Beruflich Dokumente
Kultur Dokumente
5/12/06
8:39 am
Page a
Advances in
Structural Engineering
An International Journal
ISSN 1369-4332
Editor-in-Chief
Professor J. G. Teng
The Hong Kong Polytechnic University
A Multi-Science Publication
5/12/06
8:39 am
Page b
5/12/06
8:39 am
Page a
Advances in
Structural Engineering
An International Journal
CONTENTS
Page
Guest Editorial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
719
Technical Papers
The Importance of Bonding-a Historic Overview and Future Possibilities
B. Tljsten . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 721
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic
Debonding Mechanisms
D. J. Oehlers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 737
Analysis of the Load-Deformation Behaviour and Debonding for FRPStrengthened Concrete Structures
K. W. Neale, U. A. Ebead, H. M. Abdel Baky, W. E. Elsayed and A. Godat 751
Fracture Mechanics Approach to Geometrically Nonlinear Debonding
Problems in RC Beams Strengthened with Composite Materials
Oded Rabinovitch and Yeoshua Frostig . . . . . . . . . . . . . . . . . . . . . . 765
Deterioration of FRP-to-Concrete Bond under Failure Loading
Kent A. Harries, John Aidoo, Andrew Zorn and
Joshua Quattlebaum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
779
805
833
5/12/06
8:39 am
Page b
ASE 9-6-00_Editorial
4/12/06
5:12 pm
Page 719
Guest Editorial
This special issue contains extended and revised
versions of 8 papers selected from those presented at the
International Symposium on Bond Behaviour of FRP in
Structures (BBFS 2005), which was held on 79
December 2005 in Hong Kong, China. The Symposium
was organised by the Working Group on Bond between
FRP and Concrete (the Bond WG) of the International
Institute for FRP in Construction (IIFC), in conjunction
with the Department of Civil and Structural Engineering,
The Hong Kong Polytechnic University.
The IIFC Bond WG was established in November
2004 with the objectives of:
719
ASE 9-6-00_Editorial
4/12/06
5:12 pm
Page 720
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 721
Abstract: An adhesive is a substance that has the capability to hold materials together.
The manner by which adhesives are able to serve this function is due to a surface
attachment that is resistant to separation. A bond occurs when the adhesive molecules
adsorb onto a solid surface and chemically react with it. Adhesives have a very large
importance for most industries, both for structural and for non-structural use. In this
paper, an overview of the use of adhesives in different industries is presented with
focus on the building industry and on structural use. Not surprisingly, the use of
structural bonding has increased considerably during the last decade. The main reasons
for this are probably a combination of lower cost, more need of lightweight structures,
development of more durable adhesives and improved structural adhesive perf1ormance.
Additionally, adhesives often provide better load transfer than welded structures.
In this paper, an overview of the use of adhesive in the service of man is presented.
The overview is brief, but nonetheless gives an idea of the possibilities of adhering
materials by bonding. The paper briefly discusses the use of adhesives in the
aerospace, automotive and marine industries, but focuses on the construction industry.
Finally some ideas about sticking things together tomorrow are presented.
1. HISTORICAL OVERVIEW
Natural adhesives have been known since pre-historic
times. Two stones dating back at least 36,000 years
have traces of a sticky black substance that was once
used to attach them to a handle. In a chemical analysis
the glue-like material was identified as bitumen. The
oldest discovery of the use of adhesives by humans to
date was made in the Nahal Hemar Cave to the
northwest of Mount Sedom in Israel. When this cave
was excavated in 1983, many of the artefacts unearthed
were found to carry residues of collagen based material
believed to be derived from animal skins. This adhesive
has been carbon dated to over 8,000 years ago (Walker
1998; Bar-Yosef and Schick 1989). Archaeologists
studying burial sites of prehistoric tribes dating to about
4000 B.C. found foodstuffs buried with the deceased in
broken pottery vessels that had been repaired with sticky
*Corresponding author. Email address: bt@byg.dtu.dk; Fax: +45 45 88 32 82; Tel: +45 45 24 17 73.
721
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 722
history shows the use of glue fell into disuse until about
15001700 A.D. when adhesives were used in the
building of furniture. Some of the greatest furniture and
cabinet makers of all times used adhesives in their
productsnames still recognizable today like Chippendale
and Duncan Phyfe. Another notable name in history
who may owe his notoriety, at least in part to adhesives
in Genghis Khan. Around the year 1000 A.D., Genghis
Khan overcame all attackers because of the exceptional
power and range of the weaponry his men carried. Bows
were made from laminated lemon wood and bullhorn
bonded with an adhesive whose formulation has been
lost to antiquity.
By the 17th century, scientists were beginning to give
consideration to the nature of adhesion itself. Francis
Bacon in his Novum Organum suggested that there is in
all bodies a tendency to avoid breaking up, (Fay 2005).
A little later (Newton 1717) conjectured that, There are
agents in nature able to make the particles of bodies
stick together by very strong attractions. And it is the
business of experimental philosophy to find them out.
A general renaissance in the use of adhesive bonding
began around this time and is clearly demonstrated by
the changing construction methods used for furniture.
During the early days in the history of adhesives, it is
likely that the materials were produced on very small
scale, possible in the kitchen of the individual users.
However, by around the 1700s, the production of
adhesives started to undergo transformation into a major
industry. During the 18th century, the technology of
animal and fish glues advanced. In the 19th century,
rubber and nitrocellulose based cements were introduced.
In about 1700, the widespread use of glue brought about
some rapid changes in the history of adhesives. The first
commercial glue factory was started in Holland to
manufacture animal glue from hides. About 1750, the
first glue patent was issued in Britain for fish glue.
Patents were then rapidly issued for adhesives using
natural rubber, animal bones, fish, starch and milk
protein (casein). By 1900, the U.S. had a number of
factories producing glue from the aforementioned bases.
Although enormous efforts had been made in the
manufacture of animal glue in the period up to
approximately 1925, much bigger developments were
taking place which were to have the most significant
effect on the history of adhesives in thousands of years
the development of synthetic polymers (Fay 2005).
The Industrial Revolution caused an explosion in
technical breakthroughs which resulted in new materials
becoming available for use in formulating adhesives.
The first plastic polymer to be synthesized was cellulose
nitrate, a thermoplastic material derived from the
cellulose of wood. Its first use was in the manufacture of
722
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 723
B. Tljsten
Vegetable
Mineral
Elastomeric
Thermoplastic
Thermosetting
Type
Gelatin
Casein
Albumen
Starch
Cellulose
Asphalt/bitumen
Natural rubber
Synthetic rubber
PVA
Polystyrene
Cyanoacrylates
Liquid acrylic
Formaldehyde (urea/phenol)
Unsaturated polyesters
Epoxy resins
Polyurethane
Source
Mammals, fish
Milk
Blood
Corn, potatoes, rice
Cellulose
Earths crust
Tree latex
Synthetic
Synthetic
Synthetic
Synthetic
Synthetic
Synthetic
Synthetic
Synthetic
Synthetic
Use
Can labels
Plywood, blockboard
Paper, packaging
Leather, wood, china
Road pavements
Carpet making (rare)
Tyre, fabrics, bookbinding
Wood and general
Model making
Plastics, metals, glass, rubber
Structural vehicle assembly
Chipboard and plywood
Glass fiber, resin mortars
Structurally, metal to metal, concrete
Semi-structural uses with plastics,
metals wood and sandwich panel construction
723
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 724
724
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 725
B. Tljsten
Density, (kg/m3)
Youngs modulus, (GPa)
Shear modulus, (GPa)
Poissons ratio
Tensile strength, (MPa)
Shear strength, (MPa)
Compressive strength, (MPa)
Tensile strain at break, (%)
Approximate fracture energy, (Jm2)
Coefficient of thermal expansion, (106/C)
Water absorption: 7 days 25C, (% w/w)
Glass transition temperature, (C)
11001700
0.520
0.28
0.30.4
930
1030
55110
0.55
2001000
2535
0.13
4580
0.30.35
4060
2030
60120
0.53
4045
<0.2
>180
(a)
(b)
Tension
Compression
Shear
Cleavage
Peel
725
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 726
5. SURFACE TREATMENT
To achieve a satisfactory bond between the adherents,
it is usually necessary to carry out some form of surface
treatment. Many treatments are available, ranging from
a simple solvent wipe to the use of series of complex
chemical processes, (Brewis 2005). The method chosen
depends on the nature of the substrate, the conditions to
which the adhesive joint will be subjected, safety,
environmental factors and cost. Treatments may be
divided into physical and chemical methods. The
former includes solvent degreasing and grit blasting.
Physical treatments may remove cohesively weak
layers from a substrate and they may also modify
topography. Chemical treatments, which include the
flame treatment of plastics and anodising procedures
for metals, by definition cause chemical modification to
the surface involved.
It is relatively easy to achieve high initial joint
strengths. However, a satisfactory performance in
service will require careful selection of a pre-treatment.
Water in particular can cause a serious loss of joint
strength in service and pre-treatments vary markedly in
their ability to provide the necessary durability of the
joints involved, (Brewis 2005).
Primers, usually in the form of thin organic coatings,
are often used as an addition or alternative to pretreatments. Primers can provide several advantages. They
usually have much lower viscosities than adhesives and
can therefore achieve greater contact with the substrate.
They can have greater interaction with the substrate and
adhesive. They can contain corrosion inhibitors; this can
be important for metals. Finally, they can protect a surface
until the bonding process is carried out.
To pre-treat metals physical and chemical primers,
often in combination, are used. Etching of aluminium
with chromic acid CAE has been found to give
enhanced performance compared to physical methods.
(Critchlow and Brewis 1996). Much research has also
been carried out on the pre-treatment of titanium alloys.
Considerable success was achieved with an alkaline
peroxide etch. For steel grit blasting, sometimes in
conjunction with a silane, is often used to enhance the
bond properties
To achieve high joint strengths for inorganic
materials, such as glasses, ceramics and concrete, the
outermost layer often needs to be removed. For glass
this weak layer is represented by organic grease and for
concrete by the cement laitance. For glass a solvent
could be used whereas for concrete the weak surface
layer is removed by grit blasting.
Plastics and elastomers have much lower surface
energies than for example glass or metals. The
interaction between polymers and adhesives is therefore
726
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 727
B. Tljsten
Bonding
Bonding
Bonding
Figure 3. The Swedish Military Aircraft Jas Gripen, (a) Composite Parts, (b) Bonded Parts (SICOMP Workshop, 2004)
Epoxy
Toughened epoxy
Aerospace
40
Formula 1
30
Toughened acrylic
Weld bonding for cars
20
10
Flexible polyurethane
Windscreen
200
300
400
500
Strain (%)
Figure 4. Shear versus strain for different bonded parts in the automotive industry
7. AUTOMOTIVE
Adhesives have been employed in the automotive
industry since its beginnings, with the use of natural
resins to bond wood and fabric bodies (Mays and
Hutchinson 1992). Recent developments in synthetic
resin technology have resulted in a very wide range of
adhesive materials being available to the design
engineer. In Figure 4, a shear strength versus strain
diagram for various adhesive types used in the
automotive industry is shown. From this diagram it is
clear that bonding is widely used in the automotive
industry both for structural- and non-structural parts.
Adhesives in an automobile body are used to avoid
corrosion, to achieve higher body stiffness and to
enhance resistance to fatigue and crashes. To avoid
corrosion in flanges and hem flanges adhesives have to
fill the gaps to avoid the penetration of water. If there is
727
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 728
Stahl/steel
Aluminium
Magnesium
Kunststoff/plastics
Figure 5. Different adhesively bonded automotive parts for the Mercedes CL-Coupe
728
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 729
B. Tljsten
729
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 730
730
9.3. New-Built
If we discount glulam beams, structural bonding is not
very common in newly-built structures. Adhesive is often
used for filling voids or gaps and to fasten secondary,
non-load carrying elements. Adhesives are also used to
fasten bridge bearing and expansion joint nosing. On steel
decks, skid resistant surfacing materials in the form of
small gravel-filled epoxy or polyester resins are often
used. There are also examples where epoxy resin systems
have been used to bond precast concrete slab units
directly to a steel girder surface for a steel-concrete
composite bridge, (German Bonded Bridge 1979).
