1、 附录一 论文原文 ACI STRUCTURAL JOURNAL TECHNICAL PAPER Title no. 107-S61 Creep Effects in Plain and Fiber-Reinforced Polymer-Strengthened Reinforced Concrete Beams by M. M. Reda Taha, M. J. Masia, K.-K. Choi, P. L. Shrive, and N. G. Shrive The long-term deflection behavior of two reinforced concrete (RC)
2、beams with similar dimensions and material properties wasmonitored. One beam was externally strengthened with fiberreinforcedpolymer (FRP) strips, whereas the other was used as acontrol specimen. Both beams have been subjected to sustainedloading for over 6-1/2 years. The objective of the experiment
3、s wasto assess the significance of creep in the epoxy adhesive and whethersuch creep might allow the FRP strips to unload over time. Slipmovements at the ends of the FRP strips were also monitored. The experimental deflections have been compared to deflectionpredictions using ACI 209R-92 and CEB-FIP
4、 MC 90. The creepdeformations of the FRP-strengthened beam are not as predictedfrom the control beam. Two analytical approaches are used: astep-by-step in-time analysis and finite element (FE) modeling. Both techniques demonstrate that creep of the adhesive layer canaccount for the differences obser
5、ved between the predicted andactual behaviors of the beams. Keywords:creep； deflection； epoxy adhesive； fiber-reinforced polymer(FRP)； reinforced concrete. INTRODUCTION In recent years there has been much research on the use offiber-reinforced polymers (FRPs) to strengthen existingconcrete structure
6、s. One popular application, used widely inpractice, is to bond FRP strips externally to the tension face ofreinforced concrete (RC) beams to increase flexural capacity. The FRP strips are typically bonded directly to theprepared concrete surface using an epoxy adhesive. Thestrips may be anchored mec
7、hanically near their ends orsupported by additional shear reinforcement, usually in theform of U-shaped FRP sheets. If the beam is subsequentlyloaded with sustained loads, creep in the epoxy adhesivecould take place and would allow the FRP strips to unload, leaving them ineffective against the susta
8、ined load. Similarly, if the strips are prestressed as recently recommendedby some researchers (for example, Ye et al.1), creep in theepoxy may relieve some of the initial force. Hence, althoughthe FRP strips can still assist in supporting additional live load, the increased sustained loads may exce
9、ed the capacity of whatcould effectively become the original unstrengthened beam. Research into the time-dependent behavior (creep andshrinkage) of concrete beams strengthened with externallybonded FRPs is scarce. Analytical models were verified againstlimited experimental observations of RC2 and ti
10、mber3 beamsexternally reinforced with FRP strips. Similar approaches wereused for a composite glass fiber-reinforced polymer (GFRP)box girder with concrete in the compression flange and a carbonfiber-reinforced polymer (CFRP) strip bonded to the tensionface.4 In all of the aforementioned models, how
11、ever, the effectof creep in the adhesive layer bonding the FRP to the tensionface of the beams was neglected. That is, perfect bond and straincompatibility was assumed between the substrate and the FRP. Recent experiments by Choi et al.5 demonstrated thatsignificant creep under shear stresses occurs
12、 in the epoxy at theconcrete-FRP interfaces when loading is applied within 7 daysof epoxy application. Herein, the results of an experimental investigation andaccompanying analytical predictions of immediate and timedependentbeam deflections are described. The construction ofthe RC beams and the exp
13、erimental program for observing theirtime-dependent deflection are presented. The measureddeflections are compared to deflection predictions usingthe ACI and CEB-FIP methods implemented according tothe recommendations of Hall and Ghali.6 The long-term deflection data show that the timedependent (cre
14、ep) deformation of the CFRP-strengthenedbeam is a larger proportion of its immediate deformationthan the same deformation ratio for the unstrengthenedbeam. The creep of the beam with the FRP strips could notbe predicted from the creep measured on the plain beamwhen creep of concrete alone was consid
15、ered. BecauseCFRPs have not been observed to creep at the stress levelsgenerated,7,8 the additional creep may have occurred in theepoxy adhesive bonding the FRP strips to the concrete. The creep mechanism is expected to be a simple flow ofthe epoxy under the shear stress,5 which develops to createte
16、nsion in the FRP strip. While models exist for predictinglong-term deformation of RC beams strengthened with FRP (for example, Charkas et al.9), these models do not accountexplicitly for creep of epoxy adhesives. Herein, we use twodifferent approaches to determine if creep in the epoxy canaccount fo
17、r the different behaviors observed in the beams: astep-by-step in-time analysis allowing incremental creep ofconcrete and epoxy in each time step and enforcingequilibrium at the end of the time step, and finite element (FE) modeling with shear flow allowed in the epoxyadhesive layer. RESEARCH SIGNIF
18、ICANCE The potential effects of creep on RC beams strengthenedwith externally applied FRP strips are considered. It wasthought that creep in the epoxy resin might relieve stress inthe FRP, making the FRP less effective from a serviceabilitypoint of view under sustained loads. Thus, FRP strips used t
19、ostrengthen a beam, which was then subject to increasedsustained load, might end up with the extra sustained loadbeing carried by the original concrete and steel reinforcement, not the FRP. The experimental and analytical work performedrevealed that the situation is more complex. Nevertheless, creep
20、 deflections are greater than predicted from the creep ofconcrete alone, indicating contributions from creep of theepoxy. The reported experimental program was designed to identify the existence of epoxy creep rather than replicate apractical retrofit scenario. The results highlight the potentialfor
21、 epoxy creep to affect the long-term performance of FRPretrofits in practice. EXPERIMENTAL PROGRAM Test specimens and materials Two similar RC beams were cast from the same concretebatch (Fig. 1). Each beam was 3500 mm (137.8 in.) long, 280 mm (11.02 in.) wide, and 180 mm (7.09 in.) high, reinforced
22、 with four longitudinal bars (Canadian 10M-11.3 mm 0.445 in. diameter, 100 mm2 0.155 in.2 area) at an effective depth of 135 mm (5.31 in.) from the topsurface of the beam. Seven 10M stirrups were spaceduniformly in each shear span of each beam. The 28-daycompressive strength of the concrete, as dete
23、rmined from100 mm (4 in.) diameter, 200 mm (8 in.) high cylinderscastfrom the same batch of concrete as the test beamswas34.3 2.3 MPa (4900 328 psi). Fig. 1 Test specimens, test setup, and strain distribution. The two beams were cast together and storedfullysupportedfor 10 months before the CFRP str
24、ips and GFRPwraps were applied. One beam (Beam 1) was designated asa control specimen. Two CFRP strips were bonded to thetension face of the second beam (Beam 2) using an epoxyadhesive. The strips are 100 mm (3.94 in.) wide, 1.2 mm (0.047 in.) thick, and 2970 mm (116.9 in.) long. Over theshear spans at each end of Beam 2, GFRP sheets werewrapped in a U-shape to cover the two side faces and thetension face of the beam. The CFRP strips are unidirectionalwith the fibers aligned along the length of the beam. Thestrips have a modulus of
