FRP-strengthened RC Slabs Anchored With FRP Anchors

Engineering Structures 33 (2011) 1075–1087

Contents lists available at ScienceDirect

Engineering Structures

journal homepage: www.elsevier.com/locate/engstruct

FRP-strengthened RC slabs anchored with FRP anchorsScott T. Smith a,∗, Shenghua Hu a, Seo Jin Kim a, Rudolf Seracino b

a Department of Civil Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong, Chinab Department of Civil, Construction, and Environmental Engineering, North Carolina State University, USA

a r t i c l e i n f o

Article history:Received 15 June 2010Received in revised form28 October 2010Accepted 4 November 2010Available online 3 February 2011

Keywords:DebondingDeformabilityFibre-reinforced polymer compositesFlexural strengtheningFRP anchorsSlabs

a b s t r a c t

An abundance of tests over the last two decades has shown the bending capacity of flexural memberssuch as reinforced concrete (RC) beams and slabs to be enhanced by the bonding of fibre-reinforcedpolymer (FRP) composites to their tension face. The propensity of the FRP to debond, however, limitsits effectiveness. Different types of anchorages have therefore been investigated in order to delay or evenprevent debonding. The so-called FRP anchor, which is made from rolled fibre sheets or bundles of losefibres, is particularly suitable for anchoring FRP composites to a variety of structural element shapes.Studies that assess the effectiveness of FRP anchors in anchoring FRP strengthening in flexural membersis, however, limited. This paper in turn reports a series of tests on one-way spanning simply supported RCslabs which have been strengthened in flexure with tension face bonded FRP composites and anchoredwith different arrangements of FRP anchors. The load–deflection responses of all slab tests are plotted, inaddition to selected strain results. The behaviours of the specimens including the failure modes are alsodiscussed. The greatest enhancement in load and deflection experienced by the six slabs strengthenedwith FRP plates and anchored with FRP anchors was 30% and 110%, respectively, over the unanchoredFRP-strengthened control slab. Thepaper also discusses the strategic placement of FRP anchors for optimalstrength and deflection enhancement in FRP-strengthened RC slabs.

© 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Numerous experimental investigations have proven the abilityof fibre-reinforced polymer (FRP) composites to increase theflexural capacities of beams and slabs when bonded to theirtension faces [1,2]. Numerous studies have also observed the FRPto debond at strains well below its rupture strain. Such prematurefailure, which has been observed to initiate at the base of flexuraland flexural-shear cracks along the length of the member (e.g.IC debonding, [2]) or at the FRP plate end (e.g. concrete coverseparation, [2]), can occur in a relatively sudden manner andconstitutes an under-utilisation of the strength and strain capacityof the FRP. Mechanical anchorage of the FRP offers a real solutionto the debonding problem and several different systems have beentrialed to date. They include, but are not limited to, embeddedmetal threads [3], nailed plates (also known as hybrid bonding [4]),U-jackets [5], near-surface mounted rods [6], and anchors madewith FRP [7] (also known as spike anchors but herein referred toas FRP anchors or anchors). FRP anchors are versatile as they arenon-corrosive and can be applied to wide dimensioned elementssuch as slabs and walls. A recent review of FRP anchors is provided

∗ Corresponding author. Tel.: +852 2241 5699; fax: +852 2559 5337.E-mail address: [email protected] (S.T. Smith).

0141-0296/$ – see front matter© 2010 Elsevier Ltd. All rights reserved.doi:10.1016/j.engstruct.2010.11.018

in [7] while a review of other anchorage methods (including FRPanchors) is presented in [8]. The anchorage of steel strengtheningplates using metallic bolts is a related field of research (e.g. [9]),however, it is outside the scope of this paper and is therefore notconsidered further.

Fig. 1. is a schematic representation of the face of a concretemember which has been strengthened with an externally bondedFRP plate and anchored with an FRP anchor. Such an anchor isessentially made from glass or carbon fibres in which fibre sheetsare folded or rolled, or lose fibres are bundled together. One endof the anchor (herein anchor dowel) is inserted into an epoxy filledhole in the concrete substrate (Fig. 1(b)) and the other end of theanchor is passed through the externally bonded FRP strengtheningplate (herein FRP plate or plate). The free ends of the fibres (hereinanchor fan) are splayed and epoxied onto the surface of the platein order to disperse local stress concentrations. The double anchorfan arrangement (herein bow-tie) shown in Fig. 1 has been tailormade for the test slabs reported herein. As a precursor to thebow-tie anchor fan form, Smith [10] reported FRP anchors witha single fan component to increase the shear strength and slipcapacity of FRP-to-concrete joints by up to approximately 70% and800%, respectively, over unanchored control joints. The relativedifference between the strength and the behaviour of single fanand bow-tie anchors in FRP-to-concrete joint tests has also beensummarised in [10]. While Smith [10] reported both types of

http://dx.doi.org/10.1016/j.engstruct.2010.11.018

http://www.elsevier.com/locate/engstruct

http://www.elsevier.com/locate/engstruct

mailto:[email protected]

http://dx.doi.org/10.1016/j.engstruct.2010.11.018

1076 S.T. Smith et al. / Engineering Structures 33 (2011) 1075–1087

Fig. 1. FRP anchor and plate: (a) overall view; (b) cut-away view.

anchors to exhibit similar load–slip characteristics over most ofthe responses, the bow-tie anchors were ultimately able to resistmuch greater slips (while maintaining limited strength) beforefailure. Such an extensive slip capacity is a desirable feature of anFRP anchor especially when large slips are expected between theFRP strengthening and concrete substrate in structural members.Another benefit of the bow-tie anchor alternative is that slip maybe in the other direction for members where the applied load canmove. Here the minimum criterion is to position the anchor fanon the side of the anchor in the direction of load. There has beenlimited research though conducted to date on characterising thefundamental behaviour of FRP anchors (e.g. [7,11–14]) and morework is clearly required. Such work is, however, outside the scopeof this paper.

The majority of the research conducted to date on FRP an-chors has been on the anchorage of flexurally strengthened RCbeams [15,16], slabs [17,18], slab–column connections [19], con-fined columns [20], and concrete and masonry walls [21,22]. Insuch research, FRP anchors were generally shown to be effectivein enhancing the strength and deformability of the strengthenedmembers, however, the FRP anchors were generally not the focusof these studies. Also, in many cases, the FRP was not observed tofail and as a result the limits of the anchors were not established.Brief reviews of some of the literature of FRP-anchored FRP flexu-rally strengthened RC beams and slabs are provided as follows.