Applications of resins as true adhesives in new
constructions have been relatively scarce although
increasing use is being made of epoxies in the joints
between segmental precast, prestressed bridge construction
where the adhesive is being employed as a stress
distributing bonding/sealing layer between the precast
elements. This concept was pioneered in major structures
in France in the construction of the Choisy-le Roi bridge
over the River Seine near Paris in 1962 (Muller 1964).
This bridge was made up of precast concrete box units 2.5
m long. The units were match cast off-site in
continuous lengths; one unit being cast against the next
with a thin plastic film separator to ensure perfect fit and
that the thickness of the epoxy layer was kept to a
minimum. In comparison to mechanical or welded joints,
however, structural bonded joints are few. This might
change, however, with increased use of newly-built FRP
structures for building and civil purposes. By way of
example, a brief description of a FRP bridge structure
built in the UK 2002 having all structural joints made by
epoxy is presented. The ASSET (Advanced Structural
System for Tomorrows Infrastructure) bridge project,
founded by the European Commission, began in 1998
and was completed in the autumn of 2002. The project
mainly covered design, manufacture and prototype
construction of a fully polymer composite road bridge. A
European consortium designed a deck profile. The deck is
pultruded from a combination of roving, woven fabric
and mat. The deck was oriented transverse to the direction
of the traffic bonded together with epoxy. The deck
subassemblies, consisting of 5 or 6 pultruded profiles
bonded together off-site indoors, are then bonded to the
longitudinal composite girders. The mid-span crosssection of the bridge is shown in Figure 6. Below the deck
is a composite support structure consisting of four 300
mm, 11.0m long pultruded glass/polyester box beams
which are spaced 2 m apart and attached to the concrete
abutments. The box beams incorporated unidirectional
carbon fibre on the top and bottom for additional flexural
stiffness. The 10.0m long and 6.8m wide deck structure is
bonded to the box beams with epoxy. This prototype
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 731
B. Tljsten
731
Detail 2
Steel
L/1.00/V
parapet
(80 km/h)
775
400
upstand
500
verge
T
1:40
6800
1750
4 no. composite
longitudinal box beams
each comprising 20 thk.
CFRP top and bottom
flanges and 4 no. 240 240
GFRP S.H.S.
1:40
Asset profiles
spanning
transversely
30 mm at channel
92.5 mm at C
polymer concrete
2500
1750
1750
Special HB precast
KERB cut to provide
100 mm upstand and
bonded to deck
Concrete verge
2500
5000 carriageway
6800
400
upstand
775
6 mm HIM surfacing
500
verge
Proposed 150
gas main
Top_line
Bot_line
Level constant at
any transverse section
through bridge deck
225
5:11 pm
SECT
SEC
732
4/12/06
520
ASE 9-6-01_Taljesten
Page 732
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 733
B. Tljsten
Table 3. Material properties of FRP and adhesives used for the ASSET project
Product
Supplier
Property
Density, (kg/m3)
Viscosity, (mPa s)
Potlife, (min)
EL-modulus, (GPa)
ET-modulus, (GPa)
Tensile Strength, (MPa)
Compressive strength, (MPa)
Strain at failure, (%)
Profiles
Carbon Fibre
Resin
Adhesive
Fiberline A/S
Saertex
T700
1820
230
4800
2.1
NM AB
NM94/93C
1100
500
56160
3
52
100
2.1
BPE Systems AB
BPE 465/464
1498
Tix
90
7
31
103
27
9
1.2
733
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 734
(a)
(b)
Figure 9. Images of a flat-tailed house gecko and its seta. Top is a photograph of the gecko on a glass covered mirror used in the study.
At the bottom are scanning electron microscope (SEM) images of gecko setae showing the tree-like structure with
a magnification of (a) 900 times and (b) 8500 times (Sun W. et al., 2005)
734
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 735
B. Tljsten
735
ASE 9-6-01_Taljesten
4/12/06
5:11 pm
Page 736
736
Teng J.G., Smith S.T., Yao J. and Chen J.F. (2003). Intermediate
crack-induced debonding in RC beams and slabs, Construction
and Building Materials, Vol. 17, No. 6, pp 447462.
Tljsten, B. (1994). Plate Bonding: Strengthening of Existing Concrete
Structures with Epoxy Bonded Plates of Steel or Fibre Reinforced
Plastics, Doctoral thesis, Lule university of technology, Sweden.
Tljsten, B. (1997). Strengthening of beams by plate bonding,
Journal of Materials in Civil Engineering, ASCE, Vol. 9, No. 4,
pp 206212.
Tljsten, B. (2004). FRP Strengthening of Existing Concrete
Structures Design Guideline, ISBN 91-89580-03-6, Lule
University of Technology, Third ed., 228 pp.
Tremper, B. (1960) Repair of damaged concrete with epoxy resins,
Journal of the ACI, Vol. 57, pp 173182.
Vilnay, O. (1988). The analysis of reinforced concrete beams
strengthened by epoxy bonded steel plates, The International
Journal of Cement Composites and Lightweight Concrete, Vol.
10, No. 2, pp. 7378.
Wakeman, C.M., Stover, H.E. and Blye, E.N. (1962). Glue for
concrete repair, A.S.T.M, Bulletin, Vol. 2, No. 2, pp. 9397.
Walker, A.A. (1998). Oldest glue discovered, Archaeology Online
News, May 21, (http://www.archaeology.org).
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 737
1. INTRODUCTION
The main aim of this paper is to identify the generic
plate debonding mechanisms. Reinforced concrete (RC)
beams and slabs are subject to fundamental flexural and
shear deformations that are peculiar to reinforced
concrete members and which are generally well
understood by RC designers and researchers. It is these
fundamental reinforced concrete deformations that
induce fundamental internal forces which cause FRP
plates attached to the beam to debond. Hence, the first
step in understanding and identifying the generic and
fundamental debonding mechanisms is to understand
the fundamental behaviour of RC beams which is
explained for flexural, shear and axial deformations.
These generic reinforced concrete deformations, that is
flexural, shear and axial, control the plate debonding
mechanism. However, the capacity to resist the plate
debonding mechanism is controlled by the generic
material properties which are covered next. The generic
plate debonding mechanisms are then described in
terms of the reinforced concrete flexural, shear and axial
deformations.
2. GENERIC FLEXURAL
DEFORMATION MECHANISMS
A standard cross-sectional flexural analysis of an RC
beam is shown in Figure 1 (Oehlers and Seracino 2004).
For unplated RC beams with ductile reinforcing bars,
failure is always caused by concrete crushing at a strain
c. In contrast, a plated beam can also fail by plate
debonding at a strain db or plate fracture at a strain fr.
Hence, in plated beams three material strain failure
have to be considered, as shown in Figure 1(b), instead of
the one strain failure of c in unplated reinforced concrete
beams. In the analysis depicted in Figure 1, the strain
profile is then pivoted about the failure strain as shown in
Figure 1(c) where in this example plate debonding is
assumed to cause failure. The curvature of the strain
profile is then varied until from the stress distribution in
Figure 1(d), the force distribution in Figure 1(e) is in
longitudinal equilibrium. It is worth noting, that if plate
debonding strain or plate fracture strain is found to
precede the concrete crushing strain then the pivotal
strain must be changed and allowance must be made for
the concrete to remain at least partly elastic as shown in
737
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 738
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
(pivotal strains)
(stress)
(strain)
F (force)
F1 (y1)
F2 (y2)
kud
Plate stress
resultants
Tension
face plate
db
(a)
frac
(b)
F3 (y3)
F4 (y4)
db
Mplate = 0
Pplate = F4
Pivotal point
(c)
(d)
(e)
(f)
(stress)
(strain)
F (force)
Plate stress
resultants
F1
F2
Plastic zone
Elastic zone
Mplate
Neutral axis
Plate centroid
bar
F5 F4
Pplate
db
(b)
(a)
F3
Pivotal point
(c)
(d)
(e)
Point of
contraflexure
Position of
maximum moment
Vplate
Side plate
Point of
contraflexure
Continuous T-beam
Pplate
Pplate
738
3. GENERIC SHEAR
DEFORMATION MECHANISMS
The behaviour of RC beams under shear is much more
complex than that under flexure (Oehlers and Seracino
2004). What is known is that the forces in the shear
reinforcement, that is stirrups, are minimal and can be
ignored until a critical diagonal crack (CDC) forms. An
example of a CDC in a beam without stirrups is shown
in Figure 4, and it is this failure of a beam without
stirrups which is directly associated with the concrete
component of the shear capacity of the beam Vc.
The presence of stirrups tends to hide the importance
of the initial CDC as shown in Figure 5 where a series
of diagonal cracks can be seen. However, it should be
remembered that the formation of the first or initial
CDC in a beam with stirrups occurs at the same shear
load Vc as that of a beam without stirrups. Furthermore,
the formation of the first CDC often causes debonding
of longitudinal plates in beams with or without stirrups.
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 739
D. J. Oehlers
Applied load
Edge of
applied load
Flexural-shear cracks
Support
Diagonal
crack focal
point
A
B
Critical diagonal cracks
Support
Reinforcing
bar area s
Prestressing
tendon
Vc + Vc
(Pplate)tfp
(Pplate)sp
V
RC beam
Longitudinal side
739
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 740
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
Side plate
Steel
stirrup
U-jacket
(PAIC)1
(PAIC)3
Vs-2
Transverse
plate
Vs-1(PAIC)2
1.1
1
0.9
Stirrups
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Direction of IC interface
crack propagation
Intermediate crack
63%
55%
Total
ki
ke
kt
IC interface crack
NSM plates
PIC (IC)
0
0.5
1
1.5
2
2.5
3
Uniform crack width (mm)
3.5
740
B
P
Slip
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 741
D. J. Oehlers
Elastic
Micro-cracking
Debonding
5. DEBONDING MECHANISMS
IN EXISTING GUIDELINES
Most national guidelines, such as those listed in the
table in Figure 13 (Oehlers and Seracino 2004), deal
with tension face plated beams and it is felt that within
these guidelines there is already a general consensus as
to the major debonding mechanisms, even though each
guideline may use its own terminology as listed in the
table in Figure 13. All the guidelines acknowledge IC
(intermediate crack) debonding due to flexural cracks in
the beam and do distinguish between the behaviours of
IC debonding at the anchorage zone shown in column 1
and IC debonding due to a succession of flexural and
flexural/shear cracks as in columns 2 and 3; this form of
debonding is controlled in the guidelines through
limitations to the plate strains.
All the guidelines acknowledge CDC (critical
diagonal crack) debonding due to rigid body shear
deformations as in column 4 in Figure 13 and often
quantify this form of debonding by adjusting the
concrete component of the shear capacity Vc. Most
guidelines acknowledge PE (plate end) debonding due
to curvature, as in column 5, but may not give design
rules. For the European approach this is understandable,
Direction of debonding
crack propagation
Interface shear
PE debonding crack
Nplate
Pplate
Nplate
Plate strain = 0
= 0 at plate end
Mplate
741
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 742
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
Debonding mechanisms
p.o.c.
Uncracked section
p.o.c.
VC
Cracked section
Guidelines
[1]
AUST: IC
debonding
EUR mode 1:
peeling off in
uncracked
anchorage
BRIT:
debonding
failure
HK:
IC induced
debonding
USA: referenced.
