Teng2012.pdf

Journal of Constructional Steel Research 78 (2012) 131–143

Contents lists available at SciVerse ScienceDirect

Journal of Constructional Steel Research

Review

Strengthening of steel structures with fiber-reinforced polymer composites J.G. Teng a,⁎, T. Yu b, D. Fernando c a b c

Department of Civil and Structural Engineering, The Hong Kong Polytechnic University, Hong Kong, China School of Civil, Mining & Environmental Engineering, Faculty of Engineering, University of Wollongong, Northfields Avenue, Wollongong, NSW 2522, Australia Institute of Construction and Infrastructure Management (IBI), Department of Structural, Environmental and Geomatic Engineering (D-BAUG), ETH Zürich, Zürich, Switzerland

a r t i c l e

i n f o

Article history: Received 27 February 2012 Accepted 29 June 2012 Available online 30 July 2012 Keywords: Steel structures FRP composites Strengthening Retrofit Composite materials

a b s t r a c t Over the past two decades, fiber-reinforced polymer (FRP) composites have gradually gained wide acceptance in civil engineering applications due to their unique advantages including their high strength-to-weight ratio and excellent corrosion resistance. In particular, many possibilities of using FRP in the strengthening and construction of concrete structures have been explored. More recently, the use of FRP to strengthen existing steel structures has received much attention. This paper starts with a critical discussion of the use of FRP in the strengthening of steel structures where the advantages of FRP are appropriately exploited. The paper then provides a critical review and interpretation of existing research on FRP-strengthened steel structures. Topics covered by the review include steel surface preparation for adhesive bonding, selection of a suitable adhesive, bond behavior between FRP and steel and its appropriate modeling, flexural strengthening of steel beams, fatigue strengthening of steel structures, strengthening of thin-walled steel structures against local buckling, and strengthening of hollow or concrete-filled steel tubes through external FRP confinement. The paper concludes with comments on future research needs. © 2012 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3.

4.

5. 6.

7. 8. 9.

Introduction . . . . . . . . . . . . . . . . . . . . . . . Appropriate use of FRP in the strengthening of steel structures Bond behavior between FRP and steel . . . . . . . . . . . 3.1. General . . . . . . . . . . . . . . . . . . . . . . 3.2. Adhesion failure . . . . . . . . . . . . . . . . . . 3.3. Bond behavior . . . . . . . . . . . . . . . . . . . 3.3.1. Bond strength . . . . . . . . . . . . . . . 3.3.2. Bond-slip relationship . . . . . . . . . . . Flexural strengthening of steel beams . . . . . . . . . . . . 4.1. Plate end debonding . . . . . . . . . . . . . . . . 4.2. Intermediate debonding . . . . . . . . . . . . . . . 4.3. Other issues . . . . . . . . . . . . . . . . . . . . Fatigue strengthening . . . . . . . . . . . . . . . . . . . Strengthening of steel structures against local buckling . . . 6.1. Buckling induced by high local stresses . . . . . . . . 6.2. Buckling induced by other loads . . . . . . . . . . . FRP confinement of hollow steel tubes . . . . . . . . . . . FRP confinement of concrete-filled steel tubes . . . . . . . . Concluding remarks . . . . . . . . . . . . . . . . . . . . 9.1. Steel surface treatment . . . . . . . . . . . . . . . 9.2. Selection and formulation of adhesives . . . . . . . . 9.3. Bond behavior and debonding failures . . . . . . . . 9.4. Fatigue strengthening . . . . . . . . . . . . . . . . 9.5. FRP confinement of tubular structures . . . . . . . . 9.6. Other issues . . . . . . . . . . . . . . . . . . . .

⁎ Corresponding author. Tel.: +852 2766 6012. E-mail address: [email protected] (J.G. Teng). 0143-974X/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jcsr.2012.06.011

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . .

132 132 133 133 133 134 135 135 136 136 137 137 137 138 138 138 138 139 140 141 141 141 141 141 141

132

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Fiber-reinforced polymer (FRP) composites are formed by embedding continuous fibers in a polymeric resin matrix which binds the fibers together. Common fibers used in FRP composites include carbon, glass, aramid and basalt fibers while common resins are epoxy, polyester, and vinyl ester resins. The most widely used FRP composites are glass fiber-reinforced polymer (GFRP) composites and carbon fiber-reinforced polymer (CFRP) composites, while aramid fiber-reinforced polymer (AFRP) composites and basalt fiber-reinforced polymer (BFRP) composites are less frequently used. A useful general background to the composition of these materials and their mechanical properties can be found in Refs. [1–4]. Fig. 1 shows typical stress–strain responses of FRP composites in contrast with that of mild steel, where it is clearly seen that FRP composites exhibit a linear elastic stress–strain behavior before brittle failure by rupture. This linear–elastic–brittle stress–strain behavior has important implications for the structural use of FRP composites in civil engineering applications. FRP composites possess several advantages over steel, the most salient of which are their high strength-to-weight ratio and excellent corrosion resistance. The structural use of FRP in civil infrastructure is generally based on the exploitation of these advantages. In particular, FRP, being a material of high tensile strength, can generally be used to its greatest advantages, when combined with concrete which is strong in compression but poor in tension. Therefore, the use of FRP in concrete structures has been a major focus of existing research [2,4–6]. Such applications include the external bonding of FRP to concrete structures for strengthening purposes, concrete structures reinforced or prestressed with FRP, concrete-filled FRP tubes as columns and piles, as well as FRP-concrete hybrid beams/bridge decks. More recently, the use of FRP composites in combination with steel, particularly in the strengthening of steel structures, has received much attention. This paper first examines applications where the use of FRP in the strengthening of steel structures presents significant advantages and then provides a critical review and interpretation of existing research on FRP-strengthened steel structures.

reducing disturbance to services and traffic. Another significant advantage of FRP, which applies only to FRP laminates formed via the wet lay-up process, is the ability of such FRP laminates to follow curved and irregular surfaces of a structure. This is difficult to achieve using steel plates. A third advantage of FRP is that its material properties in different directions can be tailored for a particular application. As a result of the second and third advantages, FRP jackets with fibers oriented only or predominantly in the circumferential direction can be used to confine steel tubes/shells or concrete-filled steel tubes to delay or eliminate local buckling problems in steel tubes/shells, thereby enhancing the strength and/or seismic resistance of such structures (e.g. [7–12]). The method of FRP confinement is attractive not only in the strengthening of steel tubular structures, but also in the construction of new tubular columns. The combination of adhesive bonding with shape flexibility makes bonded wet lay-up FRP laminates an attractive strengthening method in a number of applications. Needless to say, steel plates can also be adhesively-bonded but bonding is less attractive for steel plates due to their heavy weight and inflexibility in shape. Furthermore, for the same tensile capacity, a steel plate has a much larger bending stiffness than an FRP laminate so a steel plate leads to higher peeling stresses at the interface between the steel plate and the steel substrate. It is also easier to anchor FRP laminates to a steel member by wrapping FRP jackets around the steel member. Steel plates can also be attached by welding to strengthen existing steel structures, but the bonding of FRP laminates is superior to the welding of steel plates in the following situations: (1) Bonding of FRP laminates for enhanced fatigue resistance has the advantage that the strengthening process does not introduce new residual stresses; (2) In certain applications (e.g. oil storage tanks and chemical plants) where fire risks must be minimized, welding needs to be avoided when strengthening a structure; bonding of FRP laminates is then a very attractive alternative; (3) High-strength steels suffer significant local strength reductions in heat-affected zones of welds, so bonded FRP laminates offer an ideal strength compensation method [13].

2. Appropriate use of FRP in the strengthening of steel structures Since steel is also a material of high elastic modulus and strength, the use of FRP in strengthening steel structures calls for innovative exploitations of the advantages of FRP. The main advantage of FRP over steel in the strengthening of steel structures is its high strength-to-weight ratio, leading to ease and speed of transportation and installation, thus 3000

Intermediate modulus CFRP

Stress (MPa)

2500

High strength CFRP

141 141

The use of both CFRP and GFRP to strengthen steel structures has been explored. For the strength enhancement of steel structures, CFRP is preferred over GFRP due to the much higher elastic modulus of the former. In particular, when the enhancement of buckling resistance is the aim, the use of high or ultra-high modulus CFRP is very attractive. Table 1 shows the properties of pultruded CFRP plates supplied by SIKA; these three types of CFRP plates are referred to herein as high strength, intermediate modulus and high modulus plates respectively and their stress–strain curves are illustrated in Fig. 1. By contrast, for the confinement of steel tubes, particularly when ductility

2000 Table 1 Properties of SIKA CFRP platesa.

High modulus CFRP

1500 1000

Mild Steel

500 0

Product

Elastic modulus (GPa)

Tensile strength (MPa)

Ultimate strain (%)

Sika CarboDur S (high strength CFRP) Sika CarboDur M (intermediate modulus CFRP) Sika CarboDur H (high modulus CFRP)

165

2800

1.70

210

2400

1.20

300

1300

0.45

GFRP

0

0.5

1

1.5

2

2.5

Strain (%) Fig. 1. Typical FRP and mild steel stress–strain curves.

3

a

Extracted from the manufacturer's product data sheet.