Teng et al. [23] reported seven cantilever RC slabs tests of700 mm span of which six slabs were strengthened in flexurewith glass FRP (GFRP) composites formed in a wet lay-up manner.The unanchored slabs were found to fail by IC debonding withdebonding initiating at the fixed end of the slab. Two of thestrengthened slabs were anchored with FRP anchors positioned150 mm and 300 mm from the fixed end. In both cases, the FRPanchors were observed to reduce the rate of debonding crackpropagation. In the first case, the GFRP plate ruptured after thedebonding crack had propagated to the second anchor. The lowtensile strength of the GFRP (i.e. 428 MPa) made it susceptible torupture failure. The second anchored slab test utilised an extra

layer of GFRP. In this case the anchors failed after the debondingcrack had propagated along the plate. In both anchored slabcases the slope of the load–deflection curve clearly decreasedas debonding propagated. The two anchored slabs experienceda 24% and 61% increase over the unanchored but strengthenedcontrol slab respectively, however, the deflection at failure for bothanchored slabs was 76% of the control.

Lam and Teng [17] then reported an additional five RCcantilever slabs tests of 700 mm span in which fours slabs werestrengthened in flexure with wet lay-up GFRP and anchored withFRP anchors positioned in the same locations as Teng et al.’s [23]test slabs. The main test variables were preloading as well asinternal tension steel ratio and position. In all strengthened slabtests the FRP was observed to rupture. In some cases, debondingwas halted by the first anchor and in other cases no debondingwasobserved.

Eshwar et al. [24] strengthened ten beams of varying soffitcurvature with carbon FRP (CFRP) tension face plates. The spanof the beams was 6 m, the length of FRP was 5.2 m, and thefailure mode was IC debonding. Of the three beams with greatestcurvatures, twowere strengthenedwith identical configurations ofwet lay-up FRP and one of these beams was additionally installedwith FRP anchors at 500 mm centres. The increase in strengthand mid-span deflection of this anchored beam to its unanchoredcounterpart was 34% and 74%, respectively. The anchored beamappeared to fail by complete debonding of the FRP followed byanchor rupture, however, the effectiveness of the FRP anchor inenhancing load and deflection had been proven.

Oh and Sim [15] reported tests on eleven simply supportedbeams each of 2 m span. Ten of these beams were strengthenedin flexure with tension face GFRP plates formed in a wet lay-upmanner. The beams were susceptible to concrete cover separationfailure, so two FRP anchors were positioned at 500 mm centres atthe end of one beam specimen. The anchors were not successfulin delaying the occurrence of concrete cover separation and asa result they did not enhance load or deformation capacity ofthe beam. More recently, Micelli et al. [16] showed FRP anchorsspaced at 250 mm centres in 2.2 m spanning beams to increasethe load carrying capacity of the FRP-strengthened beams by13% above the strengthened but unanchored control beams. Thestrengthened beams ultimately failed by IC debonding after whichthe behaviour of the beams resorted to that of the unstrengthened(and unanchored) control beam.

While Brunckhorst et al. [25] did not consider FRP anchorage,their research is still applicable and is therefore reviewedhere. Brunckhorst et al. [25] presented a diagram of a genericmoment–displacement (analogous to load–deflection) response ofan RC beam strengthened in flexure with a tension face bondedCFRP pultruded plate comprising of multi-directional fibres. Theplate was also anchored with regularly spaced metal screw-bolts.In order to install the bolts, holes were drilled through the initiallybonded (and cured) FRP plate at regular intervals along the wholelength of the plate and thenmetal bolts were inserted. The genericresponse consisted of several key features, namely (i) first crackingof the tensile concrete, (ii) initiation of debonding of the FRP platevia ‘gliding fracture’ (this translation appears to be consistent withIC debonding), (iii) a sharp drop of moment upon initiation ofdebonding, (iv) residual strength (above the plain unstrengthenedRC beam) provided by the remaining bonded FRP, and (v) residualstrength provided by the bolts (after complete plate debonding).

In light of the overall success of FRP anchors in delayingor suppressing IC debonding failures, a clear understandingsurprisingly still does not exist about the exact role the FRP anchorsplay when used in structural members. Also, there is no rationalmethodology for the design and placement of the anchors. Suchis the motivation for the experimental program reported in this

S.T. Smith et al. / Engineering Structures 33 (2011) 1075–1087 1077

(a) Elevation. (b) Section.

Fig. 2. RC slab and geometry, loading and instrumentation details { 90 mm strain gauge; LVDT; CMR constant moment region}.

(a) S1.

(b) S2.

Fig. 3. Control slab details (tension face) and strain gauge layout: (a) unstrengthened and unanchored, (b) FRP-strengthened and unanchored { support line; - - -tension and cross-bar steel; 10 mm strain gauge}.

paper. More specifically, the aims of the study reported hereinare to (i) quantify the load and deflection enhancement that FRPanchors can provide to FRP flexurally strengthened RC slabs failingby IC debonding, (ii) observe the behaviour and failuremodes of theanchored slabs, and (iii) manipulate the load–deflection responsein beneficial ways with the strategic use of FRP anchors. To achievethese objectives, eight simply supported one-way spanning RCslabs were constructed and tested to failure. One slab served asan unstrengthened and unanchored control while the remainingseven slabs were strengthened in flexure with CFRP (herein FRP)formed in a wet lay-up manner. Six of these seven strengthenedslabs were anchored with different FRP anchor types and layouts.

2. Experimental setup

2.1. Details of test slabs

The experimental program consisted of eight simply supportedone-way spanning RC slab tests. All slabswere rectangular in cross-section of nominally 150mmdepth and 400mmwidthwith a clearspan of 2400mmas shown in Fig. 2(a). The slabswere reinforced inflexure with two 10 mm diameter hot-rolled steel reinforcing barspositioned at an effective depth of 120 mm as shown in Fig. 2(b).The same steel reinforcement type was also placed on the top ofthe longitudinal reinforcement at 200 mm centres as cross-bars(Fig. 2(b), Fig. 3). The span-to-effective depth ratio of each slab was20 and the steel reinforcement ratio was 0.33%. All slabs were castfrom the same batch of ready-mix concrete.