[2]
AUST: IC
debonding
EUR mode 2:
peeling off at
flexural crack
BRIT:
debonding
failure
HK:
IC induced
debonding
USA: refer.
IC
[3]
VAy
[6]
AUST: IC
debonding
EUR mode 2:
peeling off at
flexural crack
BRIT:
debonding failure,
peeling failure
HK:
IC induced
debonding
USA: refer.
CDC
PE
[4]
[5]
AUST: CDC
debonding
EUR mode 3:
peeling off
caused by
shear cracks
EUR (mode 5):
plate-end
shear failure
BRIT: refer.
HK: plate end
interfacial
USA: refer.
AUST: PE
debonding
EUR (mode 6):
concrete rip-off
BRIT:
peeling failure
HK:
concrete cover
separation
BRIT: refer.
USA: not
recognised
742
L per ( EA) p
(1)
ASE 9-6-02_Oehlers
4/12/06
5:10 pm
Page 743
D. J. Oehlers
bp tp
2 mm
Concrete surface
Ap of EB plate
df
2 mm
bf
Failure plane EB plate
Ap of NSM plate
Lper
Concrete
element
Lper
bp
df
tp
2 mm
2 mm
bf
Partial-interaction
region
Full-interaction
region
Flexural
crack
(a) Plated
beam
D
(b) Slip
FA-B
(c) Resultant
bond force
FB-C
743
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 744
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
Herringbone IC debonding
744
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 745
D. J. Oehlers
PE debonding
Plate end
Plate end
Compression
face plate
PE debonding
745
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 746
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
PE debonding
IC debonding
Pplate
Pplate
B
2
Sequence of crack
propagation
C
3
E
4
Diagonal
crack (100 kN)
Critical diagonal
crack (188 kN)
CDC debonding crack
Tension face plate
CDC propagation
746
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 747
D. J. Oehlers
IC interface
cracking
Intermediate
crack
CDC debonding
Critical
diagonal
crack
747
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 748
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
Applied
load
CDC debonding
Support
IC debonding
Critical
diagonal crack
(PAIC)2
Inclined
plate
(PAIC)1
Internal
steel stirrup
CDC
748
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 749
D. J. Oehlers
ve or hogging region
ve or hogging region
EB
FIC
CDC
CDC
PE
FIC
VAy
NSM
Compression face
PE
CDC
SIC
VAy
PE
FIC
FIC
FIC
PE
VAy
PE
NSM
(d) Combinations
Angle section
CDC
Wrapped plate
PE
749
ASE 9-6-02_Oehlers
4/12/06
5:11 pm
Page 750
FRP Plates Adhesively Bonded to Reinforced Concrete Beams: Generic Debonding Mechanisms
APPENDIX: NOTATION
Ap
cross-sectional area of plate
AIC
axial intermediate crack
AUST Australia
bf
width of failure plane of perimeter length Lper
bp
width of plate
BRIT British
CDC critical diagonal crack
d
effective depth of beam
df
depth of failure plane of perimeter length Lper
(EA)p axial rigidity of plate
EB
externally bonded
EUR
Europe
F
component of longitudinal or axial force;
force profile
FIC
flexural intermediate crack
FRP
fibre reinforced polymer
HK
Hong Kong
IC
intermediate crack
k
interaction factor
ke
total axial force in transverse plates crossing
CDC as a proportion of Vtp
ki
total axial force in stirrups crossing CDC as a
proportion of Vs
750
ks
kt
shear slip
f
shear slip capacity
shear stress
f
maximum shear strength
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 751
Abstract: Results from nonlinear finite element analyses of fibre reinforced polymer
(FRP)-strengthened concrete beams and slabs are presented. The direct shear test, a
basic application that provides insight into FRP-concrete interfacial behaviour, is also
considered. The motivation for this work is the fact that, although there is a large
amount of experimental data available on the FRP strengthening of concrete structures,
a full understanding of the various loaddeformation behaviours and debonding
phenomenon is still lacking. The numerical models presented in this paper adopt a
displacementcontrolled solution and are capable of simulating FRP-strengthened
beams either in shear or in flexure, as well as slabs strengthened using either passive or
prestressed FRP laminates. Results of the different applications are presented and
compared with published test data, and a very good agreement in terms of the ultimate
load carrying capacities, loaddeflection behaviour and modes of failure, is obtained.
Key words: FRP, strengthening, numerical modelling, reinforced concrete, slabs, beams, flexure, shear, direct shear.
1. INTRODUCTION
The strengthening of reinforced concrete structures by
means of externally bonded fibre reinforced polymers
(FRPs) is now routinely considered as an effective method
for load capacity enhancement. However, with this
technique there is often a concern that possible failures
may occur due to debonding of the FRP from the concrete.
Laboratory investigations on FRP-strengthened concrete
beams in flexure and shear, for example, have shown that
despite the capability of achieving considerable increases
in strength capacities, premature failures by debonding
often limit the effectiveness of the strengthening schemes.
At present, our basic understanding of the mechanics of
the bond and failure between the FRP and the concrete for
such applications is somewhat limited.
As far as numerical studies on FRP-strengthened
beams are concerned, researchers have recently
751
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 752
Analysis of the Load-Deformation Behaviour and Debonding for FRP-Strengthened Concrete Structures
752
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 753
In Eqn 3
Bond
stress
se = max K 0
max
(5)
K0 = Ka Kc
(Ka + Kc ) ,
(6)
where
Slip
S0
K a = Ga ta
(7)
K c = Gc tc ,
(8)
= max
(1)
max = 1.5w ft
(2)
(3)
(2.25 b
bc
) (1.25 + b
= max exp ( s / s0 1 )
(9)
where
=
1
Gf
max s0 3
(10)
s
,
s0
w =
Here Ga and ta, and Gc and tc are the shear modulus and
the thickness of the adhesive and concrete, respectively.
The value of tc is taken as 5 mm; this corresponds to the
effective thickness whose deformation forms part of the
interfacial slip (Lu et al. 2005b).
For the descending part, i.e., s > s0
bc
),
(4)
G f = 0.308 w2 ft f ( K a )
(11)
753
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 754
Analysis of the Load-Deformation Behaviour and Debonding for FRP-Strengthened Concrete Structures
Interface
element
C.L.
Applied load
Applied load
9-node
plane stress
(concrete)
9-node
plane stress
(concrete)
Truss element
(steel)
Support edges
Interface
element
Truss element
(attached FRP)
(a) Direct shear test
C.L.
Applied load
C.L.
Applied load
C.L.
Attached FRP
Support edges
A s'
Support
Attached
U-shape FRP
Concrete
node
FRP
node
As
Detail A
concrete node
Interface element
FRP node
Detail C
754
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 755
755
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 756
Analysis of the Load-Deformation Behaviour and Debonding for FRP-Strengthened Concrete Structures
1.6
1.4
1.4
1.2
1
1
0.8
Tan (2002)
Zhao et al. (2000)
Takeo et al. (1997)
Ren (2003)
Bizindavyi and Neale (1999)
Chajes et al. (1996)
0.6
0.4
0.2
0
0
20
40
60
Specimen reference
80
Pnum /Pexp
Pnum /Pexp
1.2
0.8
Brena (2003)
Ross (1999)
Kaminska and Kotynia (2000)
Ritchie (1991)
M'Bazaa (1995)
Chicoine (1997)
0.6
0.4
0.2
0
0
100
10
15
20
25
30
Specimen reference
1.6
1
1.4
0.8
1
0.8
Adhikary and Mutsuyoshi (2004)
Pelligrino and Modena (2002)
Khalifa and Nanni (2000)
Chaallal et al. (1998)
0.6
0.4
0.2
Pnum /Pexp
Pnum /Pexp
1.2
0.6
Mosallam & Mosalam (2003)
Harajli & Soudki (2003)
Ebead & Marzouk (2004)
Longworth et al. (2004)
0.4
0.2
0
0
0
5
10
Specimen reference
15
8
10
6
Specimen reference
12
14
(d) Slabs
756
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 757
120
160
100
140
Debonding
Load (kN)
Load (kN)
120
80
60
40
Control-exp
Control-num
Po-exp
Po-num
20
100
Control-exp
Control-num
A1-exp
A1-num
A2-exp
A2-num
A3-exp
A3-num
80
60
40
20
0
0
20
40
60
80
100
10
20
30
40
Deflection (mm)
Deflection (mm)
50
60
180
Rupture
160
Debonding
Load (kN)
140
120
100
80
Po-exp
Po-num
P1-exp
P1-num
P3-exp
P3-num
60
40
20
0
0
20
40
60
80
Deflection (mm)
(c) Chicoine (1997)
400
500
300
TR30D3-exp
TR30D3-num
TR30D4-exp
TR30D4-num
TR30D2-exp
TR30D2-num
200
100
150
100
B-8-exp
50
B-8-num
0
0
12
16
12
Deflection (mm)
Deflection (mm)
16
757
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 758
Analysis of the Load-Deformation Behaviour and Debonding for FRP-Strengthened Concrete Structures
70
250
60
200
Load (kN)
Load (kN)
50
40
SA1F10-exp
30
SA1F10-num
20
150
100
SA1F5-exp
10
B2-SL1-exp
B2-SL1-num
B2-SL4-exp
B2-SL4-num
50
SA1F5-num
0
0
10
15
20
20
40
60
80
100
Deflection (mm)
Deflection (mm)
1 kN
2.28 kN
3.56 kN
4.84 kN
6.12 kN
7.4 kN
2
1
0
10
20
30
40
50
1 kN
3.2 kN
5.4 kN
7.6 kN
9.8 kN
12 kN
2
1
0
0
60
20
40
60
80
100
1 kN
3.2 kN
5.4 kN
5
7.6 kN
9.8 kN
12 kN
2
1
0
0
50
100
150
Distance along the interface (mm)
200
120
1kN
3.24 kN
5.48 kN
7.72 kN
9.96 kN
12.2 kN
2
1
0
0
50
100
150
200
Distance along the interface (mm)
250
758
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 759
1.2
C. L
0.8
b
a
0.6
0.4
0.2
0
0.2 0
200
400
600
c
800
1000
b
1200
1400
P = 30% Pcr
P = 60% Pcr
P = Pcr
Distance along the interface (mm), l
0.4
0.6
0.8
C. L
3
c
b
a
2
1
0
0
200
400
600
c
800
a
1000
b
1200
1
2
3
1400
6
c
b
a
4
2
0
0
200
400
600
c
800
1000
b
1200
1400
2
4
C. L
Figure 10. Interfacial shear stress distributions for Specimen A3 (Brena et al. 2003)
759
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 760
1 kN
2.28 kN
3.56 kN
4.84 kN
6.12 kN
7.4 kN
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
Analysis of the Load-Deformation Behaviour and Debonding for FRP-Strengthened Concrete Structures
0
0
10
20
30
40
50
Distance along the interface (mm)
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
1 kN
3.2 kN
5.4 kN
7.6 kN
9.8 kN
12 kN
60
120
1 kN
3.2 kN
5.4 kN
7.6 kN
9.8 kN
12 kN
0.2
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
1 kN
3.24 kN
5.48 kN
7.72 kN
9.96 kN
12.2 kN
0.25
Interfacial slip (mm)
20
40
60
100
80
Distance along the interface (mm)
0.2
0.15
0.1
0.05
0
50
100
150
Distance along the interface (mm)
200
50
100
150
200
Distance along the interface (mm)
250
Figure 11. Interfacial slip profiles for the direct shear application
200
188 kN
291 kN
332 kN
410 kN
200
100
300
0
0.0
0.4
0.8
Slip (mm)
1.2
1.6
150
100
59 kN
109 kN
150 kN
50
166 kN
0
0.4
0.0
Slip (mm)
0.4
Figure 12. Interfacial slip profiles along the FRP laminate depth for shear-strengthened beams
760
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 761
0.7
0.40
0.25
0.5
slip (mm)
Slip (mm)
0.30
P = 206.5 kN
P = 275.4 kN
P = 303.0 kN
P = 316.7 kN
P = 330.5 kN
0.6
P = 26.4 kN
P = 39.6 kN
P = 52.8 kN
P = 58.0 kN
P = 66.0 kN
0.35
0.20
0.15
0.4
0.3
0.2
0.10
0.1
0.05
0.00
0a
100
200
300 b
400
Distance from slab end towards the centre (mm)
0
0
b
200
400
600
800
1000
Distance from slab end towards the centre (mm)
0.45
0.40
0.35
P = 11.1 kPa
P = 27.0 kPa
P = 45.8 kPa
P = 58.0 kPa
Slip (mm)
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0 a
200
400
600
800
1000
1200 b1400
Distance from slab end towards the centre (mm)
(c) Specimen C-REP-R1
Figure 13. Slip profiles for Specimens SA1F15, GFRP-F-035% and C-REP-R1
wraps, the interfacial slips are higher at the top edge and
shift to negative values around the mid-depth. With an
increase of the applied shear force up to failure, the
interfacial positive and negative values of slip are
increased. It is thus obvious that using U-shaped FRP
wraps is more efficient in mitigating debonding, as can
be seen from comparing these slip values with those for
the specimen with side-bonded laminates.