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

enhancement is the main aim, GFRP is more attractive as it is cheaper and offers a greater strain capacity (>2%). An issue to note is that of galvanic corrosion when steel is in direct contact with CFRP [14,15], so a layer of GFRP has been advised to be sandwiched between them by some researchers (e.g. [15]). A detailed discussion of the issue of galvanic corrosion is given in Ref. [16]. Since FRP composites, particularly CFRP composites are an expensive material, in all applications, the amount of FRP material required should be minimized. For this reason, where the amount of FRP material required is small by nature of the problem (e.g. local strengthening under a concentrated force), FRP strengthening is more likely to be attractive. 3. Bond behavior between FRP and steel 3.1. General Similar to the structural use of FRP in concrete structures, the structural use of FRP with steel can be classified into two categories: (a) bond-critical applications where the interfacial shear stress transfer function of the adhesive layer that bonds the steel and the FRP together is crucial to the performance of the structure; and (b) contact-critical applications where the FRP and the steel need to remain in contact for effective interfacial normal stress transfer which is crucial to ensure the effectiveness of the FRP reinforcement. The use of FRP in the strengthening of steel structures provides good examples for both categories: externally bonded FRP reinforcement for the flexural strengthening of steel beams falls into the first category, while confinement of concrete-filled steel tubular members with FRP jackets belongs to the second category. In all bond-critical applications, the interfacial behavior between FRP and steel is of critical importance in determining when failure occurs and how effectively the FRP is utilized. An important difference in bond behavior between FRP-strengthened concrete structures and FRP-strengthened steel structures is the exact location of interfacial failure: for the former interfacial failure generally occurs in the substrate concrete and the design theory has been developed with this nature of interfacial failure implicitly or explicitly assumed; for the latter interfacial failure cannot possibly occur in the substrate steel due to the much higher tensile strength of steel than that of adhesives. As a result, for the latter, interfacial failure can only occur within the adhesive layer (i.e. cohesion failure) or at the material interfaces (adhesion failure) between the steel and the adhesive (referred to as the “steel/adhesive interface” hereafter) or between the adhesive and the FRP (referred to as the “FRP/adhesive interface” hereafter). A summary of possible failure modes is shown in Fig. 2. If adhesion failure controls the strength of FRP-strengthened steel structures, then the interfacial bond strength depends on how the steel surface and the FRP surface are treated as well as the bond capability of the adhesive. As adhesion failure depends on the method and degree of surface treatment, especially to the steel substrate, which is difficult to control on site, the development of a design theory becomes much more involved. This important issue has not been given adequate attention in previous studies, but has been focused Interlaminar failure of FRP

on in some recent research [17]. The authors thus strongly believe that in FRP-strengthened steel structures, interfacial failure should occur within the adhesive layer in the form of cohesion failure (Fig. 3), and a proper surface treatment procedure together with an appropriate adhesive should be used to ensure that such cohesion failure is critical. 3.2. Adhesion failure In an FRP-to-steel bonded joint, adhesion failure may occur at the steel/adhesive interface or at the FRP/adhesive interface. However, adhesion failure at the FRP/adhesive interface seldom occurs when the FRP is formed and applied to the structure via a wet lay-up process on site; when a pultruded FRP plate/strip is used, such failure can generally be avoided through the use of a peel-ply which is removed prior to bonding to ensure a clean and rough FRP surface for bonding [15] or by abrading and cleaning the FRP surface before bonding. By contrast, failure at the steel/adhesive interface is much more likely to happen. For various reasons, the treatment and characterization of steel surfaces for adhesive bonding has received much research attention [18–22]. The adhesion strength of a steel/adhesive interface results from both chemical bonding and mechanical bonding between the two adherends [18,21,23]. It is evident that a strong steel/adhesive interface requires the adhesive to be in intimate contact with the steel surface. This generally means that the adhesive should have a sufficiently low viscosity so that it can flow easily over the surface and fill the pores [24], and that the steel surface should be clean and should have a sufficiently large surface energy so that it can be easily wetted [20,21]. When the two adherends are in intimate contact, the strength of chemical bonding depends mainly on the chemical composition of the steel surface and that of the adhesive and whether they are chemically compatible [21]. By contrast, apart from the properties of the adhesive, the strength of mechanical bonding depends mainly on the roughness and topography of the steel surface; roughening the surface can significantly enhance the strength of mechanical bonding [23,25], but it may also reduce the level of contact between the two adherends [26,27]. Therefore, the three main properties of a steel surface, namely, surface energy, surface chemical composition and surface roughness and topography, are often used to characterize the capacity of a surface for bonding [20,28–30]. Existing approaches of steel surface treatment generally aim to enhance the two bonding mechanisms (i.e. chemical bonding and mechanical bonding) by: (1) cleaning the surface; (2) changing the properties of the surface. The most popular approaches include solvent cleaning and mechanical abrasion through grit blasting or using other tools (e.g. wire brushes, abrasive pads and wheels, and needle guns) [15,21]. Solvent cleaning removes the contaminants from the surface (e.g. grease, oil and water) but does not change the surface properties, so it alone only has a limited effect on the adhesion strength [20]. It is however a necessary step of any surface treatment process and should

FRP Rupture

CFRP Adhesive

Adhesion failure at FRP/adhesive interface

Cohesion failure in adhesive Steel Adhesion failure at steel/adhesive interface Fig. 2. Possible failure modes of FRP-to-concrete bonded joints.

133

Fig. 3. Surface of the FRP plate after cohesion failure.

134

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

normally be conducted at the beginning of the process [15,22]. It is important to use a volatile solvent (e.g. acetone) so that the contaminants on the surface (and hence their negative effects on the adhesion strength) are minimized [4,18]. Mechanical abrasion roughens the surface and removes the weak surface layer (e.g. oxide layer) which is chemically inactive [20,21], so that the surface in contact with the adhesive is sufficiently rough, clean and chemically active. Among various mechanical abrasion approaches, grit blasting appears to be the most effective [15,20,31,32] and is recommended by some existing guidelines on the FRP strengthening of metallic structures [22,33]. Tests recently conducted by Teng et al. [17] showed that with the four types of different adhesives used in their study, adhesion failure was avoided when the steel surface was grit-blasted prior to bonding. The grit used in grit blasting may be made of different materials and have different particle sizes. Existing studies [17,20,34] have shown that grit blasting can modify the chemical composition of the surface by introducing grit residues to the surface, so it is important to choose a grit material which is chemically compatible with the adhesive. The particle size of grit may have a pronounced effect on surface energy and surface roughness, but the limited existing studies [17,20] have revealed that within the range of grit particle sizes examined in these studies (i.e. from 0.125 mm to 0.5 mm), the effect of particle size on adhesion strength is limited. During the grit blasting process, fine abrasive dust is produced and becomes additional surface contaminants [15]. Therefore, it is important to clean the surface again after grit blasting. Hollaway and Cadei [15] suggested to remove the fine dust using dry-wiping or using a vacuum head instead of solvent wiping as they believed that solvent wiping is capable of only partial removal of the dust and is likely to redistribute the remaining dust on the surface. El Damatty and Abushagur [35] however showed that with the use of an excessive amount of solvent, the dust can be completely removed and a clean surface can be produced. After surface treatment, an adhesive/primer should be applied as soon as possible to avoid any contamination of the surface or formation of weak oxide layers on the surface [36]. Cadei et al. [33] recommended that the period between grit blasting and adhesive/primer application should not exceed 2 h, while Schnerch et al. [22] suggested a more practical maximum period of 24 h for the application of adhesive. Apart from the adoption of an appropriate surface treatment procedure, it is also important to characterize the surface to determine whether a sufficient adhesion strength can be developed. The following methods are available for surface characterization: (a) a VCA (video contact angle) device can be employed to obtain contact angle measurements from which the surface energy can be evaluated; (b) an SEM/EDX (scanning electron microscopy/energy dispersive x-ray) system can be used to measure the surface chemical composition; and (c) a profilometer can be used to measure the surface roughness and topography [17]. By using these devices, Teng et al. [17] showed that the characteristics of surfaces are consistent after being grit-blasted using the same grit, which suggest the possibility of developing a standard preparation process to ensure a good surface with a sufficient adhesion strength. 3.3. Bond behavior Similar to reinforced concrete (RC) structures strengthened with externally bonded FRP reinforcement, interfacial debonding failures also control the load-carrying capacity of steel structures strengthened with externally bonded FRP reinforcement in many cases. A simplysupported steel beam strengthened in flexure using a bonded soffit FRP plate is a typical bond-critical case where the following two distinct debonding failure modes can occur: (1) intermediate debonding; and (2) plate end debonding. In the former mode, debonding initiates away from the FRP plate ends and at a location where high interfacial shear stresses arise from either the presence of a defect (e.g. crack) or local yielding of the steel substrate. In the latter mode, debonding

initiates at an FRP plate end due to a combination of high interfacial shear and peeling (normal) stresses. Intermediate debonding has been observed in laboratory tests on FRP-strengthened steel beams with or without an initial defect (e.g. [37–39]) and steel sections strengthened with FRP against local buckling (e.g. [40]), while plate end debonding has been observed in laboratory tests on flexurally-strengthened steel beams (e.g. [41]) and on steel sections strengthened against end bearing loads (e.g. [42,43]) or other loads inducing local buckling (e.g. [44]). It has been widely recognized [45–50] that in order to understand and model debonding failures, the bond behavior between the substrate material and the bonded FRP reinforcement needs to be studied, commonly through pull tests on simple bonded joints (Fig. 4(a)) [48,51–53]. In a pull test, the adhesive layer is primarily subjected to interfacial shear stresses and debonding is caused by Mode II fracture in fracture mechanics terms. The interfacial behavior of such simple bonded joints is similar to that of an FRP-to-steel interface in a beam where intermediate debonding is critical, as interfacial shear stresses dominate the debonding process in both cases. This interfacial shear behavior is also an important basis for understanding the behavior of FRP-to-steel interfaces subjected to combined shear stresses and peeling stresses. Different from FRP-to-concrete bonded joints where the concrete is usually the weak link, the adhesive is the weak link in FRP-to-steel bonded joints, provided that adhesion failure at the steel/adhesive interface and the FRP/adhesive interface is avoided by careful selection of the adhesive and appropriate surface preparation of the steel and the FRP. As a result, the behavior of FRP-to-steel bonded joints is similar to that of steel-to-steel bonded joints, so available tests on the latter are also included in the discussion below to supplement the limited available studies on FRP-to-steel bonded joints [35,51,53–65]. In addition, existing studies on FRP-to-concrete bonded joints are referred to wherever appropriate, as the generic concepts (e.g. the interfacial fracture energy and the effective bond length) well established for these joints are also applicable to FRP-to-steel bonded joints. Different test methods for bonded joints have been used by different researchers [52], including single-lap pull tests (Fig. 4(a)) [51,60], double-lap pull tests (Fig. 4(b)) [54,58,65], double-lap shear tests under compression [35], and beam tests [56]. Despite the variations in the test method, most of the existing studies were focused on the two important characteristics of the interface: the ultimate load of the joint (i.e. the bond strength) and the relationship between the interfacial shear stress and the interfacial relative displacement between the two adherends at a

a

FRP plate

Adhesive

Steel substrate

b Gap Steel plates

Adhesive

FRP plate

Fig. 4. Pull tests of bonded joints. (a) Single-lap pull test. (b) Double-lap pull test.

