4.3. FRP Deck Manufacturers

A number of composite deck panel manufacturers, which were
originally dedicated to other industries such as the automotive
and aerospace industries, have been alternatively refocusing
their scope to the civil engineering industry. As mentioned in
Chapter 2, the manufacturers that have participated in most of
the developed field applications are members of the Market
Development Alliance of the FRP Composites Industry (MDA).
Each FRP deck manufacturer has a demonstrated system that is
applicable to a target application. The manufacturer members of
MDA of FRP deck panels are provided in this Section.

3TEX, Inc
This manufacturer is located in Cary, North Carolina. While
3TEX has been involved in areas of application such as the
automotive, defense, recreational, etc., it has recently begun to
manufacture low-profile composite bridge decks and pedestrian
bridges (girder spacing ranging 48 from 2 to 3 ft). Their system,
referred to as TYCOR, is composed by a foam core reinforced
in the Z-direction sandwiched by fiberglass fabric skins (Figure
3.19). This system is intended as a competitor to conventional
corrugated steel decks. This manufacturer has completed one
application in Montgomery County, Ohio, and is currently
developing a second application WPAFB, Ohio.

Figure 3.19. TYCOR bridge deck panels

Creative Pultrusions, Inc.
This manufacturer operates in two locations: Alum Bank,
Pennsylvania and Roswell, New Mexico. Their products are
manufactured using the pultrusion process. Their bridge deck
panel, referred to as Superdeck, is formed the pultrusion and
bonding of a double trapezoid and a hexagonal section to form a
bridge deck module (Figure 3.20). This deck is 20% lighter than
reinforced concrete, but the factor of safety is 6-to7 over the

design load. These deck panels are designed to comply with the
AASHTO HS25 requirements. Among the applications
developed by this manufacturer are the following bridges in
Ohio: the Laurel Lick Bridge, the Wickwire 49 Run Bridge, the
Shawnee Creek Bridge, and part of the Salem Avenue Bridge.
Another bridge in their inventory is the bridge on Laurel Run
Road in Pennsylvania

Figure 3.20. Superdeck bridge deck panels

Hardcore Composites
This company is located in New Castle, Delaware. Hardcore
composites has served mainly the marine infrastructure industry.

In 1995 the manufactured their first FRP bridge deck, which was
installed in Delaware. This manufacturer uses the Vacuum
Assisted Resin Transfer Molding (VARTM) process to
manufacture their bridge deck panels, which consist of a
honeycomb structural core (to transfer shear) sandwiched by
FRP face-skins (to provide flexural stiffness) (Figure 3.21). The
VARTM process allows for the development of monolithic
structures, and for the tailoring of the face-skins. Their decks
can be designed to satisfy AASHTO HS25 and the L/800
deflection criterion. Hardcore composites is designing and
fabricating the bridges of Project 100 (Ohio state initiative). The
following are the bridges manufactured by this company, which
are in service: Magazine Ditch Bridge (Delaware), 50
Washington School House Road Bridge (Maryland), Muddy
Run Bridge (Delaware), Bennetts Bridge (New York), Wilsons
Bridge (Pennsylvania), Greenbranch Trail Bridge (Delaware),
Mill Creek Bridge (Delaware), a bridge in Elmira (New York),
and part of the Salem Avenue Bridge (Ohio).

Figure 3.21. Hardcores bridge deck panels

Kansas Structural Composites, Inc.:
This company was formed in 1995 and it is located in Russell,
Kansas. The area of concentration of KSCI, Inc. is the
application of FRP bridge deck panels to deteriorating highway
infrastructure. Their first application in collaboration with
Infrastructure Composites, International (ICI) from San Diego,
California, is the No-Name Creek Bridge in Kansas, was

developed in 1996. Their deck system consists of a fiber

reinforced polymer honeycomb 51 (FRPH) core sandwiched by
composite panels (Figure 3.22). This companys bridge deck
meets the AASHTO HS25 standard requirements. Other
applications developed by KSCI are the two FRP bridge decks
installed on Kansas State Highway 126.

Figure 3.22. Cross-section of FRPH deck panel

Martin Marietta Composites, Inc.:
This company is a subsidiary of Martin Marietta Materials
(MMM), which is a major supplier of aggregates in the U.S.
Martin Marietta Composites, Inc. (MMC) was established to
pursue the application of advanced composites to highway
infrastructure. Their bridge deck panel is the DuraSpan

(Figure 3.23), which has been designed to satisfy stiffness

requirements. Their main goal is to minimize the amount of
material and still satisfy AASHTO HS25 deflection requirement.
DuraSpans geometry uses stitched fabrics with engineered
orientations and it is fabricated using pultrusion. MMCs
completed and active projects include: road test panels
(University of California, San Diego), DARPA Task 16 Bridge
(Ohio), INEEL Bridge 52 (Idaho), Ohios First All-Composite
Bridge (Ohio), Kings Stormwater Channel Bridge (California),
Route 418 Truss Bridge over Schroon River (New York), and
Schulyer Heim Lift Bridge (California).