4.5.3. Two-way slabs
The predicted slip profiles along the FRP/concrete
interfaces, at different load levels, are shown in Figure 13
for the Specimen SA1F15 tested by Harajli and Soudki
(2003), Specimen GFRP-F-0.35% tested by Ebead and
Marzouk (2004), and Specimen C-REP-R1 tested by
Mosallam and Mosalam (2003).
In Figure 13(a) it can be observed that the slip values
at the centre of the slab are significantly higher than
those at the end of the laminate. This suggests a punching
shear failure associated with debonding of the FRP
material in the central region of the slab. By contrast, the
slip profiles for Specimen GFRP-F-0.35% indicate that
761
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 762
Analysis of the Load-Deformation Behaviour and Debonding for FRP-Strengthened Concrete Structures
762
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 763
Lu, X.Z., Jiang J.J., Teng, J.G. and Ye, L.P. (2006). Finite element
simulation of debonding in FRP-to-concrete bonded joints,
Construction and Building Materials, Vol. 20, No. 6, pp. 412424.
Malek, A. and Saadatmanesh, H. (1998a). Ultimate shear capacity
of reinforced concrete beams strengthened with web-bonded fibre
reinforced plastic plates, ACI Structural Journal, Vol. 95, No. 4,
pp. 391399.
Malek, A. and Saadatmanesh, H. (1998b). Analytical study of
reinforced concrete beams strengthened with web-bonded
fibre reinforced plastic plates or fabrics, ACI Structural Journal,
Vol. 95, No. 3, pp. 343352.
MBazaa, I. (1995). Renforcement en flexion de poutres en bton
arm a laide de lamelles en matriaux composites: Optimisation
de la longueur des lamelles, M.A.Sc. Thesis, University of
Sherbrooke, Sherbrooke, Quebec, Canada.
Mosallam, A.S. and Mosalam, K.M. (2003). Strengthening of twoway concrete slabs with FRP composite laminates, Construction
and Building Materials, Vol. 17, No. 1, pp. 4354.
Nitereka, C. and Neale, K.W. (1999). Analysis of reinforced concrete
beams strengthened in flexure with composite laminates,
Canadian Journal of Civil Engineering, Vol. 26, pp. 646654.
Pellegrino, C. and Modena, C. (2002). Fibre reinforced polymer
shear strengthening of reinforced concrete beams with transverse
steel reinforcement, Journal of Composites for Construction,
ASCE, Vol. 6, No. 2, pp. 104111.
Ren, H.T. (2003). Study on Basic Theories and Long Time Behavior
of Concrete Structures Strengthened by Fiber Reinforced
Polymers, Ph.D. Thesis, Dalian University of Technology, China.
Ritchie, P.A., Thomas, D., Lu, L-W. and Connelly, G. (1991).
External reinforcement of concrete beams using fibre
763
ASE 9-6-03_Neale
4/12/06
5:10 pm
Page 764
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 765
Key words: Analytical modelling, debonding, energy release rate, fracture mechanics, FRP, geometrically nonlinear
analysis, RC beams, strengthening.
1. INTRODUCTION
Strengthening, upgrading, and retrofitting of existing
structures with externally bonded fiber reinforced polymer
(FRP) composite materials have gained a worldwide
acceptance in the past two decades (Neale and
Laboissiere 1997; Bakis et al. 2002). The versatility of
the strengthening method, its applicability to a wide
range of structural problems, the ease of installation
of the composite systems on site, and the improved
mechanical properties of the composite materials are
among the main advantages of the method. On the other
hand, it is also characterized by the formation of brittle
modes of failure, and, especially, by the development
of unstable and sudden debonding failure modes
(Buyukozturk and Hearing 1998). The two most dominant
765
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 766
Fracture Mechanics Approach to Geometrically Nonlinear Debonding Problems in RC Beams Strengthened with Composite Materials
Debonding growth
Unbonded regions
(a)
(b)
(c)
Figure 1. Debonding failure in RC beams strengthened with composite materials: (a) Edge debonding; (b) Debonding near shear-flexure
cracks; (c) Voids and debonded regions in a peeled-off FRP strip
766
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 767
767
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 768
Fracture Mechanics Approach to Geometrically Nonlinear Debonding Problems in RC Beams Strengthened with Composite Materials
x,uorc
zrc,wrc
c.g
ua
za,wa
rc
Yrc d
uofrp
Mfrp
xx
RC Beam
frp
Nxx
ca
zfrp,wfrp
Qfrp
xx
dfrp
azz(x,0)
(a)
_
Prc
j
nrc
qrc
_
Mrc
j
azz(x,ca)
axz(x,ca)
FRP Strip
x = xj
Debonded
region
axz(x,0)
Adhesive
Layer
_ rc
Nj
Mfrp
xx
Nfrp
xx
frp
Qxx
(b)
(c)
Figure 2. Notation and sign conventions: (a) Deflections and coordinates; (b) Loads; (c) Stresses and stress resultants
(U + V) = 0
rc
xx
xxrc dvrc
v rc
frp
frp
frp
a
a
a
a
a
xx xx dv xz xx + zz zz dv
v frp
(2)
va
768
u i ( x, zi ) = uoi z i w,ix ;
w i ( x, z i ) = w i ;
=u
i
xx
i
o, x
( )
1
+ w,ix
2
(3a-c)
2
z w
i
i
, xx
(1)
U =
FRP strip (i = frp); axz and azz are the shear and vertical
normal stress fields in the adhesive layer, respectively;
axz and azz are the shear angle and the vertical normal
strain fields in the adhesive layer, respectively; and vi
(i = rc,a,frp) is the volume of each component.
The kinematic relations for the RC beam and the
FRP strip independently follow the Bernoulli-Euler
assumption and the theory of large displacements,
moderate rotations, and small strains as follows:
zz = w,z;
a
xz = u,az + w,ax
a
(4a,b)
dx +
j =1 x = 0
rc
( N j uorc
rc
P j wrc
rc
Mj
rc
, x ) D ( x
x j )dx
(5)
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 769
where qrc, nrc, and mrc are distributed loads and bending
rc
rc
rc
moments, respectively, N j , P j , and M j are
concentrated loads and bending moments at x=xj, D is
the Dirac function, and NC is the number of the
concentrated loads.
2.2. Interfacial Compatibility and Debonding
Conditions
The compatibility conditions at the fully bonded interfaces
of the adhesive layer are:
a
a
rc
rc
rc
rc rc
u ( x, z = 0 ) = u ( x, Y ) = u o Y w , x
(6)
a
rc
rc
a
w ( x, z = 0 ) = w ( x ) = w
(7)
a
a
frp
u ( x, z = c ) = u ( x,
a
d frp
d frp frp
) = u ofrp +
u
2
2 ,x
w ( x, z = c ) = w ( x ) = w
a
frp
frp
(8)
(10)
(12)
N xxfrp,x ca b xza ( x, z a = c a ) = 0
(13)
rc rc
rc
rc a
a
M xx
)
, xx + ( N xx w , x ), x + 0 bY xz , x ( x, z = 0)
+ 0 b zza ( x, z a = 0 ) = q rc + m,rcx
(14)
d frp a
( x, z a = c a )
2 xz , x
(15)
a
a
a
ca b zz ( x, z = c ) = 0
xza ,x + zza ,z = 0
(16)
xza ,z = 0
(17)
(9)
where
is the height of the center of area axis
(reference axis) in the RC beam, dfrp is the thickness of
the FRP strip, ca is the thickness of the adhesive layer,
and za is the vertical coordinate of the adhesive
layer, measured from the adhesive-concrete interface
downwards, see Figure 2(b).
In the debonded regions, the damaged interface
cannot transfer shear or vertical tensile stresses. Hence,
the requirements of compatible longitudinal deformations
are replaced with the conditions of stress free surfaces.
For example, if the adhesive-concrete interface is
debonded, Eqn 6 is replaced with:
zza ( x, z a = 0 ) = 0
N xxrc ,x + 0 b xza ( x, z a = 0 ) = n rc
Yrc
xza ( x, z a = 0 ) = 0
where Nixx and Mixx are the in-plane stress resultant and
the bending moment in the RC beam (i = rc) and the
FRP strip (i = frp), b is the width of the adhesive layer,
i is a flag that defines the bonding condition at the
adhesive concrete interface (j = 0) or the adhesive-FRP
interface (j = ca), thus j = 1 for a fully bonded interface
and j=0 for a delaminated interface; and i is a flag that
defines the contact conditions at the adhesive concrete
interface (j = 0) or the adhesive-FRP interface (j = ca)
and equals 1 for debonding with contact and 0 for
debonding without contact. In the fully bonded region,
0 = ac = 0 = ac = 1. Thus, by using the proper
combination of flags, Eqns 1217 are valid for all
characteristic regions and sub-region.
2.4. Constitutive Relations
The constitutive relations for the RC beam are:
1
769
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 770
Fracture Mechanics Approach to Geometrically Nonlinear Debonding Problems in RC Beams Strengthened with Composite Materials
= B (u
frp
11
frp
o,x
(19)
1
)2 ) D11i w,fxxrp
+ ( w,frp
2 x
frp
frp
frp
where A11 , B11 , D11 are the extensional, coupling, and
flexural rigidities of the FRP strip (Vinson and
Sierakowski 1986) multiplied by its width, b. Note that in
most practical cases, a symmetric (and balanced) layup,
in which the coupling rigidity vanish, is used. Hence, in
frp
the formulation presented next B11 = 0 is assumed.