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

P u ¼ bp

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2Ep t p Gf when L ≥ Le

P u ¼ ϕðLÞbp

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2Ep t p Gf when L ≥ Le

3.3.2. Bond-slip relationship An accurate bond-slip model for FRP-to-steel interfaces is of fundamental importance to the understanding and modeling of the behavior of FRP-strengthened steel structures. A bond-slip model depicts the relationship between the local interfacial shear stress and the relative slip between the two adherends and can be experimentally obtained through bonded joint tests. To study the bond-slip behavior of FRP-to-concrete bonded joints, the single-lap pull test with the steel block supported at the loaded end (Fig. 4(a)) is probably the most suitable [48] and was also used in the recent studies on the full-range behavior of FRP-to-steel bonded joints [51,60,84,86]. For FRP-to-concrete bonded joints, Lu et al. [87] conducted a thorough review of bond-slip models and proposed three two-branch (an ascending branch and a descending branch) bond-slip models of different levels of sophistication. The simplest of the bond-slip models proposed by Lu et al. [87] is a bi-linear model with sufficient accuracy for practical use (Fig. 5(a)). The key parameters of the bilinear bondslip model are the maximum local bond shear stress τmax and the

a Elastic

Softening region

Debonding

Area under the curve= Gf

Slip

b Elastic Interfacial shear stress

3.3.1. Bond strength The bond strength is the ultimate tensile force that can be resisted by the FRP plate in a bonded joint test before the FRP plate debonds from the substrate [4]. Existing studies [51,53,57,59,60] have shown that the bond strength of an FRP-to-steel bonded joint initially increases with the bond length, but when the bond length reaches a threshold value, any further increase in the bond length does not lead to a further increase in the bond strength. This observation is similar to that found in tests on FRP-to-concrete bonded joints [48,50,66,67], and the threshold bond length value is commonly referred to as the effective bond length (Le) [66]. Two main approaches have been developed to predict the bond strength of FRP-to-steel bonded joints: (1) strength-based approach [22,57,68] which assumes that the bond strength is reached when the maximum stress/strain in the adhesive reaches its corresponding ultimate value; and (2) fracture mechanics-based approach [60,69] which is similar to that employed to predict the bond strength of FRP-to-concrete bonded joints [66,70] where the bond strength is related to the interfacial fracture energy. Apart from studies on FRP-to-steel joints, the strength-based approach has also been adopted in some studies on steel-to-steel bonded joints [71–73]. The failure criteria for the adhesive used in these studies include the maximum shear stress criterion [71], the maximum principal stress criterion [72] and the maximum shear strain criterion [73]. The strength-based approach generally implies that the ultimate load of the bonded joint is reached when the first crack occurs in the adhesive. However, Fernando [60] found from single-lap pull tests that the tensile force resisted by the FRP plate can still increase significantly after the initiation of the first crack in the adhesive, provided that the bond length is sufficiently long. In addition, the existence of an effective bond length is not compatible and cannot be explained with the strength-based approach. Therefore, it can be concluded that the strength-based approach does not reflect the debonding failure mechanism of an FRP-to-steel bonded joint; however, it may provide reasonable predictions when the bond length is small so that debonding failure of the bonded joint follows immediately the occurrence of the first crack in the adhesive. In applying the strength-based approach, an accurate analysis of interfacial stresses and/or strains in the adhesive is needed. Both analytical studies [71,74–79] and finite element (FE) studies [45,80,81] have been conducted to predict interfacial stresses in bonded joints, but many of them suffer from various limitations [60], including the omission of interfacial peeling stresses (e.g. [71]), the assumption of a constant stress state over the thickness of the adhesive (e.g. [76,78]), and the inaccurate simulation of the edge shape of the FRP plate end (e.g. [82]). A thorough review of interfacial stress analysis can be found in Ref. [81]. The fracture mechanics-based approach has been successfully employed to predict the bond strength of FRP-to-concrete bonded joints and steel-to-concrete bonded joints [66,70,83]. This approach provides the theoretical basis for the existence of an effective bond length which has also been observed in FRP-to-steel bonded joint tests [51,53,56,84]. In this approach, the bond strength depends on the interfacial fracture energy as given below [51,67,85] instead of the strength of the adhesive:

energy under shear (Mode II) loading, and ϕ(L) is a function of the bond length. Fernando [60] and Xia and Teng [51] recently conducted two series of single-lap pull tests aiming to understand the full-range behavior of FRP-to-steel bonded joints. Their test results clarified the effects of adhesive properties, adhesive layer thickness, and the plate axial rigidity of FRP on the bond strength, and verified the applicability of Eqs. (1) and (2) to FRP-to-steel bonded joints. Fernando [60] also proposed an equation to predict the Mode II interfacial fracture energy Gf based on the thickness and tensile strain energy (i.e. the area under the uniaxial tensile stress–strain curve) of the adhesive.

Interfacial shear stress

specific location on the interface (i.e. the local bond-slip relationship). In the following discussion, a single-lap pull test is assumed for simplicity of description and a double-lap pull test can be seen as two single-lap pull tests being conducted simultaneously.

135

Constant stress region

Softening region

Debonding

Area under the curve= Gf

ð1Þ ð2Þ

where Pu is the bond strength, bp is the plate width, Ep is the elastic modulus of the plate, tp is the plate thickness, Gf is the interfacial fracture

Slip Fig. 5. Bond-slip curves for linear and nonlinear adhesives. (a) Linear adhesives. (b) Nonlinear adhesives.

136

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

corresponding slip δ1, the ultimate slip δf when the local bond shear stress first reaches zero, and the interfacial fracture energy Gf which is equal to the area enclosed by the bond-slip curve and the horizontal axis. For FRP-to-concrete bonded joints, these parameters are generally related to the tensile strength of concrete as the concrete is usually the weak link of the joint. A two-branch bond-slip model without a plateau at the peak stress has been shown to perform well for almost all FRP-to-concrete bond joints because of the brittle nature of concrete. However, such a model may not work well for FRP-to-steel bonded joints where the weak link is the adhesive whose behavior may be brittle or ductile. As a result, the bond-slip response of FRP-to-steel interfaces may also be brittle or ductile as it depends on the material properties of the adhesive. Fernando [60] recently conducted a series of single-lap pull tests on FRP-to-steel bonded joints formed using four different adhesives. Results from Fernando’s study [60] showed that while a two-branch bond-slip model is suitable for bonded joints with a brittle linear adhesive, it is not suitable for joints with a more ductile nonlinear adhesive having a high strain capacity (up to 2.9%). The shape of the bond-slip curve for his joints with a nonlinear adhesive was shown to be trapezoidal (Fig. 5(b)). Based on these test results, Fernando [60] proposed three bond-slip models, two for linear adhesives and one for nonlinear adhesives respectively, where the parameters of both types of models are related to the material properties of the adhesive.

4. Flexural strengthening of steel beams Similar to an RC beam, a steel beam (or a composite steel-concrete beam) can be strengthened by bonding an FRP (generally CFRP) plate to its tension face (i.e. the soffit if a beam in positive bending is assumed, see Fig. 6) [37,54,57,88–97]. The bonded FRP plate can enhance not only the ultimate load but also the stiffness of the beam (especially when a high modulus CFRP is used) [90,93,98,99]; the latter means that the strains in the beam are reduced under the same load and the first yielding of the beam is delayed. A number of failure modes (Fig. 7) are possible for such FRP-plated steel beams, including: (a) in-plane bending failure [96]; (b) lateral buckling [37]; (c) plate-end debonding [41,97]; and (d) intermediate debonding due to local cracking or yielding

a

Steel I beam

FRP U-jackets Adhesive layer

FRP plate

b Steel I beam

Adhesive layer

FRP plate

Fig. 6. Strengthening of steel beams with a bonded FRP plate. (a) Side view. (b) Cross-sectional view.

Flange buckling Web buckling Beam

Plate end debonding

FRP Plate

Intermediate debonding

Adhesive

FRP rupture Fig. 7. Some of the failure modes of steel beams bonded with an FRP plate.

away from the plate ends [37]. Additional failure modes include: (e) local buckling of the compression flange; and (f) local buckling of the web. It should be noted that even in a beam for which these local buckling modes are not critical before FRP strengthening, they can become critical after strengthening, particularly when the strengthening involves only the bonding of FRP to the tension flange only. This is because the compression flange and the web now need to sustain a higher load level before the beam fails in one of the other modes, but their local buckling resistance does not benefit from the bonded FRP reinforcement. The in-plane bending capacity of an FRP-plated steel beam can be easily determined, provided that debonding does not become critical and hence the plane section assumption can still be used [33,100,101]. Many existing analytical studies [33,90,95,96,100,101] on FRP-plated steel beams adopted this simple assumption, which means that the prediction of debonding failures was beyond their scope. Nevertheless, research on debonding failures has attracted considerable attention worldwide (e.g. [16,33,91,100]) as discussed below. 4.1. Plate end debonding As described earlier, plate end debonding in an FRP-plated steel beam is due to high localized interfacial shear stresses and peeling stresses in the vicinity of the plate end. The magnitudes of these localized interfacial stresses depend on a number of factors [78,81], including the bending moment and the shear force in the beam at the plate end location. In a simply-supported beam in three- or four-point bending, plate end debonding is more likely to occur when the plate end is farther away from the adjacent support (i.e. when the plate end moment is larger) but can be delayed or even avoided when the plate end is very close to the adjacent support [41]. Besides the plate end location, the localized interfacial stresses can also be reduced using other measures. Examples include the use of a spew fillet of excess adhesive at the plate end [73], the use of a softer adhesive near the plate end [102], tapering the thickness of the plate near the plate end [22,103], and a combination of these measures [22]. Obviously, clamps or other types of mechanical anchors should be used wherever possible to prevent plate end debonding failure [98]. As plate end debonding in FRP-plated beams depends strongly on the localized interfacial stresses, many studies have been conducted on the prediction of these interfacial stresses, including both analytical solutions [76,78,79,90,104] and numerical investigations [45,81,90,105]. These studies have been based on different simplifying assumptions and thus possess different levels of sophistication [81]. Despite such differences, these existing studies generally assumed that the adhesive layer is linearly elastic. A comparison of different modeling approaches was recently presented by Zhang and Teng [81], which illustrates clearly how each assumption affects the predicted interfacial stresses. Stress singularity arises at the bi-material interfaces when a sharp square edge is assumed [45,106] but this issue cannot be properly dealt with by the existing analytical solutions. In real applications, the edge shape can be quite different from a sharp square edge because of the existence of a fillet of excess