Figure 3.23. DuraSpan deck panel

CHAPTER 5
FRP FOR NEW HIGHWAY BRIDGES
Due to severe environmental conditions and the use of salt for
de-icing roads in Canada, the use ofFRPs for bridge girders,
deck slabs and barrier walls has been used for several new
bridges. These projects were completed through networking and
collaboration between ISIS Canada and various provincial and
municipal highway departments across the country. Due to a
lack of codes and standards for the use of FRP for bridges and
structures, ISIS Canada has undertaken the challenge of
launching a comprehensive research program for each field
application using a new design approach to examine the various
aspects of the strength requirements, severiceability
performance and the durability of these materials. In the case of
the Taylor Bridge in Headingley, Manitoba, the research
included an experimental program conducted at the University
of Manitoba using full-scale models to examine behaviour and
provide design guidelines for the construction details used in the
bridge. The following section reviews the design and
construction of three bridges in Manitoba, Quebec and Alberta
which have been completed using a variety of FRPs in terms of
the type of fibre and reinforcement and which, in some cases,
have been combined with the new design concept of the steelfree deck.

5.1 TAYLOR BRIDGE

The Taylor Bridge is located on Provincial Road No. 334 over
the Assiniboine River in the parish of Headingley, Manitoba.
The total length of the bridge is 165.1 metres (540 feet), divided
into five equal spans. Each span consists of eight I-shaped
precast prestressed concrete girders as shown in Figures l and 2.

Figure : 1 Layout of the bridge

Figure 2: Cross-section of the girders at the mid-span and end

block.
Two different types of carbon FRP
reinforcements were used. Carbon fibre composite cables
(CFCC) of 15.2 mm diameter, produced by Tokyo Rope Mfg.
Co., Ltd. of Japan, were used to pretension two girders while the
other two girders were pretensioned using 10 mm diameter
indented Leadline bars, produced by Mitsubishi Chemical
Corporation of Japan, as shown in Figure 3. Two of the four
girders were reinforced for shear using 15.2 mmdiameter CFCC
stirrups and lOx 5 mm Leadline bars of rectangular cross
section. The other two beams were reinforced for shear using 15
mm diameter epoxy coated steel rebars.

Figure 3: Details or reinforcement of girder prestressed by

CFCC.
A two-lane width of the deck slab was reinforced by 10 mm
diameter indented Leadline bars similar to the reinforcement
used for prestressing, as shown in Figure 4. C-BAR (GFRP)
reinforcement of 15 mm diameter, produced by Marshall
Industries Composites Inc. of the United States, was used to
reinforce a portion of the Jersey-type barrier wall. Doubleheaded
stainless steel tension bars of 19 mm diameter were used for the

connection between the barrier wall and the deck slab. Material
properties of FRP reinforcement used in the bridge are shown in
Table 1.

Figure 4: Bridge deck of the Taylor Bridge reinforced by CFRP.

The anchorage system used for CFCC was the die cast wedge
type, where a 300 mm steel tube is attached to the cable by a
low point braze alloy. Steel wedges were used to clamp
individual cables to an anchorage head, similar to the
conventional method for prestressing cables. The 10mm
Leadline bars were anchored using 150 mm long steel wedges
and a steel anchor head. Aluminum tubes were placed between

the Leadline bars and the steel anchorage to reduce the

transverse stresses in the bars. Both anchorage systems were
supplied by the manufacturers of the reinforcement.
A typical bridge girder was
pretensioned using 12.7 mm steel strands with a cross sectional
area of 99.0 mm2 The ultimate tensile strength and modulus of
the steel were 1860 MPa and 190 GPa, respectively. Epoxycoated 15 mm steel rebars with a yield stress of 400 MPa were
used as stirrups for the other girders. The specified concrete
strength for the bridge girders was 30 MPa at release and 40
MPa after 28 days. The specified concrete strength of the deck
slab and the barrier wall is 30 MPa.
5.1.1 Monitoring System
A total of 63 single and two multiplexed fibre optic sensors were
installed on the carbon FRP, GFRP and steel reinforcement to
monitor the bridge from a central monitoring station remote
from the bridge. The 65 sensors were installed on the following
bridge components:
i) The girders reinforced by CFRP;
ii) Selective girders reinforced by steel;
iii) Deck slab portion reinforced by CFRP; and
iv) Barrier wall portion reinforced by GFRP.