Of the eight identical slabs cast, two were tested as controlslabs S1 and S2. Slab S1 was a plain RC slab which contained noFRP flexural strengthening and no FRP anchorage. Slab S2 wasstrengthened in flexure with three layers of carbon fibre sheetin a wet lay-up procedure but not anchored. The remaining sixslabs were all strengthened with the same flexural strengtheningarrangement as Slab S2 in addition to being anchored withdifferent arrangements and types of FRP anchors. Table 1 providesa summary of the key variables for all the eight slabs. Figs. 3 and4 provide a summary of the FRP flexural strengthening and FRPanchorage arrangements for control Slabs S1 and S2, and anchoredSlabs S3–S8, respectively.

The type and positioning of the FRP anchorage were the keyvariables in this study and the following comments are offered insupport of the layouts presented in Fig. 4. At the initial design stageof the project, the anchorage arrangements for Slabs S3–S7 wereintuitively selected. The anchorage arrangement for Slab S8 wasdecided upon after observing the test results of Slabs S3–S7. SlabsS3–S6 were anchored with Type 1 anchors, Slab S7 was anchoredwith Type 2 anchors and Slab S8 was anchored with a combinationof Type 1 and Type 2 anchors. Type 1 anchors contained twice theamount of fibre as Type 2 anchors, although, both anchors weremade by hand in an identical manner. In addition, no anchors wereinstalled at cross-bar locations because (i) the embedment depth ofthe anchor was greater than the cover to the cross-bar, and (ii) ascross-bars can act as crack initiators, it was not deemed logical toposition an anchor where flexural cracking is likely to occur. Theanchorage scheme for Slab S3 (Fig. 4(a)) entailed the anchor dowelbeing positioned mid-distance between the cross-bars along the


(a) S3.

(b) S4.

(c) S5.

(d) S6.

(e) S7.

(f) S8.

Fig. 4. FRP anchor layout (tension face) and strain gauge layout { Type 1 FRP anchor; Type 2 FRP anchor; support line; - - - cross-bar (tension steel omitted);10 mm strain gauge}.


Table 1FRP strengthening and FRP anchorage details.

Slab FRP strengthening FRP anchor details CommentsArrangement Fibre content (mm) Shear connection fibre (mm2)c

S1 Nil Nil Nil Nil Unstrengthened controlS2 3-layers Nil Nil Nil Unanchored controlS3 3-layers Fig. 4(a) 250a 166 Type 1 anchorS4 3-layers Fig. 4(b) 250 83 Type 1 anchorS5 3-layers Fig. 4(c) 250 166 Type 1 anchorS6 3-layers Fig. 4(d) 250 83 Type 1 anchorS7 3-layers Fig. 4(e) 125b 166 Type 2 anchorS8 3-layers Fig. 4(f) 125 and 250 166 Hybrid Type 1 and Type 2 anchorsa Type 1 FRP anchor.b Type 2 FRP anchor.c Total cross-sectional area of FRP anchor crossing the FRP-to-concrete interface in one shear span (Area = cross-sectional area of the fibre sheet in single anchor dowel

times number of anchors in one shear span).

whole length of the FRP flexural strengthening while in Slab S4(Fig. 4(b)) every second anchor was removed. In Slab S5 (Fig. 4(c)),the shear span anchors of Slab S2 were only retained. For SlabS6 (Fig. 4(d)), only the ends of the FRP plate were anchored. Theanchored regions as well as the amount of fibre used to constructthe anchors in Slabs S7 (Fig. 4(e)) and S8 (Fig. 4(f)) were identicalto Slab S5 (Fig. 4(c)). The Type 2 anchors used in Slab S7 weredouble in number and half the spacing as the Type 1 anchorsused in Slab S5. In order to avoid collision with the cross-bars, theanchors in Slab S7were offset 50mmto those of Slab S5. The hybridarrangement of anchors in Slab S8 utilised both Type 1 and Type 2anchors. The reasons for positioning the Type 1 anchors closer tothe mid-span region and the Type 2 anchors closer to the free endof the FRP plate are discussed in detail in Section 3 of this paper.

It should be noted that only one specimen was tested for eachconfiguration of FRP strengthening and FRP anchorage. The resultsand discussions presented in this paper are therefore void of thevariation inherent in experimental testing of like specimens.

In reality, serviceability rather than strength commonly governsthe design of slabs. The current experimental program, however,considers both load and deflection. Also, slabs have been selectedin this study in order to ensure the occurrence of IC debondingfailure of the FRP flexural strengthening. Such a debonding failuremode is most common in FRP-strengthened RC slabs and canprovide a useful framework for evaluation of the effectiveness ofFRP anchors. The understanding gained from this study on FRPanchors can be extended to other RC structural elements, such asbeams.

2.2. FRP anchor construction

The same roll of carbon fibre sheet of 0.166 mm nominalthickness and the same tins of two-part epoxy were used tomake all the FRP anchors and all the FRP plates in this study. Inaddition, all anchors were made from the same bow-tie form asshown in Figs. 1 and 5. The method of manufacturing the FRPanchors was essentially the same as that reported in detail in [26].The difference though was the use of a single anchor fan in [26]and a bow-tie arrangement in the present study. The followingexplanation is offered in support of the making of the anchors.

All bow-tie anchors were made from 90 mm long carbon fibresheets. The 90 mm length was to form a 50 mm long fan with a40 mm embedded portion which included a small allowance forthe 90 degree bend portion. Type 1 anchors were made from a250 mm wide sheet of fibre which was twice the width as thatused for Type 2 anchors. In preparing the anchors, the fibre sheetwas spread out on a flat surface and a 25 mm long region at oneend of the sheet was impregnated with epoxy across the wholewidth of the sheet. Both exterior edges of the sheet were thenrolled towards the centre of the sheet until the two rolled portions

a

b

c

Fig. 5. Construction of bow-tie FRP anchors: (a) schematic of fibre sheet rolling(epoxy impregnated end fibres shown in foreground); (b) schematic of completedFRP anchor; (c) actual FRP anchor.

met (Fig. 5(a)). The rolled sheets, of approximate circular cross-section, were then inserted into preformed holes of 14 mm (forType 1 anchors) or 10 mm diameter (for Type 2 anchors) in a


(a) Installed FRP anchor dowel. (b) Wet lay-up application of fibre sheet and anchor fans.

Fig. 6. FRP anchor and plate installation.

Fig. 7. Test setup.

polystyrene mould which was pre-filled with epoxy in order toproperly form the anchor dowel. The anchors were then removedfrom the mould once the epoxy had cured for at least one day.Fig. 5(b) and (c) show completed anchors.

