See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268594611

Long-term behavior of FRP

Conference Paper · December 2001

DOI: 10.1061/40596(264)10

CITATIONS READS

42 348

2 authors:

György L. Balázs Dorian Borosnyoi-Crawley

Budapest University of Technology and Economics Hilti
181 PUBLICATIONS   1,747 CITATIONS    90 PUBLICATIONS   913 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Post-installed rebars for seismic applications View project

Crack width of reinforced concrete View project

All content following this page was uploaded by Dorian Borosnyoi-Crawley on 23 March 2022.

The user has requested enhancement of the downloaded file.

 Long-term behavior of FRP

György L. Balázs* and Adorján Borosnyói**

ABSTRACT
Fiber reinforced polymer (FRP) reinforcements have some superior characteristics to steel
reinforcements (especially they are non-corrosive), however, specific considerations may
be needed for their behavior under long-term influences. Durability and time dependent
mechanical characteristics of FRPs are essential parameters for design of concrete
structures reinforced or prestressed with FRP. Based on an extensive literature review,
present paper summarizes all aspects of FRP reinforcements considering durability
(including effects of alkalis, chloride ions, UV radiation, water, elevated temperature and
freeze/thaw cycles) as well as time dependent mechanical characteristics (such as creep,
relaxation, fatigue) and their changes due to environmental effects.

INTRODUCTION
In the last decades considerable deterioration of concrete structures due to corrosion has
turned the attention of researchers, designers, producers and owners on the application of
non-metallic (FRP), therefore non-corrosive reinforcements. Short-term characteristics of
FRPs can be more or less easily determined, however, long-term properties may require
specific considerations. Lot of papers have been published so far dealing with durability
and time dependent mechanical characteristics of embedded FRP reinforcements. Present
paper intends to summarize the state-of-the-art (Table 1).

DURABILITY OF FRP REINFORCEMENTS

FRP reinforcements are made of Fiber Reinforced Polymer composites. Composites have
at least two different phases such as load carrying unidirectional fibers and a resin matrix.
Load carrying fibers of FRP reinforcements can be carbon, aramid or glass.
Embedding matrix can be epoxy, polyester or vinyl ester resin. Long-term properties of
both constituents are to be investigated.

*
MSc (CE), PhD, Professor in Structural Engineering, Head of Department
**
MSc (CE), PhD candidate
Budapest University of Technology and Economics, Hungary
Department of Construction Materials and Engineering Geology

1
Table 1 Long-term influences related to the behavior of FRP reinforcements.
Environmental influences Mechanical properties
- alkalis - thermal effects - creep
- chloride ions - fire - relaxation
- water - freeze/thaw - fatigue
- UV radiation - combined effects

Effect of alkaline environment

Concrete is highly alkaline due to the high calcium hydroxide content of hardened cement
stone (pH 12.5 to 14) that needs special attention for durability of FRP reinforcements
(Fig. 1).
Carbon fibers cannot absorb liquids and are resistant to acid, alkali and organic
solvents, therefore, do not show considerable deterioration in any kind of harsh
environments (Machida 1993, Tokyo Rope 1993).
Deterioration of glass fibers in alkaline environment is well known. Therefore, the duty
of resins is of great importance in protecting glass fibers. Experimental studies of GFRP
reinforcement embedded in concrete or under accelerated aging tests in strong alkaline
solutions have demonstrated that glass fibers show significant degradation due to alkali,
independently of the of resin (Sen et al. 1993, Tannous and Saadatmanesh 1998, Uomoto
and Nishimura 1999). Decrease in tensile capacity can be in the range of 30 to 100
percent according to saturation and acting time. Best protection is ensured with vinyl ester
resin. Results of accelerated tests usually show more deterioration than tests with
embedded reinforcement. Rate of deterioration of glass fibers in alkaline environment is
highly dependent on the type of fibers.
Aramid fibers may also suffer deterioration in alkaline environment, however, to a less
degree than glass fibers and may depend on the actual fiber product (Uomoto and
Nishimura 1999). Decrease in tensile capacity can be 25-50 percent (Rostásy 1997).
Alkaline environment can deteriorate links between molecules of resins. Similarly to
the resistance against water absorption the alkaline resistance of vinyl ester resins is the
best while epoxy and polyester resins can give sufficient and poor resistance, respectively
(Machida 1993).
200
3