The constitutive relations of the adhesive layer are:
=E ;
a
zz
a
zz
=G
a
xz
a
xz
( x, z ) = ( x ) =
a
a
xz
(21)
w frp w rc a
a c
( x, z ) =
E
+
za
,
x
a
c
2
a
zz
w a ( x, z a ) =
770
a 2
(22)
a a ( ) ( a)
z
6
Ga
E 4
a
, xx
a
(z )
w,frp
w,rcx w,rcx z a + uorc Y rc w,rcx
x
a
2c
0 ca E a ( w frp w rc )
( x, z ) =
ca
(26)
(27)
frp frp
frp frp frp
frp
a
A11 u o, xx + A11 w, x w, xx 0 ca b = n
(28)
3
rc 2 rc
rc
EA ( w, x ) w, xx
2
2
+ 0 ca
(24)
(29)
bEa
( w frp w rc ) = q rc
ca
x
0 ca
2 ,x
frp
frp frp
D11frp w ,frp
xxxx + A11 u o, x w , xx +
0 ca
,xa z a ( z a c a ) ( w frp w rc ) z a
+
+ w rc (23)
2E a
ca
u a ( x, z a ) =
(25)
(20)
a
xz ,z
xza ( x, z a ) = 0 ;
a
zz
1
)2 ) B11i w,fxxrp
N xxfrp = A11frp (uofrp, x + ( w,frp
x
2
frp
xx
(30)
bEa
( w frp w rc ) = q frp
ca
c a a (c a) 3 a
0 ca u rco u ofrp +
12 Ea , xx
Ga
c a + d frp frp
ca
+ Y rc w ,rcx
w =0
2 ,x
(31)
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 771
i( )
o
=u
i( + )
o ;
i( )
=w
i( + )
i
; w
i( )
,x
=w
i( + )
,x
(32a-c)
i
i( )
i( + )
N xx N xx = N j , M ixx( ) + M ixx( + ) = M ij
(M
(M
i( )
xx , x
+ ( N ixx( )w ,ix( ) ) + b a ( )Y i
i( + )
xx , x
+ (N
i( + )
xx
i( + )
,x ) +
a( + )
Y = Pj
i
(33a,b)
G = J = Wdy T ds
(35)
(33c)
G=
1
1 (U + V )
=
=
A
b a
b
a
(36)
771
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 772
Fracture Mechanics Approach to Geometrically Nonlinear Debonding Problems in RC Beams Strengthened with Composite Materials
qrc = q
E = 30 GPa
250
Ea = 0.7 GPa
Ga = 0.28 GPa
Debonded
region
161
a
CFRP Laminate
E11 = 120 GPa
161
L = 2400 mm
5.0
0.8
100
(a)
(b)
Figure 3. Geometry, loading, and mechanical properties: (a) Geometrical layout; (b) Cross section.
100
q (kN/m)
90
max(|wrc|)
80
70
max(|wfrp|)
60
50
40
30
20
10
w (mm)
0
2
4
6
8
10
12
100
q (kN/m)
90
80
70
60
50
40
30
20
10
0
0.05
w
Dw
w, Dw ()
0.1
(a)
0.15
(b)
0.2
0.25
100
90
80
70
60
50
40
30
20
10
0
25
q (kN/m)
Nfrp
xx (x = L/2) (kN)
20
15
10
(c)
Figure 4. External load versus structural response, a=50 mm: (a) Peak vertical deflections; (b) Relative deflection and waviness parameters;
(c) Inplane force in the FRP strip at midspan.
w = w frp w rc dx
0
Dw = w ,fxrp w ,rcx dx
0
772
frp
rc
w + w dx ;
0
frp
rc
w, x + w, x dx
0
(39)
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 773
10
8.5
w (mm)
wfrp
wrc
wfrp
wfrp
7.5
wrc wfrp
2
0
w (mm)
x (mm)
0
500
1000
1500
2000
2500
6.5
900
x (mm)
1000
1100
(a)
50
50
0
50
50
100
100
150
150
200
200
250
250
300
300
350
350
400
450
Upper face
Lower face
500
1000
1300
1400
1500
(b)
xx (MPa)
1200
xx (MPa)
400
x (mm)
1500
2000
2500
(c)
450
1150
Upper face
Lower face
1170
x (mm)
1210
1190
1230
1250
(d)
Figure 5. Vertical deflections and longitudinal stresses in the FRP strip with q=70 kN/m and a=50 mm: Vertical deflections; (a) Along the
beam; (b) zoom on the debonded region; Longitudinal normal stresses in the FRP strip: (c) Along the beam; (d) zoom on the debonded region.
773
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 774
Fracture Mechanics Approach to Geometrically Nonlinear Debonding Problems in RC Beams Strengthened with Composite Materials
a (MPa)
1.5
1.5
0.4
0.8
a (MPa)
1.2
0.5
0.5
1.6
100
200
300
0.5
0.5
1.5
1.5
x (mm)
2
0
500
1000
1500
2000
2500
2
1150
x (mm)
1170
1190
(a)
4
1210
1230
1250
(b)
2
azz (MPa)
azz (MPa)
AdhesiveConcrete
Interface
4
6
6
8
AdhesiveFRP
Interface
10
10
12
12
x (mm)
14
0
500
1000
1500
2000
2500
(c)
14
1150
x (mm)
1170
1190
1210
1230
1250
(d)
Figure 6. Shear and vertical normal stresses in the adhesive layer under q=70 kN/m and a=50 mm: Shear stresses: (a) Along the beam;
(b) Zoom on the debonded region; Vertical normal stresses: (c) Along the beam; (d) Zoom on the debonded region.
774
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 775
100
q (kN/m)
90
80
70
60
50
min (zz [za = 0])
40
max (zz [za = 0])
30
min (zz [za = Ca])
20
max (zz [za = Ca])
10
zz (MPa)
0
50 40 30 20 10
0
10
100
90 q (kN/m)
80
70
60
50
40
30
20
10
0
0.5 1
a (MPa)
1.5 2
(a)
2.5 3
3.5 4
100
q (kN/m)
90
80
70
60
50
40
30
20
10
G (J/m2)
0
100 200 300 400 500 600 700 800 900
(b)
(c)
Figure 7. External load versus structural response, a=50 mm: (a) Peak vertical normal stresses (za=0 adhesive-concrete interface;
za=ca adhesive-FRP interface); (b) Peak shear stresses; (c) Energy release rate.
300
300
a (mm)
300
a (mm)
a (mm)
max(|wrc|)
200
200
200
max(|wfrp|)
Dw
100
w, Dw ()
w (mm)
0
10
0.05 0.1
0.15
(a)
300
0.2 0.25
0.3 0.35
0
12
Nfrp
xx (x = L/2) (kN)
10
(b)
300
a (mm)
300
a (mm)
a (mm)
200
200
200
a3cr =185 mm
100
100
100
a2cr = 105 mm
a1cr = 65 mm
0
14 12 10 8
zz (MPa)
6
a (MPa)
0
0.5
(d)
1.5
(e)
(c)
2.5
Gc165 J/m2
100
100
G (J/m2)
10 20 30 40 50 60 70 80 90
(f)
Figure 8. Debonded length versus structural response at q=35 kN/m: (a) Peak vertical deflections; (b) Relative deflection and waviness
parameters; (c) Inplane force in the FRP strip at midspan; (d) Peak vertical normal stresses at adhesive layer interfaces (za=0 adhesiveconcrete; za=ca adhesive-FRP); (e) Peak shear stresses; (f) Energy release rate.
775
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 776
Fracture Mechanics Approach to Geometrically Nonlinear Debonding Problems in RC Beams Strengthened with Composite Materials
776
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 777
Bakis, C.E., Bank, L.C., Brown, V.L., Cosenza, E., Davalos, J.F.,
Lesko, J.J., Machida, A., Rizkalla, S.H. and Triantafillou, T.C.
(2002). Fiber-reinforced polymer composites for construction
state-of-the-art review, Journal of Composites for Construction,
ASCE, Vol. 6, No. 2, pp. 7387.
Broek, D. (1974). Elementary Engineering Fracture Mechanics,
Nordhoff Int., Leyden, The Netherlands.
Buyukozturk, O. and Hearing, B. (1998). Failure behavior of precracked concrete beams retrofitted with FRP, Journal of
Composites for Construction, ASCE, Vol. 2, No. 3, pp. 138145.
Buyukozturk, O., Gunes, O. and Karaca, E. (2004). Progress on
understanding debonding problems in reinforced concrete and
steel members strengthened with FRP composites, Construction
and Building Materials, Vol. 18, No. 1, pp. 119.
CEB-FIP (2001). Externally Bonded FRP Reinforcement for RC
Structures, CEB-FIP Bulletin No. 14, Fdration internationale
du bton (fib), Lausanne, Switzerland.
Deuring, M. (1993). Verstrken von Stahlbeton mit Gespannten
Faserverbundwerstoffen, EMPA Research report No. 224, Swiss
Federal Laboratories for Material Testing and Research,
Dbendorf, Switzerland. (in German).
Grace, N.F., Soliman, A.K., Abdel-Sayed, G. and Saleh, K.R.
(1999). Strengthening of continuous beams using fiber
reinforced polymer laminates, Proceedings of the Fourth
International Symposium on Fibre Reinforced Polymer
Reinforcement for Concrete Structures, SP-188, ACI, Farmington
Hills, MI, PP. 647657.
Kardomateas, G.A. and Pelegri, A.A. (1994). The stability of
delamination growth in compressively loaded composite plates,
International Journal of Fracture, Vol. 65, No. 3, pp. 261276.
Lau, K.T., Dutta, P.K., Zhou, L.M., and Hui, D. (2001). Mechanics
of bonds in an FRP bonded concrete beam, Composites Part B:
Engineering, Vol. 32, No. 6, pp. 491502.
Malek, A.M., Saadatmanesh, H. and Ehsani, M.R. (1998).
Prediction of failure load of RC beams strengthened with FRP
plate due to stress concentration at the plate end, ACI Structural
Journal, Vol. 95, No. 1, pp. 14252.
Mohamed Ali, A.S., Oehlers, D.J. and Bradford, M.A. (2002).
Interaction between flexure and shear on the debonding of RC
beams retrofitted with compression face plates, Advances is
Structural Engineering, Vol. 5, No. 4, pp. 223229.
Neale, K.W. and Laboissiere, P. (1997). State-of-the-art report on
retrofitting and strengthening by continuous fibre in Canada,
Non-metallic (FRP) Reinforcement For Concrete Structures, Vol. 1,
Japan Concrete Institute, pp. 2540.
Mukhopadhyaya, P. and Swamy, N. (2001). Interface shear stress:
a new design criterion for plate debonding, Journal of
Composites for Construction, ASCE, Vol. 5, No. 1, pp. 3543.
Oehlers, D.J. (2005). Draft - Design Guidelines for RC Structures
Retrofitted with FRP and Metal Plates: Beams and Slabs, Centre
for Infrastructure Diagnosis Assessment and Rehabilitation
(CIDAR) The University of Adelaide (http://www.civeng.
adelaide.edu.au/index.html).