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

adhesive which is introduced during the installation process; this change in the edge shape may significantly reduce the interfacial stresses, but it has seldom been appropriately considered. While existing solutions for interfacial stresses in FRP-plated beams based on the assumption of linear elastic material behavior are helpful for understanding the occurrence of plate end debonding, they cannot be used directly to predict debonding failure as debonding is controlled by the interfacial fracture energy rather than by stress values. In some existing studies (e.g. [22,91]), it was simply assumed that plate end debonding occurs when the maximum interfacial stresses found from an elastic analysis reach their corresponding material strengths; this approach may significantly underestimate the plate end debonding failure load for reasons similar to those already discussed for bonded joints. To accurately predict plate end debonding, the nonlinear and damage behavior of the interface in both the normal (i.e. peeling) direction (i.e. under Mode I loading) and the shear direction (i.e. under Mode II loading) and their interaction should be appropriately simulated. Fernando [60] made the first attempt to model plate end debonding of FRP-plated steel beams using a so-called mixed-mode cohesive law to simulate this complex behavior of the FRP-to-steel interface. Fernando’s mixed-mode cohesive law [60] was based on a bond-slip model for Mode II behavior developed from pull tests and certain assumptions for Mode I behavior and for interaction between the two modes [60]. It was shown that by using this mixed-mode cohesive law, both the process of and the ultimate load at plate end debonding can be closely predicted [60]. More recently, Chiew et al. [107] proposed an approach similar to the mixed-mode fracture criterion, where the dilatational and distortional strain energy densities are used as variables instead of the Mode I and Mode II interfacial fracture energy. Chiew et al. [107] also verified their approach using their own test results [108]. However, in Chiew et al.’s study [107], the critical values for the dilatational and distortional strain energy densities and the failure envelope accounting for the interaction between the two energy density components were both based on their own bonded joint tests where only one single adhesive was used. The wide applicability of their approach thus remains uncertain. 4.2. Intermediate debonding Intermediate debonding generally initiates at a defect (e.g. crack) [38,39] or a location of concentrated plasticity of the steel substrate [37] where the FRP plate is highly stressed; it then propagates towards a plate end. Although both plate end debonding and intermediate debonding are brittle failure modes, the latter, involving a more gradual process of debonding, is generally less brittle than the former [60]. Compared with plate end debonding, much less research is available on intermediate debonding in FRP-plated steel beams [60]. Intermediate debonding in FRP-plated steel beams is similar in nature to intermediate-crack debonding (IC debonding) in FRP-plated RC beams [47]: both initiate from a location where the FRP is highly stressed and both are dominated by interfacial shear stresses. Therefore, it can be expected that the intermediate debonding strength depends strongly on the interfacial shear fracture energy obtained from pull tests on bonded joint tests [60]. For the accurate prediction of intermediate debonding failure in an FRP-plated steel beam, an accurate bond-slip model that captures the nonlinear behavior of the FRP-to-steel interface is needed. Fernando [60] showed that with the use of a cohesive law based on a bond-slip model for Mode II behavior, both the process of and the ultimate load at intermediate debonding can be closely predicted. 4.3. Other issues Although steel beams are often prevented from lateral buckling failure by slabs and other adjacent structural members, this mode of failure is still possible in some situations. The elastic lateral buckling

137

problem has been studied by Zhang and Teng [109], but much more work is needed before a design method can be established. In the strengthening of steel or steel-concrete composite bridges, the speed of strengthening operations is of great importance when closure of traffic needs to be avoided to reduce economic losses. Hollaway et al. [110] and Zhang et al. [111] investigated the rapid strengthening of steel bridges using prepregs and film adhesive. Using this new method, a bridge may be strengthened in as short as 4 h. They also examined the effect of traffic-induced vibration during the curing of the FRP system on the performance of the strengthened structure. The effectiveness and reliability of this rapid strengthening method for steel structures were demonstrated by their study [110]. 5. Fatigue strengthening One of the most important aspects of FRP strengthening of steel structures is its capability to improve their fatigue life [112–118]. Fatigue strengthening studies have been carried out on beams [92,94,119–121], steel plates [116,117,122–126], steel rods [127] and steel connections [128–130]. Similar to the behavior of FRP-to-steel joints under static loading, Liu et al. [116,117] found that the fatigue life of FRP-strengthened steel plates initially increased with the bond length until the effective bond length Le was reached, after which any further increase in the bond length did not further increase the fatigue life. In the strengthening of steel members (e.g. plates, beams and rods), a bond length longer than Le is easy to achieve, but this may be difficult in the strengthening of steel connections where the bond length of FRP is limited. In such cases, the adhesive should be carefully selected to minimize the effective bond length. Stress intensity factors (SIFs) are commonly used in fracture mechanics to describe the stress state at a crack tip due to applied loads and/or residual stresses [122,131]. The fatigue strengthening of steel structures generally aims to reduce the SIF at a (potential) crack tip and thus increase their post-crack fatigue life. As may be expected, the use of a stiffer FRP plate (i.e. a thicker plate or a plate with a higher elastic modulus) or a stiffer adhesive (i.e. with a higher elastic modulus) can reduce the SIF [116,117,128]. One exception to this statement is that when a relative thin steel plate is strengthened on one side only, an excessively stiff plate can induce out-of-plane bending of the steel plate which can lead to premature debonding of FRP [132]. Debonding near the crack tip can lead to a significant increase in the SIF, which is detrimental to the fatigue life of the strengthened structure [114]. In addition to experimental work, a number of analytical studies [114,116,132] have been conducted on the prediction of SIFs at crack tips in FRP-strengthened steel structures. Such analysis is necessary and useful in the design of FRP systems for the fatigue strengthening of steel structures. Debonding along the CFRP-to-steel interface is also a key issue of concern in the fatigue strengthening of steel beams with CFRP, where both plate end debonding and intermediate debonding are possible. While plate end debonding may be prevented using various measures (see Section 4.1) and is often not a concern, intermediate debonding of the FRP can have a significant effect on the crack growth rate in the steel [94,114] in fatigue-strengthened steel beams. However, in most of the existing literature, debonding between the FRP and the substrate is either not considered at all or is modeled based only on a prescribed debonding shape and size as a function of the substrate crack width when the SIF is evaluated [133,134]. More research is therefore necessary to gain a better understanding of the cyclic behavior of CFRP-to-steel bonded interfaces and the interaction between intermediate debonding and fatigue crack growth in steel beams so that the detrimental effect of debonding on the fatigue life of the CFRP-strengthened steel beam can be predicted. Pre-stressing the bonded FRP reinforcement can significantly enhance the effectiveness of fatigue strengthening. By pre-tensioning

138

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

responses of a rectangular hollow section (RHS) tube subjected to an end-bearing load when five different adhesives were used to bond the CFRP. Depending on the adhesive used, the failure mode varied from the debonding initiating at a plate end to FRP rupture failure; the amount of strength enhancement achieved also varied significantly. It was shown in this study that debonding was less likely to occur when an adhesive with a larger ultimate tensile strain was used, which led to a greater load-carrying capacity of the strengthened tube [43,60]. 6.2. Buckling induced by other loads

Fig. 8. Debonding failure of CFRP-strengthened rectangular steel tube subjected to an end bearing load.

the FRP plate, compressive stresses are induced in the steel substrate to achieve crack closure, resulting in improved fatigue performance. The effect of the pre-tensioning level on the fatigue crack growth rate has been studied both experimentally and numerically [114,122,131]. By evaluating the SIF at the crack tip of the strengthened system, the pre-tensioning force needed to stop the growth of a fatigue crack can be predicted [131]. The level of pre-tensioning that can be imposed on an FRP strengthening system depends on the static and fatigue strength of the bonded joint, where a good understanding of the behavior of bonded interfaces under fatigue cyclic loading is again required.