AFRP
Water absorption, kg/m

60
2
Tensile strength, kgf/mm

150 Kevlar/Epoxy
CFRP 40
100

50 20
GFRP Epoxy
in NaOH solution Carbon/Epoxy
0 Glass/Epoxy
0
50 0
100 0 5 10 15
Immersed duration, days Time, days
Fig. 1 Effect of alkali on tensile strength Fig. 2 Water absorption of FRPs
(after Uomoto-Ohga 1996) (after Piggott 1980)

2
Effect of chloride ions
Chloride ions can penetrate into concrete in marine environment or by using of de-icing
salts and can accelerate corrosion. In the presence of chloride ions the risk of corrosion of
FRP reinforcement may also take place. CFRP and AFRP reinforcements are insensitive
to chloride ions, however aramid fibers seem to be inapplicable in marine environment
due to difficulties initiated by swelling (Sen et al. 1998a, Sen et al. 1998b). Experimental
studies demonstrated that GFRP reinforcements can be seriously deteriorated in marine
environment or in presence of de-icing salts led to corrosion induced failure
(Saadatmanesh and Tannous 1997).

Effect of ultraviolet radiation

Polymeric materials can be considerably degraded by ultraviolet radiation (Piggott 1980).
Embedded FRP reinforcements are protected from direct sunlight, however, when stored
outdoors or applied as external reinforcement can be exposed to ultraviolet radiation. On
one hand, deterioration of GFRP and CFRP materials are attributed to the degradation of
resin matrix. After 2500 hours of exposure to direct sunlight decrease in tensile strength
and in Young’s modulus was less than 10 percent in the case of GFRP and negligible in
the case of CFRP (Kato et al. 1997). Aramid fibers themselves may deteriorate due to UV
radiation.

Effect of moisture and water

In fresh concrete contact of water and FRP reinforcement is evident. Material changes
associated with water are usually of the resins. Water can absorb into polymer chains and
can create weak chemical reactions causing considerable changes in characteristics (e.g.
strength, Young’s modulus, bond). These effects are mostly reversible, however, swelling
of resin can cause micro-cracks in the matrix that can initiate fiber debonding and higher
permeability. In general it can be stated that vinyl ester resins show the best resistance to
water absorption, epoxy resins can provide sufficient resistance, while polyester resins
usually have poor performance (Machida 1993).
For what concerning fibers, carbon and glass fibers cannot absorb water on the
contrary to aramid fibers (Uomoto and Nishimura 1999). Water absorption of aramid
fibers causes reversible decrease in tensile strength, Young’s modulus or relaxation and
irreversible decrease in fatigue strength (Piggott 1980). Decrease in characteristics of
AFRP due to water absorption is about 15-25% (Gerritse 1993). According to swelling of
AFRP reinforcement bond cracking can be induced by wet/dry cycles (e.g. in splash zones
of marine structures) that cause deterioration (Sen et al. 1998a). Fig. 2 indicates water
absorption capacity of various FRPs.

Combined and other effects

Effect of water and alkali absorption is clearly accelerated by elevated temperature. Rate
of degradation of GFRP and AFRP can be doubled by the change of temperature from
20°C to 60°C (Rostásy 1997, Tannous and Saadatmanesh 1998).
Degradation of AFRP due to ultraviolet radiation can be accelerated by wet/dry cycles
of marine splash zone (Uomoto and Ohga 1996).
As it is well known, carbonation of concrete has important role on the corrosion of
ordinary reinforcement. Research work on the effect of carbonation on durability of FRP

3
reinforcement is very limited. Available data show large scatter, however, carbonation
seems to have no effect on the durability of FRP reinforcements (Sheard et al. 1997).