2.3. Application of FRP strengthening and FRP anchors

The FRP plates and FRP anchors were installed in accordancewith the procedures described in [26]. The following summary is,however, provided for completeness:

1. Drill anchor dowel holes into concrete slabs at appropriatelocations.

2. Prepare surface of concrete to be strengthened using apneumatic needle scaler. This method of surface preparationremoves the top surface of weakened concrete, exposesaggregate and also roughens the concrete surface (Fig. 6(a)). Ithas been found to be successful on numerous occasions by thefirst author of this paper (e.g. [7,26]).

3. Flush anchor dowel holes with compressed air for cleaning andthen inject epoxy. Then, insert the preformed anchor dowel(Fig. 6(a)) and allow to cure for at least half a day.

4. Slip parted carbon fibre sheet fibres over the anchor fan fibresand then form the plate from the sheet fibres in a wet lay-upmanner (Fig. 6(b)) (i.e. three layers of fibre sheet were used toform the FRP plate in this study).

5. Curve the dry fan fibres of the anchor (without kinking), splayinto the bow-tie form, and then epoxy onto the outer surface ofthe outermost plate layer.

6. Allow the plate and anchor fan fibres to cure for a period of 7days prior to testing. In the case of this experimental program,all slabs were left in a controlled laboratory environment.

Recent work by this research group [26] has identified theimportance of not impregnating the anchor fibres in the bendregion with epoxy upon installation of the FRP anchor. Dry fibresin the bend region will enable slips of the plate relative to theconcrete surface to be achieved that are generally greater than5mm. Such large slippagewill enable greater deflection of the FRP-strengthened and FRP-anchored slabs considered herein and helpprevent brittle FRP anchor rupture failure.

The specimens considered in this study are undamagedspecimens. In reality though, FRP strengthening may need tobe applied to damaged or deteriorated RC members containingcorroded reinforcement, cracked concrete, or spalled concretecover. In such cases, it would be necessary to repair and preparethe concrete substrate onto which the FRP system will be appliedaccording to the FRP manufacturer’s recommendations and designguidelines (e.g. [27,28]). In this situation, it is not anticipated thatthe behaviour of the FRP-strengthened system in question wouldsignificantly change.

2.4. Instrumentation and test procedure

Two electric strain gauges of 90 mm gauge length weremounted onto the top compressive face of each of the eight RCslabs at mid-span as shown in Fig. 2. Extensive arrays of electricstrain gauges of 10 mm gauge length were mounted onto the FRPplates for Slabs S2–S8 as shown in Figs. 3 and 4. In addition, linearvariable displacement transducers (LVDTs) were positioned alongthe length of the slab as well as above the supports in order torecord the vertical deflection as shown in Fig. 2(a) (only the resultsat mid-span are reported in this paper; also, support deflectionswere negligible). LVDTs were also placed at each end of the FRPand were reacted off the adjacent slab soffit in order to measureplate end slip for Slabs S5–S8. Such results are also not reportedherein, however, suitable comments are provided in Section 3 ofthis paper. All slabs were tested in a stiff reaction frame and loadwas applied through a 1000 kN capacity servo-controlled actuatorby displacing the ram of the actuator at a rate of 1 mm/min. Themass of the spreader beam (170 kg)was included in the load, strainand LVDT readings, however, the self-weight of the slab was not.Fig. 7 shows a typical test in progress.

2.5. Material properties

All the eight slabs were poured in one batch and tested overa ten week period. The mechanical properties of the concrete,


Table 2Concrete mechanical properties.

Slab Age (days) Propertiesa

Age (days) fcu (MPa) Ec (MPa) fct (MPa) ER (MPa)

S1 3136 51.7 28389 3.3 4.1S2 43

S3 44

S4 74 77 55.1 29299 4.2 6.1S5 79

S6 9294 56.8 29234 4.2b 5.9S7 93

S8 109a fcu = cube compressive strength; Ec = elastic modulus; fct splitting strength; ER = modulus of rupture.b Based on one test result.

which were determined in accordance with the BS 1881 suite ofstandards [29–32], are reported in Table 2. The age of the concretewhen themechanical property testswere conducted, in addition tothe age of the slabs upon testing, are also reported in Table 2. The10% increase in concrete compressive strength over the duration ofthe experimental programwas expected to have aminor influenceon the test results as none of the test slabs failed in compression.

The yield stress (0.2% proof stress) and elastic modulus of thesteel reinforcing bars, which were averaged from three specimenstested in accordance with BS EN 1000201:2001 [33], were foundto be 566 MPa (standard deviation, sd. = 5 MPa) and 198 GPa(sd. = 3.8 GPa) respectively.

Two layers of fibre sheet were used to form flat FRP couponsof 30 mm width. Five coupons were prepared and then testedto failure in accordance with ACI 440.3R-04 [34] to produce anelongation at rupture of 14,674 µε (1.5%) (sd. = 402 µε), tensilestrength at rupture of 3163 MPa (sd. = 206 MPa) and elasticmodulus of 239 GPa (sd. = 6.8 GPa).

The mechanical properties of the epoxy were tested inaccordance with the 527 series of documents published by BS ENISO 527:1996 [35]. The results of seven specimens were averagedto produce an elongation at rupture of 6716 µε (0.7%) (sd. =

391 µε), tensile strength at rupture of 28.3 MPa (sd. = 1.4 MPa)and elastic modulus of 4273 MPa (sd. = 126 MPa).

3. Experimental results

3.1. Load–deflection responses and load–deflection–strain summary

The total load (P) versus mid-span deflection responses for allthe eight slab tests are shown in Fig. 8(a). In order to furtherenhance the clarity of these results, the slabswhich experienced anincrease in both load and deflection are shown in Fig. 8(b)while theslabs which experienced a predominant increase in deflection onlyare shown in Fig. 8(c). The influence on strength and deflection ofthe FRP-strengthened RC slabs by the FRP anchorage can be clearlyobserved. Table 3 provides a summary of the enhancement in peakload and corresponding peak deflection of all slab specimens aswell as themaximummeasured compressive strain in the concreteand measured tensile strain on the FRP. Table 3 also provides asummary of the post-peak results which refer to the behaviour ofthe slabs, once the large and sudden drop in load had occurred dueto complete plate debonding. Slab S8 offers the highest increase inload and Slab S7 offers the highest increase in central deflection.