6. Strengthening of steel structures against local buckling 6.1. Buckling induced by high local stresses In practice, high stresses in a local zone often arise, due to concentrated loads and the need to introduce discrete supports, openings and other local features. Under local high compressive stresses, local buckling failure is likely to control the thickness of a thin-walled steel structure. Such local buckling failure may be prevented by bonding FRP patches. Local high tensile stresses may also be addressed in the same way. A practically important problem is the web crippling failure of thin-walled sections under a bearing force [42]. Zhao et al. [42] found from their experimental study that bonded CFRP can be an effective solution to this problem. Fernando et al. [43] further investigated the effect of adhesive properties on the effectiveness of this strengthening technique. Fig. 8 which is extracted from Ref. [43] shows the different

a

FRP, especially CFRP, has also been used in the strengthening of other steel structures against local buckling, including steel square columns [135], lipped channel steel columns [136], and steel WT compression members [44,137,138] subjected to axial compression. The FRP strengthening has been shown to be very effective [44,60] in delaying local buckling and thus enhancing the strength of the steel structure, especially when a slender section is used. While crushing of the FRP plate was observed in some experiments [135], debonding has been found to be the most likely failure mode in the strengthened structure [44,135,136]. More research is therefore needed on debonding processes in buckling failures of FRP-strengthened steel structures where the FRP is commonly loaded in compression. 7. FRP confinement of hollow steel tubes Hollow steel tubes are used in many structures. Local buckling can occur in these tubular members when they are subjected to axial compression alone or in combination with monotonic/cyclic lateral loading. For example, hollow steel tubes are often used as bridge piers and such bridge piers suffered extensive damage and even collapse during the 1995 Hyogoken-Nanbu earthquake [139]. A typical local buckling mode of circular hollow steel tubes involves the appearance of an outward bulge near the base and is often referred to as elephant’s foot buckling (Fig. 9). In typical circular tubular structures, elephant’s foot buckling appears after yielding and the appearance of this inelastic local buckling mode normally signifies the exhaustion of the load carrying capacity and the end of the ductile response. The latter is of particular importance in seismic design, as the ductility and energy absorption capacity of the column dictate its seismic resistance. In rectangular (including square) steel tubes, a similar failure mode can occur. Here, the buckling deformation is normally outwards on the flanges and inwards on the webs. The enhancement of ductility and hence seismic resistance of hollow tubular columns through confinement by an FRP jacket has been

b

Fig. 9. Elephant’s foot buckling in a steel tube or shell. (a) Failure near the base of a steel tube. (b) Failure at the base of a liquid storage tank. Courtesy of Dr. H.B. Ge, Nagoya University and Prof. J.M. Rotter, Edinburgh University.

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

explored by the authors' group [8,140,141] as an extension of Xiao’s idea of confining concrete-filled steel tubes with FRP [9]. The technique was shown to be highly effective. The failure modes of hollow steel tubes with and without FRP confinement are shown in Fig. 10(a) and (b), while the axial stress-nominal axial strain (axial shortening/tube height) curves are shown in Fig. 10(c). It is clear that through FRP confinement, the elephant’s foot mode of buckling failure is prevented and the ductility of the tube is greatly enhanced. Nishino and Furukawa [142] also explored the same technique for hollow steel tubes independently. More recent work on FRP-strengthened hollow steel tubes/cylindrical shells can be found in [143–145]. These results also show that when the jacket thickness reaches a threshold value for which inward buckling deformations dominate the behavior, further increases in the jacket thickness do not lead to significant additional benefits as the jacket provides little resistance to inward buckling deformations. It is significant to note that FRP confinement of steel tubes leads to large increases in ductility but limited increases in the ultimate load, which is often desirable in seismic retrofit of columns which are part of a larger structure, so that the retrofitted tube will not attract forces which are so high that adjacent members may be put in danger. The elephant’s foot buckling mode is not only the critical failure mode in commonly used circular steel tubular columns under axial compression and/or bending, it also occurs in much thinner cylindrical shells in steel storage silos and tanks under combined axial compression and internal pressure. This failure mode has been commonly observed in earthquakes [146] and under static loading [147]. The use of FRP

139

jackets to strengthen thin steel cylindrical shells against local elephant’s foot buckling failure at the base has also been explored through finite element analyses by Teng and Hu [141]. The limited numerical results for a thin cylindrical shell with a radius-to-thickness ratio of 1000 and subjected to axial compression in combination with internal pressure indicate that the method leads to significant increases of the ultimate load. The FRP jacketing of steel cylindrical shells can also be used in the construction of new tanks and silos to enhance their performance. A similar and related study on the strengthening of such cylindrical shells has been conducted by Chen et al. [148] where an optimally-located ring stiffener is proposed as the strengthening method. This ring stiffener may well be a CFRP cable that provides the same circumferential stiffness and the needed strength. More recent work on the local confinement of cylindrical shells against elephant’s foot buckling and on the strengthening of cylindrical shells against buckling using bonded FRP reinforcement can be found in Refs. [11,12]. 8. FRP confinement of concrete-filled steel tubes Concrete-filled steel tubes (CFSTs) are widely used as columns in many structural systems. In CFSTs, inward buckling deformations of the steel tube are prevented by the concrete core, but degradation in steel confinement, strength and ductility can result from inelastic outward local buckling. When used as columns subjected to combined axial and lateral loads, the critical regions are the ends of the column where the moments are the largest. Under seismic loading, plastic hinges form at the column ends and large plastic rotations

b

a

Elephant’s foot buckling

c 400

Axial Stress (N/mm2)

350 300 250 Bare Steel Tube

200

Single-ply FRP Jacket Two-ply FRP Jacket

150

Three-ply FRP Jacket

100 50 0 0

0.01

0.02

0.03

Nominal Axial Strain Fig. 10. Suppression of local buckling in hollow circular steel tubes. (a) Bare steel tube after test. (b) FRP-confined tubes after test. (c) Axial stress-nominal axial strain curve.

140

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

without significant degradation in stiffness and strength are demanded here. Against this background, Xiao [9] proposed a novel form of confined concrete-filled steel tubular columns, in which the end portions are confined with steel tube segments or FRP wraps. In these columns, due to the additional confinement from an FRP or steel segment, both the inward and the outward buckling deformations of the steel tube are constrained, so the ductility and strength of the column can be substantially enhanced in the end regions. In addition, the concrete is better confined with the additional confinement from the FRP or steel segment. Although Xiao's work [9] was directed at new construction, the same concept can be applied in the strengthening/retrofit of CFSTs: FRP wrapping provides a simple and effective method to enhance the load-carrying capacity and/or ductility of CFSTs, which is similar to the FRP wrapping for strengthening RC columns [6,149]. Following Xiao's initial work [9], a number of studies have been conducted by Xiao and associates [10,150,151] as well as other researchers [152–160] on the effectiveness of FRP wrapping in improving the structural behavior of both circular [10,151–155] and square/rectangular CFSTs [150,153,154,156]. The structural behavior of FRP-confined CFSTs has recently been investigated systematically by the authors' group [161–163]. Within this study, several series of laboratory tests were conducted to examine

a

the behavior of FRP-confined CFSTs under monotonic axial compression, cyclic axial compression and the combined action of constant axial compression and cyclic lateral loading. In addition, theoretical models were developed to predict the experimental observations. Existing research has indicated that FRP jacketing is highly effective in delaying or even preventing the outward local buckling and in enhancing the performance of CFSTs subjected to various loading schemes (i.e. monotonic and cyclic axial compression, and combined axial compression and cyclic lateral loading), in terms of both the strength and ductility of the column [161,162]. Fig. 11 shows the enhancement of the load-carrying capacity of CFSTs under axial compression by FRP jacketing. 9. Concluding remarks External bonding of FRP reinforcement has been clearly established as a promising alternative strengthening technique for steel structures by existing research. As more research is conducted and more reliable design guidelines become available, the technique is also expected to receive increasing acceptance in practice. Based on the discussions presented in this paper, it is recommended that future research should address the following issues with priority.

b

c 3000

CFST Confined CFST

Axial load (kN)

2500 2000 1500 1000 500 0 0

5

10

15

Axial shortening (mm) Fig. 11. Strengthening of axially-loaded concrete-filled steel tubes with FRP confinement. (a) CFST specimen after test. (b) FRP-confined CFST specimen after test. (c) Axial load– shortening curves.

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

9.1. Steel surface treatment More work should be conducted on the treatment of steel surface preparation and characterization to develop a widely accepted procedure for use in practice that can avoid adhesion failure at the adhesive/steel interface. 9.2. Selection and formulation of adhesives In FRP-strengthened steel structures, the weak link is the adhesive layer, provided adhesion failure at the adhesive/steel interface and the FRP/adhesive interface can be avoided through appropriate surface preparation. As a result, at least for bond-critical applications, the material properties of the bonding adhesive play a key role in determining the load-carrying capacity of the strengthened structure; design theory needs to reflect the mechanical properties of the adhesive. To maximize the effectiveness of FRP strengthening, the selection of an appropriate adhesive is very important. The selection process needs to consider not only the short-term mechanical performance but also its long-term durability and ease for handling on the construction site. It may also become necessary and beneficial for material researchers to explore the development of better adhesives with properties tailored to the needs of FRP strengthening of steel structures. 9.3. Bond behavior and debonding failures Debonding failures are the most challenging issue in the flexural strengthening of steel beams and the strengthening of thin-walled steel structures against local buckling. As the adhesive is the weak link, debonding failures in FRP-strengthened steel structures depend on the properties of the adhesive. More work is needed to develop accurate bond-slip models for FRP-to-steel interfaces under Mode II loading and under mixed mode loading, with parameters of bond-slip models being given in generic adhesive properties. Particular attention needs to be paid to debonding of bonded FRP plates loaded in compression, which has received little attention so far; such debonding arises often in the strengthening of steel structures against buckling failures. 9.4. Fatigue strengthening Bonded CFRP patches provide a highly effective method for fatigue strengthening of steel structures. In such applications, the elastic modulus of the FRP rather than its ultimate tensile strength/strain is the key parameter. Pre-tensioning of the FRP patch is highly desirable, but simple methods to pre-tension and anchor such FRP plates have not yet been developed. In terms of theoretical modeling, a key issue is the interaction between debonding and crack propagation. A bond-slip model for FRP-to-steel interfaces subjected to cyclic loading is expected to be the key element in fatigue life prediction of FRP-strengthened steel structures. 9.5. FRP confinement of tubular structures External FRP confinement has been found to be an effective strengthening method for circular steel tubes with or without a concrete infill, but not so effective for square or rectangular columns. Effective methods for the strengthening of the latter columns need to be developed. 9.6. Other issues In addition to the topics mentioned above, several other topics do not appear to have been explored and should be given due attention in the future, including: (1) durability of the bonding adhesive; (2) fire resistance of FRP-strengthened steel structures; (3) strengthening of steel

141

structures against blast and impact loading; (4) the use of external FRP reinforcement for combined strengthening and corrosion protection.

Acknowledgment The authors are grateful for the financial support from The Hong Kong Polytechnic University provided through its Niche Area Funding Scheme, through a Postdoctoral Fellowship to the second author and an International Scholarship for PhD Studies to the third author. In preparing this paper, they have benefited from the list of references compiled on the topic by Prof. X.L. Zhao of Monash University which was made available to members of the Working Group on FRP-Strengthened Metallic Structures of the International Institute for FRP in Construction.