In addition, 20 thermocouples were used at different location of

the bridge to compensate for the temperature change. A 32channel fibre optic grating strain indicator (FLS 3500R), shown
in Figure 5, is used for strain measurements. The system is
connected to a computer to download the strain readings using a
telephone line. A general description for the fibre optic sensing
technique used for Taylor Bridge is given in Figure 6.

Figure 5. Fibre optic multiplexing and recording units.

Figure 6: Optical sensing technology.

5.2 Crowchild Bridge

The Crowchild Bridge is located in Calgary, Alberta. The
bridge's superstructure and the prestressed concrete box girders
were demolished in May 1997 and replaced by a new
superstructure system using steel girders, a steel-free concrete
deck for the intermediate deck spans and glass FRP for the
cantilever sidewalks. C-Bar (GFRP) reinforcements of 15 mm
diameter, produced by Marshall Industries Composites Inc. of
the United States, was used to reinforce the two cantilever
sidewalks including the top reinforcements of the adjacent slabs
as shown in Figure 7. The composite action between the steelfree deck and the steel girder was achieved by using Nelson
studs welded to the top flange of the steel girder and the steel
straps required for the arch action mechanism in the steel-free
deck as shown in Figure 8. The bridge was instrumented with 81

strain gauges, 19 embedded gauges, five thermisters, three smart

glass rebars and two fibre optic gauges.
A wireless data acquisition system which consists of a 24-bit
data acquisition chip, a radio transmitter and trigger is currently
being developed to remotely monitor the bridge from the office
of an engineer. A static truck load test and an ambient vibration
test have been performed on the bridge and the preliminary
results show extremely encouraging results.

Figure 7: Use of C-Bars in combination with the steel-free deck

system in Crowchild Bridge.

Figure 8: Steel straps for the arch action mechanism of the steelfree deck.

5.3 Joffre Bridge :

The Joffre Bridge, located over the Saint-Francois River in
Sherbrooke, consists of five spans (26 to 35 meters). The
superstructure is supported by steel girders spaced 3.7 meters in
a composite action of the deck slab. Construction started in
August 1997 and the bridge opened to traffic on December 6,
1997.
Carbon FRP NEFMAC grids were used to
reinforce the deck slabs. The NEFMAC was produced by
Autocon Composites Inc. of North York, Ontario. Some of the
CFRP NEFMAC grid was instrumented using structurally-

integrated fibre optic sensors during the manufacturing process,

as shown in Figure 9. The grids were used to reinforce a portion
of the deck slab of the bridge side-by-side with conventional
steel reinforcements to examine the effectiveness of this material
in increasing the service life of the bridge, as shown in Figure
10. The bridge is extensively instrumented using fibre optic
sensors, vibrating wire strain sensors and electric resistance
strain. All sensors are connected to telephone lines for
continuous monitoring of the bridge from remote locations.

Figure 9: Structurally-integratedfibre optic sensors in the

NEFMAC.

Figure 10: A portion of the deck slab of the Joffre Bridge

reinforced by carbon NEFMAC.

CHAPTER 6
STRENGTHENING WITH FRP
More than 40 percent of the bridges operating in Canada were
built over 30 years ago and most are in urgent need of
replacement or rehabilitation. Many of the structural deficiencies
are due to deterioration of the concrete as a result of corrosion of
the steel. Other bridges have become functionally obsolete due
to an increase in service loads and traffic volumes which exceed
those for which they were designed. FRPs provide an excellent
solution to repairing and/or strengthening bridges. For bridge
piers, wrapping can significantly improve the strength and
ductility. FRPs are also used for strengthening the superstructure
of the bridge by strengthening the flexure and shear capacity of
the girders and slab. The following are selected demonstration
projects illustrating the use of FRP for strengthening bridges.

6.1 Champlain Bridge

In October 1996, the Jacques Cartier and Champlain Bridges
Inc. proceeded with the rehabilitation of a concrete pier of the
Champlain Bridge located in Montreal. The pier is 1.37 meters
in diameter and was repaired over a 4 meter length measured
from the base. The rehabilitation began with restoration of the
column surface using shotcrete. After curing, the surface was
prepared using a sand blast technique followed with a resin
application. Nine layers of glass fibre, type E (SEH51ITYFO S),

from Composite Retrofit International Inc. were used, see Figure

11. FRP was needed in this case to increase the confmement of
the concrete, as well as providing a protective surface to the
crumbled and cracked concrete surface.

Figure 11: Strengthening of the pier of the Champlain Bridge.