The load–deflection responses of Slabs S2–S8, prior to completedebonding of the FRP strengthening, were predominantly of a tri-linear nature [36]. The two main turning points, which definethe ends of the first two linear portions, are due to the concretecracking and yield of the internal tension steel reinforcement. Theresponse of the third portion was a direct consequence of the FRP

Fig. 8. Load–deflection responses of anchored slabs relative to control Slabs S1 andS2: (a) all anchored slabs; (b) anchored slabswith increasing strength anddeflection(c) anchored slabs with predominantly increasing deflection.

strengthening as well as the influence of the FRP anchors. Thedramatic drop in load upon the peak load being reachedwas due tocomplete debonding of the FRP plate on one side of the slab. Mostslabs then sustained a post-peak reserve of strength. This reservewas due to frictional resistance from the intact anchors clampingthe debonded plate to the rough concrete substrate failure plane.


Table 3Summary of load and deflection (mid-span) enhancement and peak strains.

Slab Load (peak) Deflection (peak) Load (post-peak) Strain (peak)Peak, P (kN) P/PS1a (%) P/PS2b (%) Peak, δ (mm) δ/δS2

b (%) Peak, Ppp (kN) Ppp/Ppp,S1a (%) εconc (µε) (concrete)d εfrp (%c) (µε) (FRP)

S1 20.32 0 NA NA NA 20.32 0 −1872 NAS2 41.66 105 0 25.53 0 20.90 3 −1237 6649 (45.3%)S3 51.22 152 23 41.58 63 26.87 32 −1566 7676 (52.3%)S4 43.90 116 5 37.34 46 23.82 17 −1543 8025 (54.7%)S5 51.80 155 24 41.99 64 26.20 29 −1300 8884 (60.5%)S6 40.59 100 −3 31.78 24 22.47 11 −1141 6696 (45.6%)S7 51.47 153 24 53.65 110 19.78 −3 −2564 11566 (78.8%)S8 54.27 167 30 48.76 91 22.35 10 −1528 11348 (77.3%)

NA = not applicable.a % increase over unstrengthened control (S1); e.g. (P − Ps1)/Ps1 × 100%.b % increase over unanchored control (S2); e.g. (P − Ps2)/Ps2 × 100%.c % of flat coupon capacity (flat coupon capacity = 14,674 µε).d compressive (−’ve) strain.

All anchored slabs (i.e. Slabs S3 to S8) except Slab S2 were onthe whole stiffer on account of the anchor fan fibres increasing thethickness of the FRP plate. The relative enhancement in stiffnesswas a function of the amount of fan fibre (i.e. dependent on thenumber of anchors) with the stiffer slabs experiencing slightlyhigher cracking and steel yielding loads. The maximum mid-span deflection of all anchored slabs prior to complete FRP platedebonding was about 54 mm. As it served no purpose in this studyto then displace the slabs until the concrete crushed, the loadingwas stopped and released once amid-spandeflection of 80mmhadbeen reached. At this stage the strain in the compressive concretewas on average about 1600 µε and well below its crushing strain(except Slab S7 due to high deflection).

The level of utilisation of the tensile strain capacity of the FRPwas 45% for the unanchored control slab (Slab S2). This level wasincreased by up to 79% (i.e. Slab S7) upon the addition of FRPanchors.

3.2. Behaviour and failure modes

Fig. 9 provides a schematic representation of the debondedregions for all anchored slab tests in addition to the condition of theanchors immediately following complete FRP plate debonding. Theanchors were found to fail in one of three different modes, namely(i) complete rupture at the bend region, (ii) partial rupture atthe bend region, and (iii) pull-out from the concrete substrate.Fig. 10(a) shows the condition of a typical completely rupturedanchor while Fig. 10(b) shows a pulled-out anchor.

The following comments are offered in support of the behaviourand failure of all test slabs in the context of the results containedin Figs. 8 and 9 and Table 3.

3.2.1. Slab S1The unanchored–unstrengthened control Slab S1 behaved in a

manner befitting that of an under-reinforced ductilemember. Firstcracking of the tensile concrete occurred at a load of 11 kN andthen the internal tension steel yielded at a load of 19 kN. The slabthen deflected extensively until the test was stopped at 80 mmof mid-span deflection. Theoretical calculations [36] estimated theconcrete to crush (i.e. concrete strain to reach 3000 µε) at 95 mmof mid-span deflection. In addition, the corresponding strain inthe tension steel at concrete crushing was 4.2% which fits withinthe ACI 318M-05 [37] definition of a tension-controlled failure (i.e.ductile section).

3.2.2. Slab S2The flexurally strengthened but unanchored control Slab S2

failed by IC debonding of the FRP plate. Such a failure mode hasbeen well documented in the open literature [1,2,38]. In the SlabS2 test, first cracking occurred at a load of 13 kN with tension

steel reinforcement yielding at a load of 34 kN. The maximummeasured tensile strain on the FRP was about 6650 µε (Table 3)upon complete debonding of the FRP. This strain is identical tothe predicted debonding strain of 6650 µε from Teng et al. [38](using their best-fit factor of 0.753) and close to 7650 µε ascalculated from ACI 440.2R-08 [28]. The FRP plate increased theload capacity of the slab, compared to Slab S1, by 105%. Uponcomplete debonding, the FRP was no longer effective, and Slab S2then behaved in a very similar manner to the control Slab S1.

3.2.3. Slab S3Localised debonding of the FRP initiated at a load of ap-

proximately 39 kN (upon observation of the first drop in theload–deflection response) at one of the applied load positionsonce the third linear portion of the load–deflection response hadbeen reached (Fig. 8). The gradient of the third portion of theload–deflection response then started to decrease as the debond-ing crack propagated. The FRP anchors, however, decreased therate of propagation of debonding compared to Slab S2. The platethen completely debonded at a load of 23% and deflection of 63%in excess of the respective load and deflection at debonding in SlabS2. Upon complete plate debonding, the load suddenly decreasedalthough at this stage none of the anchorswere found to have com-pletely failed. The three anchors adjacent to the debonded plateend had partially ruptured (Fig. 9(a)) while the fourth anchor fromthe plate end had pulled-out due to a major crack passing rightthrough the anchor dowel region (Fig. 10(b)). The stepped descentof the load–deflection responsewas believed to be due to the resid-ual restraint offered by the intact FRP anchors. The slab then sus-tained a post-peak reserve of strength 32% in excess of the strengthof control Slab S1.