References [1] Holloway LC, Head PR. Advanced polymer composites and polymers in the civil infrastructure. Elsevier; 2001. [2] Bank LC. Composites for construction: structural design with FRP materials. John Wiley & Sons; 2006. [3] ACI-440. Report on fiber-reinforced polymer (FRP) reinforcement for concrete structures. ACI‐440; 2007. [4] Hollaway LC, Teng JG. Strengthening and rehabilitation of civil infrastructures using fibre-reinforced polymer (FRP) composites. England: Woodhead Publishing and Maney Publishing; 2008. [5] Hollaway LC, Leeming MB. Strengthening of reinforced concrete structures using externally-bonded FRP composites in structural and civil engineering. Cambridge, UK: Woodhead Publishing Limited; 1999. [6] Teng JG, Chen JF, Smith ST, Lam L. FRP strengthened RC structures. UK: John Wiley & Sons; 2002. [7] Tani K, Matsumura M, Kitada T, Hayashi H. Experimental study on seismic retrofitting method of steel bridge piers by using carbon fiber sheets. Proceedings of the Sixth Korea-Japan Joint Seminar on Steel Bridges. Tokyo, Japan; 2000. p. 437–45. [8] Teng JG, Hu YM. Suppression of local buckling in steel tubes by FRP jacketing. Proceedings of the Second International Conference on FRP Composites in Civil Engineering. Adelaide, Australia; December 8–10 2004. p. 749–53. [9] Xiao Y. Applications of FRP composites in concrete columns. Adv Struct Eng 2004;7(4):335–43. [10] Xiao Y, He W, Choi K. Confined concrete-filled tubular columns. J Struct Eng ASCE 2005;131(3):488–97. [11] Batikha M. Strengthening of thin metallic cylindrical shells using fibre reinforced polymers. UK: The University of Edinburgh; 2008. [12] Batikha M, Chen J, Rotter J, Teng J. Strengthening metallic cylindrical shells against elephant's foot buckling with FRP. Thin-Walled Struct 2009;47(10): 1078–91. [13] Jiao H, Zhao X. CFRP strengthened butt-welded very high strength (VHS) circular steel tubes. Thin-Walled Struct 2004;42(7):963–78. [14] Tavakkolizadeh M, Saadatmanesh H. Galvanic corrosion of carbon and steel in aggressive environments. J Compos Constr 2001;5(3):200–10. [15] Hollaway LC, Cadei J. Progress in the technique of upgrading metallic structures with advanced polymer composites. Prog Struct Eng Mater 2002;4(2):131–48. [16] Schnerch D, Dawood M, Rizkalla S, Sumner E, Stanford K. Bond behavior of CFRP strengthened steel structures. Adv Struct Eng 2006;9(6):805–17. [17] Teng JG, Fernando D, Yu T, Zhao XL. Treatment of steel surfaces for effective adhesive bonding. Proceedings of the 5th International Conference on FRP Composites in Civil Engineering, CICE-2010. Beijing, China; September 27–29 2010. [18] Mays GC, Huthcinson AR. Adhesives in civil engineering. New York: Cambridge University Press; 1992. [19] McKnight SH, Bourban PE, Gillespie JWJ, Karbhari VM. Surface preparation of steel for adhesive bonding in rehabilitation applications. ASCE Materials Engineering Conference. San Diego, CA; 1994. p. 1148–55. [20] Harris AF, Beevers A. The effects of grit-blasting on surface properties for adhesion. Int J Adhes Adhes 1999;19(6):445–52. [21] Baldan A. Adhesively-bonded joints and repairs in metallic alloys, polymers and composite materials: adhesives, adhesion theories and surface pretreatment. J Mater Sci 2004;39(1):1–49. [22] Schnerch D, Dawood M, Rizkalla S, Sumner E. Proposed design guidelines for strengthening of steel bridges with FRP materials. Constr Build Mater 2007;21(5):1001–10. [23] Packham DE. Surface energy, surface topography and adhesion. Int J Adhes Adhes 2003;23(6):437–48. [24] Rosen SL. Fundamental principles of polymeric materials. John Wiley and Sons; 1993. [25] Gent AN, Lai SM. Adhesion and autohesion of rubber compounds — effect of surface-roughness. Rubber Chem Technol 1995;68(1):13–25. [26] Tamai Y, Aratani K. Experimental study of relation between contact angle and surface-roughness. J Phys Chem 1972;76(22):3267–71. [27] Hitchcock SJ, Carroll NT, Nicholas MG. Some effects of substrate roughness on wettability. J Mater Sci 1981;16(3):714–32.

142

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143

[28] Gent AN, Lin CW. Model studies of the effect of surface-roughness and mechanical interlocking on adhesion. J Adhes 1990;32(2–3):113–25. [29] Amada S, Satoh A. Fractal analysis of surfaces roughened by grit blasting. J Adhes Sci Technol 2000;14(1):27–41. [30] Lavaste V, Watts JF, Chehimi MM, Lowe C. Surface characterisation of components used in coil coating primers. Int J Adhes Adhes 2000;20(1):1–10. [31] Sykes JM. Surface analysis and pretreatment of plastics and metals. Essex, England: Applied Science Publishers Ltd; 1982. [32] Schnerch D. Strengthening of steel structures with high modulus carbon fiber reinforced polymer (CFRP) materials. Raleigh (NC): North Carolina State University; 2005. [33] Cadei JMC, Stratford TJ, Hollaway LC, Duckett WG. Strengthening metallic structures using externally bonded fibre-reinforced polymers- C595. London: CIRIA; 2004. [34] Gettings M, Kinloch AJ. Surface analysis of polysiloxane-metal oxide interfaces. J Mater Sci 1977;12:2511–8. [35] El Damatty AA, Abushagur M. Testing and modeling of shear and peel behavior for bonded steel/FRP connections. Thin-Walled Struct 2003;41(11):987–1003. [36] Allan RC, Bird J, Clarke JD. Use of adhesives in repair of cracks in ship structures. Mater Sci Technol 1988;4(10):853–9. [37] Sallam HEM, Ahmad SSE, Badawy AAM, Mamdouh W. Evaluation of steel I-beams strengthened by various plating methods. Adv Struct Eng 2006;9(4):535–44. [38] Kim YJ, Garrett B. Interaction between CFRP-repair and initial damage of wide-flange steel beams subjected to three-point bending. Compos Struct 2011;93(8):1986–96. [39] Kim YJ, Kent AH. Predictive response of notched steel beams repaired with CFRP strips including bond-slip behavior. Int J Struct Stab Dyn 2012;12(1):1–21. [40] Silvestre N, Young B, Camotim D. Non-linear behaviour and load-carrying capacity of CFRP-strengthened lipped channel steel columns. Eng Struct 2008;30(10): 2613–30. [41] Deng J, Lee MMK. Behaviour under static loading of metallic beams reinforced with a bonded CFRP plate. Compos Struct 2007;78:232–42. [42] Zhao XL, Fernando D, Al-Mahaidi R. CFRP strengthened RHS subjected to transverse end bearing force. Eng Struct 2006;28(11):1555–65. [43] Fernando D, Yu T, Teng JG, Zhao XL. CFRP strengthening of rectangular steel tubes subjected to end bearing loads: effect of adhesive properties and finite element modelling. Thin-Walled Struct 2009;47(10):1020–8. [44] Harries KA, Peck AJ, Abraham EJ. Enhancing stability of structural steel sections using FRP. Thin-Walled Struct 2009;47(10):1092–101. [45] Teng JG, Zhang JW, Smith ST. Interfacial stresses in reinforced concrete beams bonded with a soffit plate: a finite element study. Constr Build Mater 2002;16(1):1–14. [46] Teng JG, Chen JF, Smith ST, Lam L. Behaviour and strength of FRP-strengthened RC structures: a state-of-the-art review. Proc Inst Civil Eng 2003;156(1):51–62. [47] Teng JG, Smith ST, Yao J, Chen JF. Intermediate crack-induced debonding in RC beams and slabs. Constr Build Mater 2003;17(6–7):447–62. [48] Yao J, Teng JG, Chen JF. Experimental study on FRP-to-concrete bonded joints. Compos B 2005;36(2):99–113. [49] Yao J, Teng JG, Lam L. Experimental study on intermediate crack debonding in FRP-strengthened RC flexural members. Adv Struct Eng 2005;8(4):365–96. [50] Yao J, Teng JG. Plate end debonding in FRP-plated RC beams-I: experiments. Eng Struct 2007;29(10):2457–71. [51] Xia SH, Teng JG. Behavior of FRP-to-steel bond joints. International Symposium on Bond Behaviour of FRP in Structures (BBFS 2005). Hong Kong, China; 2005. [52] Zhao XL, Zhang L. State-of-the-art review on FRP strengthened steel structures. Eng Struct 2007;29(8):1808–23. [53] Yu T, Fernando D, Teng JG, Zhao XL. Experimental study on CFRP-to-steel bonded interfaces. Compos B 2012;43(5):2279–89. [54] Miller TC, Chajes MJ, Mertz DR, Hastings JN. Strengthening of a steel bridge girder using CFRP plates. J Bridge Eng, ASCE 2001;6(6):523–8. [55] Sebastian WM. Nonlinear proportionality of shear-bond stress to shear force in partially plastic regions of asymmetric FRC-laminated steel members. Int J Solids Struct 2003;40(1):25–46. [56] Nozaka K, Shield CK, Hajjar JF. Design of a test specimen to assess the effective bond length of carbon fiber-reinforced polymer strips bonded to fatigued steel bridge girders. J Compos Constr 2005;9(4):304–12. [57] Nozaka K, Shield CK, Hajjar JF. Effective bond length of carbon-fiber-reinforced polymer strips bonded to fatigued steel bridge I-girders. J Bridge Eng, ASCE 2005;10(2):195–205. [58] Colombi P, Poggi C. Strengthening of tensile steel members and bolted joints using adhesively bonded CFRP plates. Constr Build Mater 2006;20(1–2):22–33. [59] Fawzia S, Al-Mahaidi R, Zhao XL, Rizkalla S. Strengthening of circular hollow steel tubular sections using high modulus CFRP sheets. Constr Build Mater 2007;21(4):839–45. [60] Fernando D. Bond behaviour and debonding failures in CFRP-strengthened steel members. PhD, Hong Kong Polytechnic University, Kowloon; 2010. [61] Haghani R. Analysis of adhesive joints used to bond FRP laminates to steel members - a numerical and experimental study. Constr Build Mater 2010;24(11): 2243–51. [62] Al-Zubaidy H, Al-Mahaidi R, Zhao XL. Experimental investigation of bond characteristics between CFRP fabrics and steel plate joints under impact tensile loads. Compos Struct 2012;94(2):510–8. [63] Hidekuma Y, Kobayashi A, Okuyama Y, Miyashita T, Nagai M. Experimental study on debonding behavior of CFRP for axial tensile reinforced steel plate by CFRP strand sheets. Proceedings of the Third Asia-Pacific Conference on FRP in Structures (APFIS2012), Sapporo, Japan; 2012.