6.2 Maryland Bridge

Carbon FRP is planned to be used to strengthen the shear
capacity of the I-shaped concrete AASHTO girders of the

Maryland Bridge in Winnipeg, Manitoba. The bridge was built

27 years ago. Analysis of the precast prestressed concrete girders
indicates a deficiency in the shear capacity using the current
AASHTO code. A 1 :35 scale model of the bridge girder was
tested using three different types ofCFRP and six configuration
schemes, as shown in Figure 12. The diagonal configuration for
the carbon FRP was found to be the most effective configuration
in reducing the tensile force in the stirrup.

Figure 12: Proposed configuration ofCFRP sheets for Maryland

Bridge.

CHAPTER 7

CONCLUSION AND FUTURE OF FRP FOR

BRIDGE APPLICATIONS
In spite of the lack of codes and standards, several bridges have
been built around the world using carbon FRPs for prestressing
and/or reinforcing of the concrete structural girders, deck slab
and barrier wall. The design is based on a rational approach of
the material characteristics which was relatively simple due to
the linear behaviour of the FRPs to failure. FRPs were also used
to strengthen existing bridges due to the increased demand for
heavier truck loads. The strengthening was in the form of
wrapping columns to increase the strength and ductility as well
as to increase flexure and shear capacity of the girders. Most of
the field applications have been instrumented for continuous
monitoring to provide data related to the material's long-term
behaviour and to ensure safety of the bridge by monitoring their
performance under service loading conditions.
The high strength, high fatigue
resistance, lightweight, and corrosion resistance of
composites are highly desirable characteristics for bridge
applications.Currently,
these new materials are a direct
technology transfer from the aerospace industry, and they are far
more advanced than those required by civil applications. Most
of the advanced composite materials that are cured at high

temperature produce high quality components and possess

excellent characteristics.
In bridge applications, resins as
the binders for the fibre and adhesives for joints and
connections that can adequately cure at ambient temperature
and still offer comparable quality and characteristics are
more desirable and practical. More research is needed to
develop the most effective and durable resin formulations.
More efficient manufacturing and effective production
methods for large volume panels and higher modulus
materials are needed to make it more cost effective for
composites to compete in the civil infrastructure. At the present
time, the direct use of fibre composites from the aerospace
industry is not cost effective as compared to conventional
materials in bridge applications.
In the paper rehabilitation and
strengthening of three bridges by using FRP composites
were considered. The results were tested and found to be
satisfactory in all the cases. In ROB at JNPT static and
dynamic load tests were carried out and it was found that
the strains and accelerations were reduced substantially. In
the case of Mithi River Bridge at Mumbai Airport load tests
were carried out to test whether it will be able to take the load of
larger aircrafts and the results were positive. The bridge at
NH7, Bangalore showed substantial reduction in live cracks
and no distress is seen to present date.

These results have proven the

effectiveness of these FRP materials for usage in civil structures
especially in the case of bridges. The main concern with FRP
composites is long-term durability because the materials do
not have sufficient historical performance data in bridge
applications. There is a concern among bridge engineers for
the long-term integrity of bonded joints and components under
cyclic fatigue loading. There are concerns with improper curing
of the resins and moisture absorption and/or ultraviolet light
exposure of composites that may affect the strength and stiffness
of the structural system. Certain resin systems are found
ineffective in the presence of moisture. In the case of a
glass fibre composite, moisture absorption may affect the
resin and allow the alkali to degrade the fibres.
If the cost constraint is kept aside, the fiber
wrapping system has been proved to be a system which has
many added advantages over conventional strengthening
processes. It has been proved in laboratory as well in real civil
projects that this system is effective and is useful in real life.
As the economy is moving ahead and infrastructure
development is catching its pace, demand for fiber
reinforced polymer in civil construction is slowly increasing
and becoming acceptable.

CHAPTER 8

REFERENCES
References
1) Radomski W., Bridge Rehabilitation, Imperial College Press
2) ACI Committee 440. State-of-the-Art Report on FRP for
Concrete Structures.
ACI 440R-96, Manual of Concrete
Practice, ACI, Farmington Hills, MI, 1996, 68 pp.
3) CALTRANS
Division
of
Structures,
1996,
Prequalification
Requirements
for
Alternative Column
Casings for Seismic Retrofit (Composites), Section 10.1,
California Department of Transportation.
4) Hawkins, G. F.; Steckel, G. L.; Bauer, J. L.; and Sultan, M.,
1998, Qualification of Composites for Seismic Retrofit of
Bridge Columns, Proceedings of the First International
Conference on Durability of Composites for Construction,
Aug., Sherbrooke, Quebec, pp. 25-36.
5) Meier H., Clenin R., Basler M., bridge strengthening with
advanced composite systems, Sika Services AG, Zrich,
Switzerland
6) http://en.wikipedia.org/wiki/Nhava_Sheva
7) Singh A.K., Rai G., Jangid R.S. , Strengthening of runway
bridge at mumbai airport using frp.