3.2.4. Slab S4The anchorage scheme of Slab S4 involved the removal of

every second anchor from the Slab S3 anchorage scheme. Thistranslated into Slab S4 having 55% of the number of anchors asSlab S3. In addition, the spacing between the anchors in Slab S4was twice that over Slab S3. The effect of such an arrangementof anchors in Slab S4 reduced the slope of the third portion ofthe load–deflection response to virtually zero once the load anddeflection corresponding to failure of the control Slab S2 wasreached. More specifically, debonding first initiated at a load of40 kN (deflection 23 mm) as observed from the large drop in theload in Fig. 8. When the debonding crack had propagated to theplate end, the central deflection had increased by 46% above SlabS2. Once the plate completely debonded, the load dropped downto the capacity of the plain control Slab S1 and at this stage noanchors had failed (Fig. 9(b)). The post-peak reserve of strengthwas increased by 17% in excess of the capacity of Slab S1. Theshear span anchors in the debonded plate portion were observedto be partially ruptured when the load was stopped at a mid-spandeflection of 80 mm as shown in Fig. 9(b).


Fig. 9. FRP plate and FRP anchor conditions post-plate debonding. { debonding crack; Type 1 FRP anchor; Type 2 FRP anchor; - - - load point; CMR= constantmoment region; completely ruptured anchor; partially ruptured anchor; pulled-out anchor; undamaged anchor}.

a

b

Fig. 10. Typical FRP anchor failure modes: (a) completely ruptured anchor (Slab S8, anchor no. 1); (b) pulled-out anchor (Slab S3, anchor no. 4).

3.2.5. Slab S5Slabs S5 and S3 were identical apart from the omission of

three anchors in the constant moment region in the former. Bothslabs therefore had exactly the same number and same type ofanchors in the shear span. As a result, Slab S5 followed virtuallythe same load–deflection response as Slab S3, however, the formerwas on the whole slightly less stiff than the latter due to the

difference of anchors in the mid-span region. The peak loadand deflection of Slab S5 was about 1% to that of Slab S3. Thesimilarity in slab behaviour between the anchorage layouts forSlabs S3 and S5 suggests that anchors located in the shear spanare the most effective. The effect of anchorage in the constantmoment region can be speculated to have enabled the steppedunloading response observed in Slab S3. The large release of energy


a

b

c

Fig. 11. Typical FRP anchor conditions after complete plate debonding: (a) Type 1anchors (Slab S5); (b) Type 2 anchors (Slab S7); (c) Hybrid Type 1 and Type 2 anchors(Slab S8).

associated with complete debonding of the plate (and the lackof a stepped descent) caused the four anchors in the debondedplate end to partially rupture (Fig. 9(c)) and Fig. 11(a) shows thetest specimen immediately after plate debonding. Unfortunatelythe LVDT measuring plate end slip at the debonded plate enddetached due to the large release of energy upon plate debonding.Measurements made during the test with a hand-held scalerevealed the slip at the plate immediately post-plate debonding tobe in excess of 5 mm. The ability of the anchors to cater for suchplate slip was due to the ability of the fibres in the bend regionto deform. This was a direct result of not impregnating the bendregion fibres with epoxy as discussed in Section 2.3. Finally, thepartially ruptured anchors and the remaining intact anchors wereable to contribute to the 29% post-peak reserve in strength overSlab S1.

3.2.6. Slab S6Slab S6 contained the least amount of anchorage and the two

anchors positioned at each end of the FRP proved to be leasteffective in enhancing the load and deflection. Just prior to thesharp drop in load in the load–deflection response of Slab S6at 40 kN, the plate had been progressively debonding from theconcrete substrate in a similar manner to that of Slab S2. After all,Slabs S6 and S2 were identical apart from the end anchorages inthe former. Once the debonding crack had become unstable, thecrack then rapidly propagated to the inner anchor, thus accountingfor the drop of load at a deflection of 25 mm. The load was thensteadily increased again until the capacity of Slab S2 was reachedapproximately, upon which the debonding crack passed the FRPanchors and the plate subsequently completely debonded. Theanchor at the debonded plate end completely ruptured, however,the second anchor from the plate end had partially ruptured(Fig. 9(d)). Slab S6 then sustained a small post-peak reservestrength of at most 11% above Slab S1.

3.2.7. Slab S7The logic behind this anchorage scheme was to have anchors

spaced closer together in a bid to further slow down the rate ofpropagation of the debonding crack. Slab S7 therefore containedanchorage spread over the same shear span region as Slab S5,however, the number of anchors in Slab S7 was doubled and theamount of fibre used tomake each anchorwas halfwhen comparedwith Slab S5.

The load–deflection response of Slab S7 was softer than SlabS5. This was due to the Type 2 anchors of Slab S5 containingless anchor fan fibre than the Type 1 anchors of Slab S5. Theanchorage scheme of Slab S7 caused the third linear portion ofthe load–deflection response to considerably soften compared tothat of Slab S5, although the closer spaced anchors were successfulin slowing down the propagation of the debonding crack enoughfor a 28% enhancement in central deflection to be achieved overSlab S5. Interestingly, the load prior to complete debonding of theFRP plate was virtually identical to the peak load of Slab S5. Uponcomplete plate debonding, the sudden release of energy causedall the anchors in the critical shear span to completely rupture asshown in Figs. 9(e) and 11(b). The reduced amount of fibre used inthe Type 2 anchors caused them all to rupture and the failure ofSlab S7 was the most catastrophic one out of all the anchored slabtests reported herein. No post-peak reserve of strength was able tobe maintained on account of all the anchors failing. Fig. 11(b) alsoshows that debonding occurred at the FRP-to-concrete interface inthe concrete. In fact, all FRP-strengthened slabs (S2–S8) failed atthe same desired FRP-to-concrete interface in the concrete.

3.2.8. Slab S8Slab S8 was anchored with a combination of anchor types

(i.e. Type 1 and Type 2) and anchor spacings. The anchorage wasdesigned after observing the behaviour of all preceding anchoredslab tests S3–S7. Slabs S3 and S5 demonstrated the effectivestrength gains caused by Type 1 anchors positioned in the shearspan region. The results of Slab S7 showed the effectiveness ofcloser spaced anchors in enhancing deflection. The closer spacedanchors of Slab S7, however, contributed to strength gains atgreater deflections (i.e. a softer load–deflection response) overSlabs S3 and S5. Finally, the results of Slabs S4 and S6 showed thatanchors spaced too far apart did not lead to significant strengthgains but did lead to deflection gains. As a result, the total amountof fibre used to make all the anchors in Slab S8 was identical to theamount of fibre used to make all the anchors in Slabs S5 and S7.The resulting arrangement of anchors in Slab S8 produced the earlystrength gains in the third linear portion of the load–deflectionresponse associated with Type 1 anchors near the peak moment


a

b

Fig. 12. Propagation of debonding cracks (Slab S8): (a) load–deflection response (Slabs S1, S2, S8); (b) debonding cracks at specific level of load anddeflection. { debondingcrack; Type 1 FRP anchor; Type 2 FRP anchor; - - - load position; CMR = constant moment region; completely ruptured anchor; partially ruptured anchor;

undamaged anchor}.

region (Fig. 8). The increase in deflection of about 16% above thatobserved in Slabs S3 and S5 proved the effectiveness of the closespaced Type 2 anchors nearer to the ends of the FRP plate. SlabS8 achieved the greatest enhancement in strength (i.e. 30% abovethat of Slab S2) and the second greatest mid-span deflection. Uponcomplete debonding of the plate, all the Type 2 anchors ruptured(Figs. 9(f) and 11(c)) however the partially and fully intact Type 1anchors contributed to the post-peak reserve of strength gainof 10%.