THERMAL ACTIONS ON FRP REINFORCEMENTS

General considerations
In the case of FRPs thermal actions can influence both mechanical characteristics and
bond behavior. Coefficient of thermal expansion (CTE) of fibers, resins, FRPs and
concrete are considerably different as indicated in Table 2. In the longitudinal direction
FRPs have lower or nearly identical CTEs than that of concrete, however, in the
transverse direction – governed mostly by the resin – reach 5 to 8-times higher values. In
specific cases, when high temperature variation takes place the large difference between
CTEs can lead to high radial pressure on the surface of the reinforcement that can cause
longitudinal splitting of concrete cover. It reflects the importance of sufficient concrete
cover especially if AFRP reinforcement is applied. Critical concrete cover (whenever
splitting occurs) of AFRP tendons with sand coated surface was found to have 2.8×rebar
diameter (Taerwe and Pallemans 1995). Authors found sufficient concrete cover of
2.5×rebar diameter to CFRP tendons with sand coated surface for pretensioned
application with ten hours heat curing at a maximum temperature of 75°C (Balázs and
Borosnyói 2001).

Table 2 Coefficients of thermal expansion of fibers, resins, FRPs and concrete

Coefficient of thermal expansion,
Material ×10-6 1/K
longitudinal transverse
carbon fiber –0.9…+0.7 8…18
aramid fiber –6.0…–2.0 55…60
glass fiber 5…15 5…15
resins 60…140
CFRP –0.5…1.0 20…40
AFRP –2.0…–1.0 60…80
GFRP 7…12 9…20
concrete 6…13

Thermal effects can also have influence on aging of resins, in this way the residual
strength of FRP reinforcements. Experimental data on change of long-term residual
strength of FRPs due to thermal cycles are not available.
Tensile strength and Young’s modulus of FRPs may be influenced by the temperature.
Under service temperature of concrete structures (from -20 to +60°C) the reduction in
Young’s modulus of CFRP is negligible, however, slight reduction of Young’s moduli of
AFRP and GFRP can be observed (Rostásy 1996). Change in tensile strength is attributed
to only higher temperatures when deterioration of resin occurs.

Effect of elevated temperature

Polymeric materials are usually flammable or harming in the case of fire, therefore,
basically resin determines the temperature/fire resistance of FRPs. Resins soften, melt or
catch fire above 150-200°C. Fibers themselves are more or less able to resist to higher

4
temperatures: aramid to 200°C, glass to 300-500°C while carbon in non-oxidizing
environment up to 800-1000°C (Rostásy 1996).
Experiments demonstrated that – in addition to fiber type – the surface configuration of
FRP reinforcement has a very important role on fire resistance and behavior under
elevated temperature. Due to the temperature independence of carbon fibers themselves,
CFRP shows the most favorable behavior. Decrease in tensile capacity of CFRP strands
and braided tendons at 400°C was about 20 percent on the contrary to CFRP round rods
that had no deterioration (Tanao et al. 1997). Decrease in tensile capacity of AFRP
reinforcements at 400°C was about 60 percent independently of the configuration (Tanao
et al. 1997). It has to be also noticed that CFRP specimens examined after cooling to room
temperature from 300°C showed no decrease in tensile capacity (Tokyo Rope 1993).
Deflection of concrete beams reinforced with FRPs under high-temperature loading is
also attributed to the surface configuration of reinforcement used (Sakashita 1997).
Deflections of GFRP reinforced beams and beams reinforced with braided AFRP or
braided CFRP tendons were greater than that of steel reinforced member. Good results
were produced by beams reinforced with spirally wounded or straight CFRP rods with
deflections of about one-fifth that of steel reinforced member.

Effect of freezing and thawing

In a lot of civil engineering applications reinforced concrete members are subjected to
high number of freezing/thawing cycles (mostly combined with water and chloride ion
penetration into the concrete). However, experimental data on the influence of such
effects on the durability of FRP reinforcements is very limited. Due to freezing and
thawing cycles (combined with water and chloride ion diffusion) degradation of fibers,
resin and interfacial bond is possible. According to micro cracking of concrete under
freezing and thawing cycles bond between concrete and FRP can be also influenced.