777
ASE 9-6-04_Frostig
4/12/06
5:10 pm
Page 778
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 779
of Pittsburgh, Pittsburgh PA
Institute of Technology, Terre Haute IN,
3DMJM Harris, Pittsburgh PA
4S&ME, Columbia SC
2Rose-Hulman
1. INTRODUCTION
Debonding of externally bonded fiber reinforced
polymer (FRP) composite materials used for repair of
reinforced concrete flexural elements is commonly
observed. The most common debonding mode reported
in earlier experimental literature is the plate end mode
(PE) which initiates at the curtailment of the FRP plate
and propagates toward the midspan of the beam
(Oehlers 2005). PE debonding may be effectively
mitigated by providing anchorage to resist the peel
forces that are generated normal to the FRP or by
simply extending the FRP plate across the entire shear
span. In beams having relatively long shear spans, or
where the PE mode has been effectively mitigated,
debonding initiates at flexural (FIC) and/or flexural/
shear (SIC) cracks near the region of maximum moment
(Oehlers 2005). Under loading, these cracks open and
induce high interfacial shear stress accompanied by a
small amount of peeling causing delamination, which
propagates across the shear span in the direction of
decreasing moment (Sebastian 2001; Oehlers 2005;
Quattlebaum et al. 2005; Aidoo et al. 2006). This mode
of debonding is properly termed intermediate [flexural
779
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 780
780
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 781
Table 1. Summary of current guidelines for maximum allowable stress in bonded FRP subject
to fatigue loading
ACI 440.2R-02
Concrete Society TR55
JSCE Recommendations
CNR 200/2004
fib Bulletin 14
ffuf
ffu
ffud
Carbon FRP
Glass FRP
Aramid FRP
781
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 782
Concrete strength, fc
Reinforcing steel yield strength, fy
Reinforcing steel tensile strength, fu
CFRP strength, ffu
CFRP tension modulus, Ef
CFRP rupture strain, fu
CFRP thickness, tf
Adhesive type
Adhesive strength, fau
Adhesive tension modulus, Ea
Adhesive rupture strain, au
Adhesive thickness, ta
1Zorn
Z-series1
Z-series1
Q-series2
A-series3
23.3 MPa
429 MPa
667 MPa
2800 MPa
155 GPa
0.018
1.4 mm
L
14 MPa
2227 MPa
0.063
1.6 mm
23.3 MPa
429 MPa
667 MPa
2800 MPa
155 GPa
0.018
1.4 mm
H
25 MPa
4482 MPa
0.010
1.6 mm
29.6 MPa
446 MPa
735 MPa
2800 MPa
155 GPa
0.018
1.4 mm
T
47 MPa
2069 MPa
0.033
1.6 mm
35 MPa
364 MPa
530 MPa
2800 MPa
155 GPa
0.018
1.4 mm
T
47 MPa
2069 MPa
0.033
1.6 mm
152 mm
254 mm
2 - #3
Z-H1
Z-L1
Z-H2
Z-L2
Q-CL
Q-CH
Z-H2x1
Z-L2x1
Z-H4
Z-L4
4 - #4
25
51
25 51 25
102
#4
A-C
Tested over
simple span
of 8025 mm
660
2 #8
95
3 #10
95
3 #11
76
204
343
782
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 783
222 kN actuator
Load cell
Midspan deflection
L = 4570 mm
Load cell
Midspan deflection
8025 mm simple span
(b) A-series test set-up.
Cycles to
Specimen
failure, Nf
N=1
N = NF
N=1
N = NF
Z-H1
Z-L1
Z-H2
Z-L2
Z-H2x1
Z-L2x1
Z-H4
Z-L4
Q-CL
424,422
400,892
1,128,006
2,000,0001
2,000,0001
447,695
2,000,0001
2,000,0001
2,000,000
587,089
523,000
2,000,0001
239
248
199
207
220
207
175
174
150
250
230
217
241
250
230
196
201
204
203 (0.072ffu)
190 (0.068ffu)
188 (0.067ffu)
170 (0.061ffu)
199 (0.071ffu)
177 (0.063ffu)
143 (0.051ffu)
144 (0.051ffu)
232 (0.080ffu)
219 (0.078ffu)
205 (0.073ffu)
205 (0.073ffu)
190 (0.068ffu)
223 (0.080ffu)
188 (0.067ffu)
168 (0.060ffu)
159 (0.057ffu)
341 (0.120ffu)
320
993
400
993
341 (0.120ffu)
85 (0.030ffu)
387 (0.140ffu)
96 (0.034ffu)
Q-CH
A-C
CFRP debonding
observed?
no
no4
no4
no
yes
no4
yes
yes
no
yes
yes (Figure 4)
no
2,000,000 fatigue cycles followed by monotonic load to failure (see Table 4).
No instruments capable of assessing rebar strain operational following 2,000,000 cycles.
3 A-series specimens were recovered from a decommissioned bridge, thus reinforcing bar stresses are calculated rather than observed.
4 CFRP debonding observed only following fatigue-induced reinforcing bar rupture.
2
783
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 784
Z-L2
Z-H2x1
test designation1
44.3
56.9
1.70
6688
45.5
67.3
2.05
4300
6688
7444
Z-H4
Z-L4
45.2
56.1
1.70
3200
45.3
51.3
1.45
3909
49.2
47.4
1.33
2850
53.8
52.6
1.30
3259
51.8
72.2
1.91
4540
6863
6970
4813
5860
6595
A-C
F
51.3
60.5
1.67
3854
823
96.0
5.20
4500
809
58.4
2.50
4000
5807
9500
6900
1M
784
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 785
Debonding
27.6 kNm
Cracking moment = 5.9 kNm
Bending moment at maximum applied load during fatigue cycling (22.2 kN)
Inclined
inclined crack
crack
Debonding
debonding
Fatigue
fracture
fatigue fracture
of 13 mm bar
bar
4000
Q-CH
Strain, microstrain
3500
CFRP
Fatigue failure
3000
Rebar
2500
Rebar yield
2000
CFRP
1500
1000
Rebar
500
Q-CL
Data for final 600,000
cycles of Q-CL is not
available
1000000
2000000
Number of cycles
3000000
785
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 786
1000
60
Rebar
800
Z-L4 monotonic
50
CFRP
600
400
0
500000
1000000
1500000
Number of cycles
2000000
786
Strain, microstrain
Z-L4 fatigue
40
20
Z-L4 fatigue conditioning
10
N = 1 to 2,000,000
0
0
20
40
60
Midspan deflection (mm)
80
100
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 787
900
A-C monotonic
800
700
A-C fatigue
600
500
300
N = 1 to 2,000,000
200
100
0
0
25
50
75
100
125
150
175
200
787
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 788
788
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 789
APPENDIX: NOTATION
a
shear span length
DL dead load (AASHTO 2004)
Ea tensile modulus of adhesive
Ef tensile modulus of FRP
fau strength of adhesive
fc concrete strength
ffu ultimate capacity of FRP
ffud stress limit in FRP to mitigate debonding under
monotonic load conditions
ffuf stress limit in FRP under fatigue load conditions
fu reinforcing tensile strength
fy reinforcing yield strength
h
overall height (depth) of beam
IM impact load (AASHTO 2004)
LL live load (AASHTO 2004)
N number of cycles of fatigue loading
Nf number of fatigue cycles to failure
S
stress range
ta
thickness of adhesive
tf
thickness of FRP
au rupture strain of adhesive
fu rupture strain of FRP
789
ASE 9-6-05_Harries
4/12/06
5:10 pm
Page 790
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 791
2Department
Abstract: Generally, it is relatively easy to obtain a high joint strength with most
modern bonding systems if these are cured under ideal factory conditions. However,
civil engineering construction joints are cured on site and are required to last the
lifetime of the structural member in harsh environmental conditions. This implies that
site joints might not be constructed and cured as well as those fabricated in the factory.
The paper compares two possible methods for bonding an FRP composite
patch/plate to a steel adherend suitable for construction sites. The first method utilises
the accepted technique of bonding two dissimilar materials using a two-part cold cure
adhesive. The second method employs a pre-impregnated FRP composite plate in
conjunction with a compatible film adhesive; these comparisons are undertaken by
examining the results of double-strap butt joint tests. A possible site technique using
the pre-impregnated FRP composite to upgrade a steel beam whilst it is under a low
frequency vibration load is investigated; this represents the repair of a steel bridge
constantly under traversing traffic.
The butt joint test results show that on average, the pre-impregnated composite in
conjunction with the film adhesive leads to a failure load which is 15% higher than that
of the cold setting adhesive technique. The results of the rehabilitated beam tests show
that the bonded joint between the pre-impregnated CFRP composite and the steel
adherend did not suffer any significant damage from the low frequency vibrations
imposed upon the steel beam during the cure period.
Key words: Polymer composites, pre-impregnated composites, carbon fibres, glass fibres, film adhesive, cold
cure adhesives.
1. INTRODUCTION
Currently, one of the major uses of fibre reinforced
polymer (FRP) composites in the civil infrastructure is
in the repair and rehabilitation of highway and railway
bridge structures constructed from concrete and steel
materials. It has been estimated that in the USA, more
than one third of all bridges are substandard and of these
more than 43% are made from steel (Tavakkolizadeh
791
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 792
Advances in Adhesive Joining of Carbon Fibre/Polymer Composites to Steel Members for Repair and Rehabilitation of Bridge Structures
792
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 793
L.C. Hollaway, L. Zhang, N.K. Photiou, J.G. Teng and S.S. Zhang
793
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 794
Advances in Adhesive Joining of Carbon Fibre/Polymer Composites to Steel Members for Repair and Rehabilitation of Bridge Structures
794
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 795
L.C. Hollaway, L. Zhang, N.K. Photiou, J.G. Teng and S.S. Zhang
Material
UHM-CFRP (Unidirectional)
HM-CFRP (Unidirectional)
GFRP (45 to line of action
of load)
Film Adhesive
Thickness
per layer
(mm)
Tensile
strength
(MPa)
0.3
0.6
0.4
1120
2110
215
0.1
32
Elastic
modulus
(GPa)
270
135
16
3.7
Ultimate
strain
(%)
Poisson
ratio
0.4
1.6
1.7
0.32
0.28
0.15
0.9
0.37
Material
Tensile
strength
MPa
Modulus of
elasticity
GPa
Shear
strength
MPa
Temperature
range
C
Sikadur 31
14.8
6.87.3
21
530
795
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 796
Advances in Adhesive Joining of Carbon Fibre/Polymer Composites to Steel Members for Repair and Rehabilitation of Bridge Structures
796
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
A
B
C
H-M CFRP/GFRP
H-M CFRP/GFRP
UH-M CFRP/GFRP
UH-M CFRP/GFRP
UH-M CFRP
UH-M CFRP
H-M CFRP
H-M CFRP
Adhesive film
Adhesive film
Adhesive film
Adhesive film
Adhesive film
Adhesive film
Adhesive film
Adhesive film
Adhesive film
Sikadur 31
80C/4 hours
65C/16 hours
80C/4 hours
65C/16 hours
80C/4 hours
65C/16 hours
80C/4 hours
65C/16 hours
65C/16 hours
Ambient
temperature
10
H-M CFRP
H-M CFRP
Adhesive
Type
Thickness
(mm)
37.02
36.57
42.0
46.0
32.0
25.4
29.5
31.8
43.93
32.89
43.46
39.4
35.4
34.4
44.5
49.0
47.0
41.16
38.18
40.71
43.46
37.71
43.23
32.2
29.5
29.1
36.4
33.1
36.4
34.26
Failure
load
(kN)
41.47
40.02
46.83
36.4
40.23
28.90
44.0
35.95
35.3
30.27
Average
Failure load
(kN)
Failure
mode
5:09 pm
2*
A
B
C
A
B
C
A
Specimen
Curing
condition
4/12/06
1*
Series
Composite
plate
Steel
plate
ASE 9-6-06_Zhang
Page 797
L.C. Hollaway, L. Zhang, N.K. Photiou, J.G. Teng and S.S. Zhang
797
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 798
Advances in Adhesive Joining of Carbon Fibre/Polymer Composites to Steel Members for Repair and Rehabilitation of Bridge Structures
798
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 799
L.C. Hollaway, L. Zhang, N.K. Photiou, J.G. Teng and S.S. Zhang
799
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 800
Advances in Adhesive Joining of Carbon Fibre/Polymer Composites to Steel Members for Repair and Rehabilitation of Bridge Structures
300 mm
300 mm
150 mm
150 mm
CFRP
Adhesive
(or CFRP/GFRP composite)
(Two-part adhesive or film adhesive)
Steel adherend
80
56
50
15
35
120
Strain
gauges
The adhesive film was cut to size and laid onto the
beam, as were the various laminates in the stacking
sequence, as described above. The FRP composites on
the beam were covered by a Halar film and breather
blanket, and the whole was then covered with a heating
blanket before placing it in a vacuum bag. The bag
under a vacuum-assisted pressure of 1 bar was exposed
to the relevant temperature (either 65C for 16 hours
under the vibrating load condition or 80C for 4 hours
without imposed loads). After curing, the temperature of
the heating system was reduced to room temperature at
a steady rate.