3.3. Debonding crack propagation

The presence of the anchors influenced the rate of propagationof debonding cracks along the length of the FRP plate. Overall,debonding cracks were found to propagate more slowly foranchors spaced more closely together. For example, the rateof crack propagation for Slab S7 was much reduced than thatof Slab S6. As a result, Slab S7 was able to achieve muchgreater enhancement in strength and deflection than Slab S6. Theenhancement in strength in this case was also due to the use ofmore FRP anchors. When the same total amount of fibre was usedto cross the shear plane, the closer spaced anchors in Slab S7 were

found to increase the deflection of the slab over the larger spacedanchors in Slab S5. The strength enhancements for Slabs S5 and S7were, however, virtually identical (see column 4 of Table 3).

Fig. 12 provides a detailed account of the propagation ofdebonding cracks in relation to the applied load (and deflection)for Slab S8. This slab is singled out as it is considered one ofthe most optimally designed (i.e. greatest strength enhancement,second greatest mid-span deflection, partially intact anchoragepost-plate debonding). In Fig. 12, the extent of debonding inaccordance with the level of load and deflection are related toeach other and the fifteen data points correspond to observedpropagation of the debonding cracks. However, it is important tostress the limitations of presenting such data. First, it is basedon visual observation. Second, one edge of the FRP plate wasmonitoredwith strain gauges. The significance of this second pointis that the extent of debonding through the width of the platemay not be uniform. Regardless, Fig. 12 provides a qualitativeoverview of the debonding crack propagation process. Overallcomments to be made include (i) debonding initiates well beforecomplete plate debonding (i.e. adequate warning of failure isprovided), (ii) debonding initiates near the loaded region andpropagates towards the free ends of the FRP plate, (iii) debonding


a

b

Fig. 13. FRP strain distributions for Slab S8: (a) complete load range; (b) selected load levels (no bond, partial debond, near complete debond) {CMR = constant momentregion}.

in the anchored regions propagates in lengths approximatelyequal to the anchor spacing, and (iv) complete debonding of theplate signifies a complete loss of load carrying capability of thestrengthened slab although post-peak reserves of strength (abovethe unstrengthened slab) due to intact anchorage can be sustained.

Further inspection of Fig. 12 leads one to the conclusionthat the addition of FRP anchors introduces a considerabledegree of robustness into the FRP-strengthened slabs by allowingthe debonding crack growth process to be controlled. Duringdebonding, the internal tension steel has also yielded. Amplewarning of distress is thus provided well before completedebonding of the FRP.

3.4. Load–strain response

To enable comparison with the results contained in Fig. 12,the distribution of strain along the FRP for the same Slab S8 isprovided in Fig. 13. In Fig. 13(a), the strains at load levels of10–40 kN are provided in addition to the strains at the 15 differentdebonding crack levels identified in Fig. 12. Fig. 13(b) providesselected results from pre-debonding, early debonding, to near finaldebonding as well as corresponding drawings showing the extentof debonding cracks. Strain gauge results are notoriously difficultto interpret especially once the concrete has cracked and the FRP

has debonded. The following observations are, however, made inlight of Fig. 13. The FRP anchor fan influences the strain results (thestrain in the anchored regions is less due to the use of more FRPmaterial). Also, the strain profile becomes flatter as the debondingcracks propagate to the ends of the plate. The strain profile doesnot completely flatten though when a large proportion of theFRP has debonded (i.e. debond crack 14 in Fig. 13(b)) due to theclamping effect of the FRP anchor which sustains interaction alongthe debonding crack by aggregate interlock and friction.

4. Conclusions

The effectiveness of FRP anchors in increasing the strengthand deflection of FRP flexurally strengthened RC slabs has beenreported in this paper. The greatest increase in strength anddeflection recorded, over the unanchored but strengthened controlcounterparts, was 30% and 110%, respectively. In addition, theusable strain in the FRP plates was increased from 45% of thecapacity of a flat coupon (for the unanchored but strengthenedcontrol slab) to almost 80% (for optimally designed anchorageschemes). The experimental results have revealed the followingissues of importance.

1. Anchorage of the FRP strengthening plate with FRP anchors canbuild robustness into the member.


2. Anchors positioned in the shear span were found to be mosteffective (anchors in the constant moment region were largelyineffective).

3. Closer spaced anchors were found to reduce the rate of debond-ing crack propagation and also enabled higher deflections to beachieved.

4. Anchors spaced far apart led to gains in deflection capacity butlimited gains in strength.

5. Anchors of greater fibre content positioned closer to the peakbending moment region in addition to anchors of lesser fibrecontent but spaced close together near the free ends of theFRP plate produced the greatest enhancement in strength withsignificant deflection capability.

Acknowledgements

Funding provided by the Hong Kong Research Grants Councilin the form of General Research Fund Grant HKU 716308E isgratefully acknowledged. The following individuals from TheUniversity of Hong Kong are thanked for their assistance withthe experimental work: Messers Huawen Zhang (Ph.D. candidate),Jiaqi Yang (M.Phil. candidate), Kin Lung Jason Li (2009–10undergraduate final year project student), and the LG laboratorytechnicians. In addition, Mr. Jacob Schmidt (Ph.D. candidate) ofthe Technical University of Denmark is thanked for his assistancewith the experimental work and also for producing the three-dimensional drawings.

References

[1] OehlersDJ, SeracinoR.Design of FRP and steel platedRC structures: retrofittingbeams and slabs for strength, stiffness and ductility. UK: Elsevier; 2004. 228pages.

[2] Hollaway LC, Teng JG. Strengthening and rehabilitation of civil infrastructuresusing fibre-reinforced polymer (FRP) composites. Cambridge (UK):WoodheadPublishing Limited; 2008.