[64] Matsumoto Y, Long ND, Yamada S, Matsui T, Matsumoto Y, Long ND. Mechanical characteristics of CFRP reinforcement for corroded steel under axial tension. Proceedings of the Third Asia-Pacific Conference on FRP in Structures (APFIS2012), Sapporo, Japan; 2012. [65] Wu C, Zhao XL, Duan WH, Al-Mahaidi R. Bond characteristics between ultra high modulus CFRP laminates and steel. Thin-Walled Struct 2012;51:47–157. [66] Chen JF, Teng JG. Anchorage strength models for FRP and steel plates bonded to concrete. J Struct Eng, ASCE 2001;127(7):784–91. [67] Yuan H, Teng JG, Seracino R, Wu ZS, Yao J. Full-range behavior of FRP-to-concrete bonded joints. Eng Struct 2004;26(5):553–65. [68] Bocciarelli M. Response of statically determined steel beams reinforced by CFRP plates in the elastic–plastic regime. Eng Struct 2009;31(4):956–67. [69] Bocciarelli M, Colombi P. Elasto-plastic debonding strength of tensile steel/CFRP joints. Eng Fract Mech 2012;85:59–72. [70] Yuan H, Wu Z. Theoretical solutions on interfacial stress transfer of externally bonded steel/composite laminates. Proceedings of the Symposium of China and Japan: Science and Technology of 21st Century, Tokyo; 1999. [71] Volkersen O. Die nietkraftverteilung in zugbeanspruchten Nietverbindungen mit konstanten laschenquerschnitten. Luftfahrtforschung 1938;15:41–7. [72] Adams RD, Peppiatt NA. Stress analysis of adhesively bonded lap joints. J Strain Anal Eng Des 1974;9(3):185–96. [73] Hart-Smith LJ. Development in adhesives-2. London: Applied Science Publishing; 1981. [74] Goland M, Reissner E. The stresses in cemented joints. J Appl Mech 1944;11: A17–27. [75] Adams RD, Peppiatt NA. Effect of Poisson's ratio strains in adherends on stresses of an idealized lap joint. J Strain Anal Eng Des 1973;8(2):134–9. [76] Rabinovitch O, Frostig Y. Closed-form high-order analysis of RC beams strengthened with FRP strips. J Compos Constr 2000;4(2):65–74. [77] Shen HS, Teng JG, Yang J. Interfacial stresses in beams and slabs bonded with thin plate. J Eng Mech, ASCE 2001;127(4):399–406. [78] Smith ST, Teng JG. Interfacial stresses in plated beams. Eng Struct 2001;23(7): 857–71. [79] Yang J, Ye JQ. An improved closed-form solution to interfacial stresses in plated beams using a two-stage approach. Int J Mech Sci 2010;52(1):13–30. [80] Adams RD. Stress analysis: a finite-element analysis approach. In: Kinloch AJ, editor. Developments in adhesives-2. London: Applied Science Publishers; 1981. [81] Zhang L, Teng JG. Finite element prediction of interfacial stresses in structural members bonded with a thin plate. Eng Struct 2010;32:459–71. [82] Harrison NL, Harrison WJ. Stresses in an adhesive layer. J Adhes 1972;3(3): 195–212. [83] Taljsten B. Defining anchor lengths of steel and CFRP plates bonded to concrete. Int J Adhes Adhes 1997;17(4):319–27. [84] Akbar I, Oehlers DJ, Mohamed-Ali MS. Derivation of the bond-slip characteristics for FRP plated steel members. J Constr Steel Res 2010;66(8–9):1047–56. [85] Wu ZS, Yuan H, Niu H. Stress transfer and fracture propagation in different kinds of adhesive joints. J Eng Mech, ASCE 2002;128(5):562–73. [86] Ibrisam A, Oehlers DJ, Ali MSM. Derivation of the bond–slip characteristics for FRP plated steel members. J Constr Steel Res 2010;66(8–9):1047–56. [87] Lu XZ, Teng JG, Ye LP, Jiang JJ. Bond-slip models for FRP sheets/plates bonded to concrete. Eng Struct 2005;27(6):920–37. [88] Tavakkolizadeh M, Saadatmanesh H. Strengthening of steel-concrete composite girders using carbon fiber reinforced polymers sheets. J Struct Eng, ASCE 2003;129(1):30–40. [89] Al-Saidy AH, Klaiber FW, Wipf TJ. Repair of steel composite beams with carbon fiber-reinforced polymer plates. J Compos Constr 2004;8(2):163–72. [90] Colombi P, Poggi C. An experimental, analytical and numerical study of the static behavior of steel beams reinforced by pultruded CFRP strips. Compos B 2006;37(1):64–73. [91] Lenwari A, Thepchatri T, Albrecht P. Debonding strength of steel beams strengthened with CFRP plates. J Compos Constr 2006;10(1):69–78. [92] Deng J, Lee MMK. Fatigue performance of metallic beam strengthened with a bonded CFRP plate. Compos Struct 2007;78(2):222–31. [93] Schnerch D, Rizkalla S. Flexural strengthening of steel bridges with high modulus CFRP strips. J Bridge Eng 2008;13(2):192–201. [94] Shaat A, Fam A. Repair of cracked steel girders connected to concrete slabs using carbon-fiber-reinforced polymer sheets. J Compos Constr 2008;12(6):650–9. [95] Fam A, MacDougall C, Shaat A. Upgrading steel-concrete composite girders and repair of damaged steel beams using bonded CFRP laminates. Thin-Walled Struct 2009;47(10):1122–35. [96] Linghoff D, Haghani R, Al-Emrani M. Carbon-fibre composites for strengthening steel structures. Thin-Walled Struct 2009;47(10):1048–58. [97] Haghani R, Al-Emrani M. A new design model for adhesive joints used to bond FRP laminates to steel beams: part B: experimental verification. Constr Build Mater 2012;30:686–94. [98] Sen R, Liby L, Mullins G. Strengthening steel bridge sections using CFRP laminates. Compos B:Eng 2001;32(4):309–22. [99] Tavakkolizadeh M, Saadatmanesh H. Repair of damaged steel-concrete composite girders using carbon fiber-reinforced polymer sheets. J Compos Constr 2003;7(4):311–22. [100] Moy SSJ. FRP composites: life extension and strengthening of metallic structures. Institution of Civil Engineers Design and Practice Guide. London: Thomas Telford; 2001. [101] Deng J, Lee MMK, Li S. Flexural strength of steel–concrete composite beams reinforced with a prestressed CFRP plate. Constr Build Mater 2011;25(1): 379–84.