60
P
Distribution beam
Steel beam
400 mm
FRP composite
1600 mm
50 mm
50 mm
800
Load (kN)
120
Control beam
80
40
10
20
30
40
Deflection at mid-span (mm)
50
60
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 801
L.C. Hollaway, L. Zhang, N.K. Photiou, J.G. Teng and S.S. Zhang
140
120
40 kN
100 kN
140 kN
170 kN
180 kN
100
80
60
40
20
0
100007500 5000 2500 0
2500 5000 7500 10000
Microstrain
140
40 kN
70 kN
90 kN
120 kN
135 kN
120
100
80
60
40
20
0
3000
2000
1000
1000
2000
3000
Microstrain
801
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 802
Advances in Adhesive Joining of Carbon Fibre/Polymer Composites to Steel Members for Repair and Rehabilitation of Bridge Structures
802
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 803
L.C. Hollaway, L. Zhang, N.K. Photiou, J.G. Teng and S.S. Zhang
803
ASE 9-6-06_Zhang
4/12/06
5:09 pm
Page 804
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 805
Wiss, Janney, Elstner Associates, Inc., 245 First Street, Suite 1200, Cambridge, Massachusetts, USA
Department of Civil, Construction, and Environmental Engineering, North Carolina State University, USA
3 Sutton-Kennerly and Associates, Greensboro, North Carolina, USA
Key words: high modulus CFRP, steel bridge, strengthening, rehabilitation, bond stresses.
1. INTRODUCTION
Innovative methods are required for the strengthening
and rehabilitation of steel structures that are deficient
due to the demand to increase the specified load and/or
deterioration as a result of corrosion. A considerable
body of research has established the successful use of
carbon fiber reinforced polymer (CFRP) materials for
strengthening concrete structures. With the introduction
of new high modulus CFRP materials, the possibility for
providing a solution to the ongoing problem of
infrastructure deterioration may be extended to steel
structures as well. Bond stresses may be much more
critical for steel structures than for concrete structures
since more strengthening material is needed for steel
structures to achieve a similar increase in strength due to
the inherent high strength of steel and also since the
debonding failure does not occur in the substrate as in
concrete structures. Further complications may arise due
to the potential for galvanic corrosion between the
carbon and steel materials. Despite these challenges,
805
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 806
806
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 807
807
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 808
1997). The second reason is that salts can leach out of the
glass fibers themselves. This causes a concentration
gradient that can draw more water into the interface or into
voids within the joint. The pressure generated by this
process can cause the voids to blister, resulting in damage
to the surrounding material (Frieze and Barnes 1996).
Tucker and Brown (1989) have found that glass fibers
placed within a carbon fiber composite result in the
blistering of the composite by creating conditions
favorable for the development of a strong osmotic
pressure within the composite. Clearly, water being drawn
within the bond line by osmotic pressure is not favorable
for maintaining a durable bond. Other materials may be
more suitable for this purpose. Hollaway and Cadei (2002)
reported that a polyester drape veil was installed to
provide insulation between the carbon fiber and the cast
iron to prevent direct contact between the CFRP and the
steel, although no durability information was given for this
combination of materials. Finally, although fiber-glass or
epoxy films can be used to provide effective insulation,
Sloan and Talbot (1992) note that few materials retain
their insulating properties for more than a few years due
to wear, chemical breakdown or electrolyte absorption. A
monitoring program could also be initiated to identify
cathodic sites so that galvanic corrosion damage could be
stopped or mitigated.
2.4. CFRP Detailing
For ease of shipping and handling CFRP strips are
typically manufactured in finite lengths that are suitable
for strengthening most typical short span girders. To
facilitate the implementation of the strengthening
system to longer span girders, it is necessary to develop
an effective technique to splice adjacent lengths of the
CFRP materials. The use of a bonded splice cover plate
is a promising technique to ensure continuous transfer
of forces across splice joints. However, the use of this
technique requires a careful consideration of localized
bond stress concentrations that may occur at or near
splice locations. Other researchers have demonstrated
that careful detailing of the ends of a CFRP strip can
significantly reduce the bond stress concentrations that
typically occur at or near the strip ends. Tapering of the
CFRP strips at their edge avoids imposing a local stress
concentration at the boundary of the joint. For lapped
joints it was recommended that peel stresses should be
designed out of the joint by tapering the ends of the
overlap (Hart-Smith 1980). In the case of joints made to
FRP adherends, it was also noted that this would also
reduce the possibility of an interlaminar failure within
the FRP. Allan et al. (1988) recommended finishing
steel to CFRP joints with a 10:1 taper (5.7 degrees) at
their ends to reduce stress concentrations.
808
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 809
809
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 810
38 mm
102 mm
6.5 mm
2.5 mm
3.3 mm
57 mm
102 mm
Development length
Length of CFRP strip
813 mm
864 mm
810
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 811
200 mm
6 mm
50 mm
100 mm
50 mm
100 mm
Strain gauge
Figure 3. Detail of splice joint configurations for tension test specimens (side view)
Table 1. CFRP strip strain at rupture/debonding for tested adhesives and development lengths
Development Length
Adhesive
203.2 mm
152.4 mm
127 mm
101.6 mm
76.2 mm
50.8 mm
Weld-On SS620
SP Spabond 345
Vantico Araldite 2015
Jeffco 121
Fyfe Tyfo MB2
Sika Sikadur 30
3.077 rupture
2.878 rupture
3.094 rupture
2.981 rupture
3.470 rupture
2.814 debond
2.964 rupture
2.943 rupture
2.980 rupture
3.276 rupture
3.060 debond
2.662 rupture*
3.161 rupture
3.111 rupture
2.820 rupture
2.438 debond
2.096 debond
2.903 rupture
2.433 debond
2.772 debond
2.589 debond
1.833 debond
av = E frp t frp
2 1
x2 x1
(1)
811
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 812
Table 2. Maximum shear stress (MPa) and failure mode for beams strengthened by adhesive bonding
of CFRP strips using different development lengths
Development Length
Resin
Weld-On SS620
SP Spabond 345
Jeffco 121
plys
254 mm
203.2 mm
127 mm
101.6 mm
1
1
2
1
2
49.9 rupture
61.8 rupture
21.3 rupture
17.7 rupture
36.7 rupture
13.3* debond
*This average shear stress was determined over the last 25.4 mm of the CFRP strip, unlike the remaining values that were determined over the last 6.4 mm.
812
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 813
406.4 mm
50.8 mm
152.4 mm
60
50
40
30
20
Predicted from bond model
10
50.8
101.6
152.4
203.2
254.0
304.8
355.6
406.4
Figure 4. Comparison of predicted shear stress distribution and shear stress distribution determined from testing
of beam using 101.6 mm development length
813
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 814
250
Strain guage
16
200
12
Configuration C
Load (kN)
150
10
Configuration B
100
6
Configuration A
14
50
2
0
0.0000
0.0005
0.0010
Strain (mm/mm)
0
0.0020
0.0015
400
400
350
350
Splice plate axial stress (MPa)
300
250
200
Calculated
Measured
150
100
50
300
250
Calculated
200
Measured
150
100
50
0
50
50
150
250
350
(a) Joint configuration A
450
0
50
50
150
250
350
(b) Joint configuration C
450
Figure 6. Comparison of measured and calculated axial stress distribution in the splice cover plate
814
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 815
40
30
At 80 kN
Load level
20
10
10
20
30
50.0
50.0
Joint A
150.0
Joint B
250.0
Joint C
350.0
450.0
815
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 816
REFERENCES
Albat, A.M. and Romilly, D.P. (1999). A direct linear-elastic
analysis of double symmetric bonded joints and reinforcements,
Composites Science and Technology, Vol. 59, pp. 11271137
Allan, R.C., Bird, J. and Clarke, J.D. (1988). Use of adhesives in
repair of cracks in ship structures, Materials Science and
Technology, Vol. 4, pp. 853859.
Allen, K.W., Chan, S.Y.T. and Armstrong, K.B. (1982). Coldsetting adhesives for repair purposes using various surface
preparation methods, International Journal of Adhesion and
Adhesives, Vol. 2, No. 4, pp. 239247.
Belingardi, G., Goglio, L. and Tarditi, A. (2002). Investigating the
effect of spew and chamfer size on the stresses in metal/plastics
adhesive joints, International Journal of Adhesion and
Adhesives, Vol. 22, pp. 273282.
Brown, A.R.G. (1974). Corrosion of CFRP to metal couples in
saline environments, Proceedings of the 2nd International
Conference on Carbon Fibres, London, England, February, Paper
No. 35, pp. 230241.
Brown, S.R. and De Luccia, J.J. (1977). Corrosion characteristics of
naval aircraft metals and alloys in contact with graphite-epoxy
composites, Proceedings of the Environmental Degradation of
Engineering Materials Conference, Virginia Polytechnic Institute
and State University, Blacksburg, Virginia, October, pp. 277288.
Buyukozturk, O., Gunes, O. and Karaca, E. (2004). Progress on
understanding debonding problems in reinforced concrete and
steel members strengthened using FRP composites, Construction
and Building Materials, Vol. 18, No. 1, pp. 919.
Cadei, J.M.C., Stratford, T.J., Hollaway, L.C. and Duckett, W.G.
(2004). Strengthening Metallic Structures Using Externally
Bonded Fibre-Reinforced Polymers. Publication C595,
Construction Industry Research and Information Association
(CIRIA), London, UK, 234 pp.
Choqueuse, D., Davies, P., Mazeas, F. and Baizeau, R. (1997).
Aging of composites in water: Comparison of five materials in
terms of absorption kinetics and evolution of mechanical
properties, High Temperature and Environmental Effects on
Polymeric Composites: 2nd Volume, ASTM Special Technical
Publication 1302, Thomas S. Gates and Abdul-Hamid Zureick,
eds., American Society for Testing and Materials, pp. 7396.
Dawood, M. (2005). Fundamental Behavior of Steel-Concrete
Composite Beams Strengthened with High Modulus Carbon
Fiber Reinforced Polymer Materials, Masters Thesis, North
Carolina State University, 213 pp.
El Damatty, A., Abushagur, M. and Youssef, M.A. (2003).
Experimental and analytical investigation of steel beams
rehabilitated using GFRP Sheets, Steel and Composite
Structures, Vol. 3, No. 6, pp. 421438.
Evans, U.R. and Vera, E.R. (1958). Corrosion and its Prevention at
Bimetallic Contacts, Her Majestys Stationery Office, London,
England.
Francis, R. (2000). Bimetallic Corrosion: Guides to Good Practice
in Corrosion Control, National Physical Laboratory, Teddington,
816
Middlesex, 15 pp.
Frieze, P.A. and Barnes, F.J. (1996). Composite materials for
offshore application new data and practice, Proceedings of the
28th Annual Offshore Technology Conference, Construction and
Installation, Houston, Texas, May, Vol. 3, pp. 247253.
Garden, H.N., Quantrill, R.J., Hollaway, L.C., Thorne, A.M. and
Parke, G.A.R. (1998). An experimental study on the anchorage
length of carbon fibre composite plates used to strengthen
reinforced concrete beams, Construction and Building
Materials, Vol. 42, No. 2, pp. 175188.
Gettings, M. and Kinloch, A.J. (1977). Surface analysis of
polysiloxane/metal oxide interfaces, Journal of Materials
Science, Vol. 12, No. 12, pp. 25112518
Harris, A.F. and Beevers, A. (1999). The effects of grit blasting on
surface properties for adhesion, International Journal of
Adhesion and Adhesives, Vol. 19, No. 6, pp. 445452.
Hart-Smith, L.J. (1980). Further developments in the design and
analysis of adhesive-bonded structural joints, ASTM Special
Technical Publication 749, Joining of Composite Materials: A
Symposium, K.T. Kedward, Ed., April, pp. 331.