[3] Sharif A, Al-Sulaimani GJ, Basunbul IA, BaluchMH,Ghaleb BN. Strengthening ofinitially loaded reinforced concrete beams using FRP plates. ACI Struct J 1994;91(2):160–8.

[4] Yun YC, Wu YF, Tang WC. Performance of FRP bonding systems under fatigueloading. Eng Struct 2008;30(11):3129–40.

[5] Smith ST, Teng JG. Shear-bending interaction in debonding failures of FRP-plated RC beams. Adv Struct Eng 2003;6(3):183–99.

[6] Gose SC, Nanni A. Anchorage system for externally bonded FRP laminatesusing near surfacemounted rods. Report, centre for infrastructure engineeringstudies. USA: University of Missouri-Rolla; 2000.

[7] Smith ST. FRP anchors: recent advances in research and understanding. In:Proceedings, second Asia-Pacific conference on FRP in structures. APFIS 2009.2009. p. 35–44.

[8] Ceroni F, Pecce M, Matthys S, Taerwe L. Debonding strength and anchoragedevices for reinforced concrete elements strengthened with FRP sheets.Composites B 2008;29(3):429–41.

[9] Su RKL, Siu WH, Smith ST. Effects of bolt–plate arrangements on steel platestrengthened reinforced concrete beams. Eng Struct 2010;32(6):1769–78.

[10] Smith ST. Strengthening of concrete, metallic and timber constructionmaterials with FRP composites. In: Proceedings, fifth international conferenceon FRP composites in civil engineering. CICE 2010. 2010. p. 13–9.

[11] Kim SJ, Smith ST. Behaviour of handmade FRP anchors under tensile load inuncracked concrete. Adv Struct Eng 2009;12(6):845–65. Special issue on bondbehaviour of externally bonded FRP reinforcement.

[12] Kim SJ, Smith ST. Pullout strength models for FRP anchors in uncrackedconcrete. J Compos Constr, ASCE 2010;14(4):406–14.

[13] Orton SL, Jirsa JO, Bayrak O. Design considerations of carbon fibre anchors.J Compos Constr, ASCE 2008;12(6):608–16.

[14] Ozbakkaloglu T, Saatcioglu M. Tensile behaviour of FRP anchors in concrete.J Compos Constr, ASCE 2009;13(2):82–92.

[15] Oh H-S, Sim J. Interface debonding failure in beams strengthened withexternally bonded GFRP. Compos Interfaces 2004;11(1):25–42.

[16] Micelli F, Rizzo A, Galati D. Anchorage of composite laminates in RC flexuralbeams. Struct Concr, FIB 2010;11(3):117–26.

[17] Lam L, Teng JG. Strength of RC cantilever slabs bonded with GFRP strips.J Compos Constr, ASCE 2001;5(4):221–7.

[18] SeliemHM, Seracino R, Sumner EA, Smith ST. A case study on the restoration offlexural capacity of continuous one-way slabs with cutouts. J Compos Constr,ASCE 2010 [in press].

[19] Widianto Bayrak O, Jirsa JO, Tian Y. Seismic rehabilitation of slab–columnconnections. ACI Struct J 2010;107(2):237–47.

[20] Ozcan O, Binici B, Ozcebe G. Seismic strengthening of rectangular reinforcedconcrete columns using fiber reinforced polymers. Eng Struct 2010;32(4):964–73.

[21] Antoniades KK, Salonikios TN, KapposAJ. Evaluation of hysteretic response andstrength of repaired R/C walls strengthened with FRPs. Eng Struct 2007;29(9):2158–71.

[22] Tan KH, Patoary MKH. Strengthening of masonry walls against out-of-planeload using fiber-reinforced polymer reinforcement. J Compos Constr, ASCE2004;8(1):79–87.

[23] Teng JG, Lam L, Chan W, Wang J. Retrofitting of deficient RC cantilever slabsusing GFRP strips. J Compos Constr, ASCE 2001;4(2):75–84.

[24] Eshwar N, Ibell TJ, Nanni A. Effectiveness of CFRP strengthening on curvedsoffit RC beams. Adv Struct Eng 2005;8(1):55–68.

[25] Brunckhorst L, Knudsen PJ, Poulson E, Thorsen T. Forstærkning af betonkon-struktioner med bolte-limede kulfiberbånd. En elementœr, teknisk introduk-tion. Esbjerg (Denmark): Rosendahls Bogtrykkeri A/S; 2007 [in Danish].

[26] Zhang HW, Smith ST, Kim SJ. Optimisation of carbon and glass FRP anchordesign. Constr Build Mater, FRPRCS9 special edition. 2010 [in press].

[27] El-Reedy M. Steel-reinforced concrete structures. Assessment and repair ofcorrosion. Florida (USA): CRC Press; 2008.

[28] ACI 440.2R-08. Guide for the design and construction of externally bonded FRPsystems for strengthening concrete structures. Farmington Hills (MI, USA):American Concrete Institute. ACI; 2008.

[29] BS 1881-116. Method for determination of compressive strength of concretecubes. UK: British Standards; 1983.

[30] BS 1881-121. Method for determination of static modulus of elasticity incompression. UK: British Standards; 1983.

[31] BS 1881-117. Method for determination of tensile splitting strength. UK:British Standards; 1983.

[32] BS 1881-118. Method for determination of flexural strength. UK: BritishStandards; 1983.

[33] BS EN 10002-1:2001. Metallic materials—tensile testing. UK: British Stan-dards; 2001.

[34] ACI 440.3R-04. Guide test methods for fiber-reinforced polymers (FRPs) forreinforcing or strengthening concrete structures. Farmington Hills (MI, USA):American Concrete Institute. ACI; 2004.

[35] BS EN ISO 527:1996. Plastics—determination of tensile properties. UK: BritishStandards; 1999.

[36] Smith ST, Kim SJ. Deflection calculation of FRP-strengthened RC flexuralmembers. Aust J Struct Eng 2010;11(2):75–86. Special edition on fibrereinforcement in concrete structures.

[37] ACI 318M-05. Building code requirements for structural concrete andcommentary. Farmington Hills (MI, USA): American Concrete Institute. ACI;2005.

[38] Teng JG, Smith ST, Yao J, Chen JF. Intermediate crack induced debonding in RCbeams and slabs. Constr Build Mater 2003;17(6–7):447–62.

Documents

FRP-strengthened RC Slabs Anchored With FRP Anchors