TIME DEPENDENT MECHANICAL CHARACTERISTICS OF FRP REINFORCEMENTS

Time dependent mechanical characteristics of FRP reinforcements can be different from
that of ordinary steel reinforcements due to not only the different materials but also to the
composite behavior: time dependent phenomena can take place within the material phases
(fiber, matrix) and on the interface (bond, delamination).

Creep
Two different issues have to be distinguished related to creep: the creep strain under
sustained load and the long-term tensile strength under sustained load (often called as
stress rupture, residual strength or creep rupture strength).
CFRP shows excellent creep behavior in terms of creep strain: in general it can be
stated that creep strain of CFRP at room temperature and humidity remains under 0.01%
after 3000 hours at a tensile stress of even 80 percent of the tensile strength (Machida
1993, Saadatmanesh and Tannous 1999a, Tokyo Rope 1993). AFRP and GFRP give
much higher creep strain than CFRP: 0.15-1.0% for AFRP and 0.3-1.0% for GFRP under
conditions verified above (Gerritse 1993, Machida 1993, Piggott 1980).
Long-term tensile strengths of FRPs have a very important role in defining allowable
sustained service stress in structural members. After 106 hours long-term tensile strength
of CFRP, AFRP and GFRP can be estimated as 80-95%, 50-70% and 40-70% of the

5
 short-term tensile strength, respectively, considerably depending on the product examined
(Fig. 3) (Ando et al. 1997, Machida 1997, Rostásy 1997, Uomoto et al. 1995, Wolff and
Miesseler 1993, Yamaguchi et al. 1997).

Max. stress (N/mm )

2
100
Stress ratio, %

CFRP
1200
AFRP
80
800

60 CFRP steel GFRP(S-glass)

AFRP 400
GFRP(E-glass)
GFRP (E-glass)
GFRP (S-glass)
40 0
-2 0 2 4 6 2 3 4 5 6 7
10 10 10 10 10 10 10 10 10 10 10
Time to failure, hours Number of cycles
Fig. 3 Creep rupture of FRPs Fig. 4 S – N curves of FRPs
(after Machida 1997) (after Machida 1997)

Relaxation
Relaxation of CFRP, AFRP and GFRP tendons after 50 years of loading can be estimated
as 2-10%, 11-25% and 4-14 %, respectively, depending on the initial tensile stress (Ando
et al. 1997, Machida 1997, Saadatmanesh and Tannous 1999a, 1999b, Tokyo Rope 1993,
Wolff and Miesseler 1993). Relaxation after 1000 hours can be estimated as 0.5-1.0%,
5.0-8.0% and 1.8-2.0%, respectively.

Fatigue
Although limited number of fatigue tests of FRP reinforcements is available some obvious
tendencies can be verified. Fatigue resistance of GFRP is usually less than that of
prestressing steel (Wolff and Miesseler 1993, Machida 1993, 1997). AFRP tendons show
similar or higher fatigue strength than prestressing steel with especially higher stress
amplitudes (Gerritse 1993). Both GFRP and AFRP show similar dependency of stress
level on fatigue strength like prestressing steel does (Uomoto et al. 1995). CFRP has
excellent fatigue strength: 3 to 4 times higher than that of prestressing steel (Tokyo Rope
1993, Saadatmanesh and Tannous 1999a, 1999b). Fatigue of CFRP seems to be
independent of stress level and amplitude (Uomoto et al. 1995). Fig. 4 indicates S-N
curves of various FRPs.

Environmental effects on time dependent mechanical characteristics

Environmental effects can influence considerably time dependent mechanical
characteristics of FRPs. Due to ingress of water, alkali or chloride ions into resins all
types of FRPs are expected to deteriorate in their characteristics. Limited experimental
data are available, however, basic assumptions can be drawn (Gerritse 1993, Rostásy
1997, Saadatmanesh and Tannous 1999a, 1999b). Relaxation of AFRP and CFRP is
increased with 10-30 percent by absorption of water or solutions of chlorides or alkali.