J.G. Teng et al. / Journal of Constructional Steel Research 78 (2012) 131–143 [102] Fitton M, Broughton I. Variable modulus adhesives: an approach to optimised joint performance. Int J Adhes Adhes 2005;25(4):329–36. [103] Haghani R, Al-Emrani M, Kliger R. Effect of tapering on stress distribution in adhesive joints - effect of tapering length and material properties. J Compos Mater 2010;44(3):287–302. [104] Stratford T, Cadei J. Elastic analysis of adhesion stresses for the design of a strengthening plate bonded to a beam. Constr Build Mater 2006;20(1–2):34–45. [105] Linghoff D, Al-Emrani M. Performance of steel beams strengthened with CFRP laminate - part 2: FE analysis. Compos B 2010;41(7):516–22. [106] Rabinovitch O, Frostig Y. On edge stresses control in strengthened RC beams with FRP strips: adhesive layer profile effect. J Eng Mech 2001;127(4):317–25. [107] Chiew SP, Yu Y, Lee CK. Bond failure of steel beams with FRP laminates - part 1: model development. Compos B 2011;42(5):1114–21. [108] Yu Y, Chiew SP, Lee CK. Bond failure of steel beams with FRP laminates - part 2: verification. Compos B 2011;42(5):1122–34. [109] Zhang L, Teng JG. Elastic lateral buckling of I‐section beams bonded with a thin plate. Proceedings of the Third International Conference on Advanced Composites in Construction (ACIC 2007), University of Bath, UK; April 2–4 2007. p. 441–6. [110] Hollaway LC, Zhang L, Photiou NK, Teng JG, Zhang SS. Advances in adhesive joining of carbon fibre/polymer composites to steel members for repair and rehabilitation of bridge structures. Adv Struct Eng 2006;9(6):791–803. [111] Zhang L, Hollaway LC, Teng JG, Zhang SS. Strengthening of steel bridges under low frequency vibrations. Proceedings of the Third International Conference on FRP Composites in Civil Engineering. Miami, Florida, USA; December 13–15 2006. [112] Bassetti A, Nussbaumer A, Hirt MA. Crack repair and fatigue life extension of riveted bridge members using composite materials. Bridge Engineering Conference 2000: Past Achievements, Current Practices, Future Technologies. Egypt: Sharm El-Sheikh; 2000. [113] Bassetti A, Nussbaumer A, Hirt MA. Fatigue life extension of riveted bridge members using prestressed carbon fiber composites. International Conference on Steel Structures of the 2000's, Istanbul, Turkey; 2000. p. 375–80. [114] Colombi P, Bassetti A, Nussbaumer A. Crack growth induced delamination on steel members reinforced by prestressed composite patch. Fatigue Fract Eng Mater Struct 2003;26(5):429–37. [115] Dawood M, Rizkalla S, Sumner E. Fatigue and overloading behavior of steel-concrete composite flexural members strengthened with high modulus CFRP materials. J Compos Constr 2007;11(6):659–69. [116] Liu H, Al-Mahaidi R, Zhao XL. Experimental study of fatigue crack growth behavior in adhesively reinforced steel structures. Compos Struct 2009;90:12–20. [117] Liu H, Xiao Z, Zhao XL, Al-Mahaidi R. Prediction of fatigue life for CFRP-strengthened steel plates. Thin-Walled Struct 2009;47:1069–77. [118] Jiao H, Zhao XL, Mashiri F. Improving fatigue performance of CFRP strengthened steel beams by applying vacuum pressure in the wet layup of CFRP woven sheets. Proceedings of the Third Asia-Pacific Conference on FRP in Structures (APFIS2012), Sapporo, Japan; 2012. [119] Tavakkolizadeh M, Saadatmanesh H. Fatigue strength of steel girders strengthened with carbon fiber reinforced polymer patch. J Struct Eng, ASCE 2003;129(2): 186–96. [120] Egilmez OO, Yormaz D. Cyclic testing of steel I-beams reinforced with GFRP. Steel Compos Struct 2011;11(2):93–114. [121] Ghafoori E, Motavalli M, Botsis J, Herwig A, Galli M. Fatigue strengthening of damaged metallic beams using prestressed unbonded and bonded CFRP plates, Int J Fatigue (in press). [122] Colombi P, Bassetti A, Nussbaumer A. Analysis of cracked steel members reinforced by pre-stress composite patch. Fatigue Fract Eng Mater Struct 2003;26(1):59–66. [123] Nakamura H, Maeda K, Suzuki H, Irube T, Nakamura H, Maeda K, Suzuki H, Irube T. Monitoring for fatigue crack propagation of steel plate repaired by CFRP strips. Proceedings of the 4th International Conference on Bridge Maintenance, Safety and Management, IABMAS'08; 2008. p. 2943–50. [124] Bocciarelli M, Colombi P, Fava G, Poggi C. Fatigue performance of tensile steel members strengthened with CFRP plates. Compos Struct 2009;87(4):334–43. [125] Colombi P, Giulia F. Fatigue behaviour of tensile steel/CFRP joints. Compos Struct 2012;94(8):2407–17. [126] Wu C, Zhao XL, Al-Mahaidi R, Emdad M, Duan WH. Fatigue tests of cracked steel plates strengthened with UHM CFRP plates, Adv Struct Eng (in press). [127] Jones SC, Civjan SA. Application of fiber reinforced polymer overlays to extend steel fatigue life. J Compos Constr 2003;7(4):331–8. [128] Nakamura H, Jiang W, Suzuki H, Maeda K, Irube T. Experimental study on repair of fatigue cracks at welded web gusset joint using CFRP strips. Thin-Walled Struct 2009;47(10):1059–68. [129] Chen T, Yu QQ, Gu XL, Zhao XL. Study on fatigue behavior of strengthened non-load-carrying cruciform welded joints using carbon fiber sheets. Int J Struct Stab Dyn 2012;12(1):179–94. [130] Xiao ZG, Zhao XL. CFRP repaired welded thin-walled cross-beam connections subject to in-plane fatigue loading. Int J Struct Stab Dyn 2012;12(1):195–211. [131] Taljsten B, Hansen CS, Schmidt JW. Strengthening of old metallic structures in fatigue with prestressed and non-prestressed CFRP laminates. Constr Build Mater 2009;23(4):1665–77.

143

[132] Tsouvalis NG, Mirisiotis LS, Dimou DN. Experimental and numerical study of the fatigue behaviour of composite patch reinforced cracked steel plates. Int J Fatigue 2009;31(10):1613–27. [133] Megueni A, Bouiadjra BB, Belhouari M. Disbond effect on the stress intensity factor for repairing cracks with bonded composite patch. Comput Mater Sci 2004;29:407–13. [134] Ouinas D, Serier B, Bouiadjra BB. The effects of disbonds on the pure mode II stress intensity factor of aluminum plate reinforced with bonded composite materials. Comput Mater Sci 2007;39:782–7. [135] Shaat A, Fam AZ. Slender steel columns strengthened using high-modulus CFRP plates for buckling control. J Compos Constr 2009;13(1):2–12. [136] Silvestre N, Camotim D, Young B. On the use of the EC3 and AISI specifications to estimate the ultimate load of CFRP-strengthened cold-formed steel lipped channel columns. Thin-Walled Struct 2009;47(10):1102–11. [137] Kim YJ, Harries KA. Behaviour of tee-section bracing members retrofitted with CFRP strips subjected to axial compression. Compos B:Eng 2011;42(4):789–800. [138] El-Tawil S, Ekiz E, Goel S, Chao S. Retraining local and global buckling behavior of steel plastic hinges using CFRP. J Constr Steel Res 2011;67(3):261–9. [139] Kitada T, Yamaguchi T, Matsumura M, Okada J, Ono K, Ochi N. New technologies of steel bridges in Japan. J Constr Steel Res 2002;58(1):21–70. [140] Hu YM. Behaviour of FRP-confined steel tubes under axial compression. BEng Thesis, Department of Civil and Structural Engineering, The Hong Kong Polytechnic University, April; 2004. [141] Teng J, Hu YM. Behaviour of FRP-jacketed circular steel tubes and cylindrical shells under axial compression. Constr Build Mater 2007;21(4):827–38. [142] Nishino T, Furukawa T. Strength and deformation capacities of circular hollow section steel member reinforced with carbon fiber. Proceedings of the Seventh Pacific Structural Steel Conference. American Institute of Steel Construction; March 2004. [143] Haedir J, Zhao XL, Grzebieta RH, Bambach MR. Non-linear analysis to predict the moment-curvature response of CFRP-strengthened steel CHS tubular beams. Thin-Walled Struct 2011;49:997–1006. [144] Haedir J, Zhao XL. Design of short CFRP-reinforced steel tubular columns. J Constr Steel Res 2011;67:497–509. [145] Bhetwal KK, Yamada S. Effects of CFRP reinforcements on the buckling behavior of thin-walled steel cylinders under compression. Int J Struct Stab Dyn 2012;12(1):131–51. [146] Manos GC, Clough RW. Tank damage during the Coalinga earthquake. Earthquake Eng Struct Dyn 1985;13:449–66. [147] Rotter JM. Local collapse of axially compressed pressurized thin steel cylinders. J Struct Eng, ASCE 1990;116(7):1955–70. [148] Chen JF, Rotter JM, Teng JG. A simple remedy for elephant’s foot buckling in cylindrical silos and tanks. Adv Struct Eng 2006;9(3):409–20. [149] Teng JG, Lam L. Behavior and modeling of fiber reinforced polymer-confined concrete. J Struct Eng, ASCE 2004;130(11):1713–23. [150] Mao XY, Xiao Y. Seismic behavior of confined square CFT columns. Eng Struct 2006;28(10):1378–86. [151] Shan JH, Chen R, Zhang WX, Xiao Y, Yi WJ, Lu FY. Behavior of concrete filled tubes and confined concrete filled tubes under high speed impact. Adv Struct Eng 2007;10(2):209–18. [152] Gu W, Guan CW, Zhao YH, Cao H. Experimental study on concentricallycompressed circular concrete filled CFRP-steel composite tubular short columns. J Shenyang Archit Civil Eng Univ 2004;20(2):118–20 (in Chinese). [153] Tao Z, Han LH, Zhuang JP. Axial loading behavior of CFRP strengthened concrete-filled steel tubular stub columns. Adv Struct Eng 2007;10(1):37–46. [154] Wang ZB, Tao Z, Han LH. Performance of CFRP-strengthened concrete filled steel tubes under axial compression. Proceedings, 10th International Symposium on Structural Engineering for Young Experts, Hunan University, Changsha, China; 2008. p. 689–94. [155] Liu L, Lu YY. Axial bearing capacity of short FRP confined concrete-filled steel tubular columns. J Wuhan Univ Technol Mater Sci Ed 2010;25(3):454–8. [156] Park JW, Hong YK, Choi SM. Behaviors of concrete filled square steel tubes confined by carbon fiber sheets (CFS) under compression and cyclic loads. Steel Compos Struct 2010;10(2):187–205. [157] Li SQ, Chen JF, Bisby LA, Hu YM, Teng JG. Strain efficiency of FRP jackets in FRP-confined concrete-filled circular steel tubes. Int J Struct Stab Dyn 2012;12(1):75–94. [158] Park JW, Hong YK, Hong GS, Kim H, Choi SM. Design formulas of concrete filled circular steel tubes reinforced by carbon fiber reinforced plastic sheets. Procedia Eng 2011;14:2916–22. [159] Sundarraja MC, Prabhu GG. Experimental study on CFST members strengthened by CFRP composites under compression. J Constr Steel Res 2012;72:75–83. [160] Tao Z, Wang ZB, Han LH, Uy B. Fire performance of concrete-filled steel tubular columns strengthened by CFRP. Steel Compos Struct 2011;11(4):307–24. [161] Hu YM., 2011. Behaviour and modelling of FRP-confined hollow and concrete-filled steel tubular columns. PhD thesis, Department of Civil and Structural Engineering, The Hong Kong Polytechnic University, Hong Kong, China. [162] Hu YM, Yu T, Teng JG. FRP-confined circular concrete-filled thin steel tubes under axial compression. J Compos Constr ASCE 2011;15(5):850–60. [163] Teng JG, Hu YM, Yu T. Stress–strain model for concrete in FRP-confined steel tubular columns [Submitted to Engineering Structures].