Hashim, S.A. (1999). Adhesive bonding of thick steel adherends for
marine structures, Marine Structures, Vol. 12, No. 6, pp. 405423.
Hildebrand, M. (1994). Non-linear analysis and optimization of
adhesively bonded single lap joints between fibre-reinforced
plastics and metals, International Journal of Adhesion and
Adhesives, Vol. 14, No. 4, pp. 261267.
Hollaway, L.C. and Cadei, J. (2002). Progress in the technique
of upgrading metallic structures with advanced polymer
composites, Progress in Structural Engineering Materials, Vol.
4, No. 2, pp. 131148.
Hutchinson, A.R. (1987). Surface pretreatment the key to durability,
Proceedings of the International Conference on Structural Faults &
Repair, University of London, July, pp. 235244.
Institution of Structural Engineers (1999). A Guide to the Structural
Use of Adhesives, The Institution of Structural Engineers,
London, UK, 51 pp.
Karbhari, V.M. and Shulley, S.B. (1995). Use of composites for
rehabilitation of steel structures - determination of bond
durability, Journal of Materials in Civil Engineering, Vol. 7,
No. 4, pp. 239245
Lenwari, A., Thepchatri, T. and Albrecht, P. (2006). Debonding
strength of steel beams strengthened with CFRP plates, Journal
of Composites for Construction, ASCE, Vol. 10, No. 1, pp. 6978.
Mays, G.C. and Hutchinson, A.R. (1992). Adhesives in Civil
Engineering, Cambridge University Press, New York, New York,
333 pp.
McKnight, S.H., Pierre, E.B., John, W., Gillespie, Jr. and Karbhari,
V.M. (1994). Surface preparation of steel for surface bonding
applications, Infrastructure: New Materials and Methods of
Repair, Proceedings of the 3rd Materials Engineering
Conference, ASCE, Kim D. Basham, Ed., Nov., San Diego,
California, pp. 11481155.
Miller, T.C., Michael, J.C, Dennis, R.M. and Jason, N.H. (2001).
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 817
APPENDIX: NOTATION
Efrp Modulus of elasticity of FRP
tfrp
thickness of FRP strip
shear stress
av
average shear stress
817
ASE 9-6-07_Schnerch
4/12/06
5:09 pm
Page 818
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 819
Abstract: The use of advanced composite materials to strengthen and repair existing
structures is increasing rapidly. One specific area in which the technique has recently
been introduced is the strengthening of metallic structures with bonded carbon-fibre
laminates. In this paper, the behaviour of composite steel-CFRP members is studied
experimentally and using FE analysis. A new type of test specimen has been developed
to examine the effect of various material parameters on the behaviour and strength of
bonded steel-CFRP elements. Finite element analysis of the tested elements was also
employed to examine the type and magnitude of interfacial stresses in the bond line of
these elements. Moreover, different types of fracture mode could be examined by
testing composite elements with different combinations of CFRP-laminates and
adhesives. The tested composite elements displayed different behaviour and a large
difference in strength and ductility could be observed.
Key words: bonding, strengthening, steel, CFRP, adhesive, interfacial, shear stresses.
1. INTRODUCTION
The use of composite materials for the strengthening and
repair of existing structures has increased significantly
during the last decade. The high durability and fatigue
resistance of these materials and their superior strengthto-weight ratio enable them to compete easily with other
traditional building materials in this field. After having
initially been used for concrete structures, the use of
composite materials for strengthening and repair work
has been extended to include timber and masonry and
more recently also metallic structures. Several field
applications have been reported in which steel and
wrought-iron structures (mostly bridges) have been
strengthened or repaired using adhesively bonded
carbon fibre laminates (CFRP) (Luke 2001; Miller et al.
2001). In some cases, pre-stressing of the laminates was
employed to obtain a more effective strengthening
819
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 820
Experimental and Numerical Investigation of the Behaviour and Strength of Composite Steel-CFRP Members
40
Adhesive B
Adhesive A
35
Stress (MPa)
30
25
20
15
10
5
0
0
0.002
0.004
0.006
0.008
0.01
Strain
Figure 1. Stress-strain curves obtained at room temperature for the adhesives used in the tests
820
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 821
E [GPa]
ult [MPa]
ult [x103]
14
6.5
32
24
3
7
0.27
0.35
Thickness
1.2
1.43
1.95
4.4
E//
//,ult
155
174
383
362
1932
1855
1252
(1252)*
*No reliable value for the ultimate tensile strength of this type of laminate
could be obtained. As it was made of the same material as B17, the same
tensile strength was assumed for both laminates.
yx
yy
Before
yielding
yx
yy
After
yielding
Figure 2. Schematic illustration of the principal load effects in a steel beam strengthened with bonded CFRP
821
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 822
Experimental and Numerical Investigation of the Behaviour and Strength of Composite Steel-CFRP Members
900
150
100
200
T = 10
25
90
36
R = 400
R = 400
Figure 3. The final shape and dimensions of the test specimen developed to study the behaviour and strength of composite steel-CFRP
Steel
CFRP
Adhesive
(a)
Steel
CFRP
Adhesive
(b)
Figure 4. The two finite-element models used in the development of the test specimen: (a) Model I with 20-node brick elements; (b) Model
II with 8-node brick elements
822
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 823
Table 3. The main features of the two models examined for FE analysis
of the test specimens
Model
No. of nodes
20-node brick
8-node brick
I
II
Element type
No. of elements
17207
73935
3324
65380
2
8
1400
FEM, P = 63,2 kN
FEM, P = 215 kN
Measured P = 63 kN
Measured P = 215 kN
1200
1000
800
600
400
200
0
50
100
150
200
250
300
Figure 5. Measured and calculated longitudinal strain (presented as stresses) in the laminate for specimen B17 (The values correspond to
two load levels before and after the yielding of the steel plate, namely 63 and 215 kN respectively)
823
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 824
Experimental and Numerical Investigation of the Behaviour and Strength of Composite Steel-CFRP Members
100
150
200
300
250
1
2
3
4
250
260
270
280
290
300
3
Analytical
5
5
6
Mid-adhesive
Steel-adhesive
Laminate-adhesive
Figure 6. The variation in shear stress along the bond line in specimen B12 analytical and FE results
EL tL
sinh( x )
( x ) = P
cosh(L)
Es ts + 2 EL tL
(1)
824
Ga 1
2
+
E s t s
ta EL tL
(2)
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 825
A12
B12
B17
B40
Mid-adhesive
Laminate-adhesive
max
max
max
max
max
max
4.4
4.3
7
8.8
6.8
5.8
8.8
11.0
5.9
5.3
8.2
9.6
6.3
5.4
8.1
10.4
6.6
6.4
10
12.0
4.1
3.8
6.0
8.3
6
Laminate-adhesive
Mid-adhesive
Steel-adhesive
0
275
280
285
290
295
300
305
2
Distance from specimen centre line (mm)
Figure 7. An example of the variation in interfacial normal stress near the ends of specimen B12 (P = 63 kN) as obtained
from the FE analysis
825
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 826
Experimental and Numerical Investigation of the Behaviour and Strength of Composite Steel-CFRP Members
50
100
150
200
250
300
4
6
8
10
P = 63 kN
P = 150 kN
12
P = 184 kN
14
16
Figure 8. The variation in interfacial shear stress along the bond line in specimen A12 at three different levels of applied load (The vertical
dotted lines show the progress of yielding from the middle of the specimen (x = 0))
low shear stresses are present in the bond line near the
middle of the specimen due to the change in the steel
plate width, as has previously been discussed. With the
successive yielding of the steel plate, higher shear
stresses are created near the middle of the specimen.
These are increased as the yielding in the plate
progresses (due to increased plastic strain) and the
location of maximum shear stress shifts towards the end
of the laminate. At higher loads (P = 184 kN), the
magnitude of maximum shear stress in the middle of the
specimen is higher than that at the ends of the laminate.
Debonding (cohesive or interfacial) might therefore
begin in the middle of the specimen before failure at the
end of the laminate is obtained. This observation was
also made for specimen A12.
6. TEST RESULTS
Figure 9 shows the load-displacement curves obtained for
the four composite elements. The results obtained from
the reference specimen (steel only) are also shown for
comparison. The same results are listed in Table 5. The
following observations can be made.
1. As one might expect, the delay in the onset of
yielding in the composite specimens in
comparison to the reference specimen is highest
for the laminates with a high E modulus (the
difference between B17 and B40 being attributed
to the higher thickness in the latter), see Table 2.
2. The highest degree of strengthening (i.e.
increase in the ultimate load) was obtained for
specimen B12. This specimen contains the
laminates with the lowest stiffness and the
lowest thickness. The ductility was also highest
826
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 827
Table 5. The strengthening effects of bonding CFRP to the steel plates in the tested specimen
Specimen
Steel
A12
B12
B17
B40
Yielding
load (kN)
Increase in
yielding load
(%)
Ultimate load
(kN)
Increase in
ultimate load
(%)
Deformation
at failure
(mm)
101
114
123
147
14
23
47
>57
140
186
251
247
157
33
80
76
12
12.51
3.34
7.51
2.10
0.95
Specimen
A12
Load
(kN)
Debonding
(adhes./steel) in the
middle of the specimen;
extended towards the
ends
Debonding at the ends
of the laminates
extending inwards
B12
B17
B40
Final failure
mode
Load
(kN)
150
184
244
230
210
247
157
300
B-17, Pmax = 247 kN
250
Load (kN)
200
150
100
50
0
0
6
8
Displacement (mm)
10
12
14
827
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 828
Experimental and Numerical Investigation of the Behaviour and Strength of Composite Steel-CFRP Members
Figure 10. The fracture surfaces at one end of the laminate in specimen B40: (a) Laminate surface; (b) Steel plate surface where the carbon
fibres are still bonded to the adhesive
30
25
20
Shear stress
15
10
x 1
FL(X)
FP(X)
0
0
50
100
150
5
B-12, x = 2.5 mm
10
P1
250
1 =
1
.E .t
x 1 L L
350
B-17, x = 1.5 mm
B-40, x = 3.0 mm
A-12, x = 3.0 mm
15
Applied load (kN)
Figure 11. Shear stresses at the end of the laminates in the four tested specimens as derived from strain measurements (It should be noted
that these stresses are derived at slightly different distances (x) from the end of the laminates, as the distances from the end of the laminate
to the first strain gauge (x) were slightly different. Notations are in accordance with Eqns 1 and 2)
828
ASE 9-6-08_Kliger
4/12/06
5:08 pm
Page 829
Final debonding
failure
Final debonding
failure
829
ASE 9-6-08_Kliger
4/12/06
5:09 pm
Page 830
Experimental and Numerical Investigation of the Behaviour and Strength of Composite Steel-CFRP Members
300
250
200
150
100
B-17
50
0
0.0
200.0
400.0
600.0
800.0
1000.0
B-12
B-40
1400.0
1600.0
A-12
1800.0
2000.0
Figure 13. The variation in maximum axial stress in the middle of the laminate in the four composite specimens
830
ASE 9-6-08_Kliger
4/12/06
5:09 pm
Page 831
APPENDIX: NOTATION
El E modulus of the laminate
Es E modulus of the steel
Ea E modulus of the adhesive
Ga Shear modulus of the adhesive
L
Laminate length
P
Applied load
ta
Thickness of the adhesive
tl
Thickness of the laminate
ts
Thickness of the steel plate
Strain
Constant
Shear stress
831
ASE 9-6-08_Kliger
4/12/06
5:09 pm
Page 832
4/12/06
5:08 pm
Page 833
Jin-Guang Teng
Editor-in-Chief
On behalf of the Local Editorial Team
833
4/12/06
5:08 pm
Page 834