6
Creep of AFRP and CFRP increases with 30-40 percent by alkaline environment. Long-
term tensile strength of AFRP can be considerably decreased by alkali.

CONCLUSIONS
Based on an extensive literature review the following conclusions can be drawn
considering long-term behavior of FRP reinforcements:
• In general, FRP reinforcements show better creep, relaxation and fatigue behavior than
steel reinforcements. However, influence of environmental effects on the time
dependent mechanical properties needs further considerations.
• Liquids (water, alkali and salt solutions) can diffuse into resins of FRPs that can cause
deterioration of mechanical characteristics. Least diffusion can be found in vinyl ester
resins. Glass fibers tend to deteriorate by alkali and chloride ions resulted in corrosion-
induced failure. Aramid fibers can absorb water that cause reversible loss in strength,
however, can cause bond failure by splitting due to swelling. Aramid fibers also may
deteriorate in alkaline environment. Carbon fibers are resistant to any kind of harsh
environment. UV radiation can be harmful on aramid fibers and resins.
• Coefficients of thermal expansion of FRPs in the longitudinal direction are smaller, in
the transverse direction 5 to 8 times greater than that of concrete. This may have an
influence on the minimum concrete cover against splitting. Behavior of FRPs under
elevated temperature is governed by not only the resin and the fibers but also the
surface configuration of tendons.

REFERENCES
Ando, N. – Matsukawa, H. – Hattori, A. – Mashima, M. (1997) “Experimental Studies on
the Long-Term Tensile Properties of FRP Tendons”, Proc. 3rd Int. Symp. FRPRCS-3,
JCI, 1997, Vol. 2., pp. 203-210.
Balázs, G. L. – Borosnyói, A. (2001) “Prestressing with CFRP Tendons”, Proceedings of
the UEF International Conference on High Performance Materials in Bridges and
Buildings, July 29 – August 3, 2001, Kona, Hawaii (in printing)
Gerritse, A. (1993) “Aramid-based prestressing tendons”, Alternative Materials for the
Reinforcement and Prestressing of Concrete, Ed. Clarke, Chapman & Hall, London,
1993, pp. 172-199.
Kato, Y. – Yamaguchi, T. – Nishimura, T. – Uomoto, T. (1997) “Computational Model
for Deterioration of Aramid Fiber by Ultraviolet Rays”, Proc. 3rd Int. Symp. FRPRCS-3,
Japan Concrete Institute, 1997, Vol. 2., pp. 163-170.
Machida, A. (1993) “State-of-the-Art Report on Continuous Fiber Reinforcing Materials”,
JSCE, Tokyo, 1993.
Machida, A. (1997) “Recommendation for Design and Construction of Concrete
Structures Using Continuous Fiber Reinforcing Materials”, JSCE, Tokyo, 1997.
Piggott, M. R. (1980) “Load Bearing Fibre Composites”, Pergamon Press, Oxford, 1980.
Rostásy, F. (1996) “State-of-the-Art Report on FRP Materials”, FIP Report, Draft, 1996
(unpublished)
Rostásy, F. (1997) “On Durability of FRP in Aggressive Environment”, Proc. 3rd Int.
Symp. FRPRCS-3, JCI, 1997, Vol. 2., pp. 107-114.
Saadatmanesh, H. – Tannous, F. E. (1997) “Durability of FRP Rebars and Tendons”,
Proc. 3rd Int. Symp. FRPRCS-3, JCI, 1997, Vol. 2., pp. 147-154.

7
 Saadatmanesh, H. – Tannous, F. E. (1999) “Relaxation, Creep, and Fatigue Behavior of
Carbon Fiber Reinforced Plastic Tendons”, ACI Materials Journal, V. 96, No. 2, March-
April 1999, pp. 143-153.
Saadatmanesh, H. – Tannous, F. E. (1999) “Long-Term Behavior of Aramid Fiber
Reinforced Plastic (AFRP) Tendons”, ACI Materials Journal, V. 96, No. 3, May-June
1999, pp. 297-305.
Sakashita, M. – Masuda, Y. – Nakamura, K. – Tanao, H. – Nishida, I. – Hashimoto, T.
(1997) “Deflection of Continuous Fiber Reinforced Concrete Beams Subjected to
Loaded Heating”, Proc. 3rd Int. Symp. FRPRCS-3, JCI, 1997, Vol. 2., pp. 51-58.
Sen, R. – Mariscal, D. – Shahawy, M. (1993) “Durability of Fiberglass Pretensioned
Beams”, ACI Structural Journal, V. 90, No. 5, September-October 1993, pp. 525-533.
Sen, R. – Shahawy, M. – Rosas, J. – Sukumar, S. (1998) “Durability of Aramid
Pretensioned Elements in a Marine Environment”, ACI Structural Journal, V. 95, No. 5,
September-October 1998, pp. 578-587.
Sen, R. – Shahawy, M. – Rosas, J. – Sukumar, S. (1998) “Durability of Carbon
Pretensioned Elements in a Marine Environment”, ACI Structural Journal, V. 95, No. 6,
November-December 1998, pp. 716-724.
Sheard, P. – Clarke, J. – Dill, M. – Hammersley, G. – Richardson, D. (1997)
“EUROCRETE – Taking Account of Durability for Design of FRP Reinforced Concrete
Structures”, Proc. 3rd Int. Symp. FRPRCS-3, JCI, 1997, Vol. 2., pp. 75-82.
Taerwe, L. – Pallemans, I. (1995) “Force Transfer of AFRP Bars in Concrete Prisms”,
Proc. 2nd Int. Symp. FRPRCS-2,, E & FN Spon, London, 1995, pp. 154-163.
Tanao, H. – Masuda, Y. – Sakashita, M. – Oono, Y. – Nonomura, K. – Satake, K. (1997)
“Tensile Properties at High Temperatures of Continuous Fiber Bars and Deflections of
Continuous Fiber Reinforced Concrete Beams Under High-Temperature Loading”,
Proc. 3rd Int. Symp. FRPRCS-3, JCI, 1997, Vol. 2., pp. 43-50.
Tannous, F. E. – Saadatmanesh, H. (1998) “Environmental Effects on the Mechanical
Properties of E-Glass FRP Rebars”, ACI Materials Journal, V.95, No. 2, March-April
1998, pp. 87-100.
Tokyo Rope (1993), “Technical Data on CFCC®”, Tokyo Rope Mfg. Co., Ltd. Manual,
Tokyo, October 1993.
Uomoto, T. – Nishimura, T. – Ohga, H. (1995) “Static and Fatigue Strength of FRP Rods
for Concrete Reinforcement”, Proc. 2nd Int. Symp. FRPRCS-2, E & FN Spon, London,
1995, pp. 100-107.
Uomoto, T. – Ohga, H. (1996) “Performance of Fiber Reinforced Plastics for Concrete
Reinforcement”, Proc. 2nd ACMBS Conference, Montreal, CSCE, 1996, pp. 125-131.
Uomoto, T. – Nishimura, T. (1999) “Deterioration of Aramid, Glass and Carbon Fibers
Due to Alkali, Acid and Water in Different Temperatures”, Proc. 4th Int. Symp.
FRPRCS-4, ACI SP-188, American Concrete Institute, 1999, pp. 515-522.
Wolff, R. – Miesseler, H.-J. (1993) “Glass-fibre prestressing system“, Alternative
Materials for the Reinforcement and Prestressing of Concrete, Ed. Clarke, Chapman &
Hall, London, 1993, pp. 127-152.
Yamaguchi, T. – Kato, Y. – Nishimura, T. – Uomoto, T. (1997) “Creep Rupture of FRP
Rods”, Proc. 3rd Int. Symp. FRPRCS-3, JCI, 1997, Vol. 2., pp. 179-186.

View publication stats