Seismic retrofitting of severely damaged RC connections made with recycled concrete using CFRP sheets

Abstract

An experimental and numerical program is carried out in this research to investigate the influence of CFRP sheets on the cyclic behavior of unconfined connections made with recycled concrete. Cement is partially replaced by silica fume, iron filling and pulverised fuel ash using two different percentages: 15% and 20%. Each specimen is partially loaded at the first stage and then specimens are repaired using CFRP sheets. The repaired specimens are then laterally loaded until failure. In addition, a finite element model is built in ABAQUS and verified using the experimental results. The experimental results have shown that the repaired specimens have regained almost double the capacity of the un-repaired specimens and hence the adopted repair configuration is recommended for retrofitting seismically vulnerable RC connections. Increasing cement replacement percentage by silica fume, fuel ash or iron filling from 15% to 20% has reduced joint carrying capacity and weakened the joint. It is recommended using 15% pulverised fuel ash or silica fume as cement partial replacement to enhance the strength and ultimate drift of beam-column joints under cyclic loading. Iron filling concrete is also recommended but the enhancement is relatively less than that found with pulverised fuel ash concrete and silica fume concrete.

Keywords

retrofitting / CFRP sheets / recycled concrete / pulverised fuel ash / silica fume / cyclic / beam-column connections

Introduction

Many existing reinforced concrete (RC) structures are seismically vulnerable due to insufficient shear reinforcement in joint panel, beams and columns [¹]. Joint shear failure is a brittle failure and has caused building collapse during earthquake events. Experimental studies [²–¹⁰] and numerical models [¹¹] have been done to investigate the behavior of beam-column connections strengthened with carbon fiber reinforced polymer (CFRP) plates and sheets under cyclic loads.

The repair of seismically vulnerable RC structures using externally bonded fiber composite materials (FRP) has been expanded. Retrofitting existing structures can increase their load carrying capacity, ductility and reduce their vulnerability. Fibers have advantages over other retrofitting materials due to their high strength to weight ratio, high resistance to corrosion, excellent durability, ease and flexibility in strengthening seismically vulnerable members [¹²].

Previous experimental studies [¹³–¹⁶] have shown that CFRP configuration has significant effect on the load carrying capacity, ductility and mode of failure of RC element. CFRP sheets and plates became an acceptable solution in retrofitting existing structures and they have proven their efficiency in increasing the load carrying capacity and ductility of RC elements. However, this enhancement is significantly influenced by CFRP configuration and orientation.

In addition, a great deal of research has been carried out over the last three decades to investigate the mechanical behavior of recycled concrete specimens. However, only few studies have been conducted to investigate the structural performance of recycled concrete elements under cyclic loading. Different types of waste materials have been recently used to enhance the mechanical and durability characteristics of concrete such as iron fillings, rice husk ash, fly and fuel ash, etc., where the enhancement ratio varies based on the type and percentage of the implemented material. Noori and Ibrahim [¹⁷] have replaced sand by iron waste in concrete cylinder specimens and they found that 12% is the optimum replacement percentage that increases compressive and flexural strength of iron waste concrete while increasing this percentage has decreased concrete strength. Other researchers [¹⁸] have found that the fracture behavior of silica fume concrete is brittle while others shown that silica fume can enhance the flexural and compressive strength of concrete [¹⁹]. Pulverised fuel ash concrete achieved comparable strength to ordinary concrete at earlier ages, with much higher strengths at later ages [²⁰].

An experimental program is conducted by Murad et al. [²¹,²²] to investigate the cyclic behavior of RC connections using different waste materials (pulverized fuel ash, silica fume and iron fillings) as cement partial replacement. The specimens are tested under partial cyclic loading. The experimental results have shown that the proper replacement percentage of cement is 15% for the pulverized fuel ash, silica fume and iron fillings.

This research experimentally and numerically investigates the cyclic behavior of rehabilitated RC beam-column connections that made with recycled concrete using new configuration of CFRP sheets. The novelty part is represented by the fact that no other tests available in the literature that study the behavior of repaired beam-column joints made with iron-filling concrete, silica fume and fuel ash concrete under full cyclic loading. Furthermore, the adopted repair configuration with CFRP sheets is limited in the literature. In addition, numerical modeling has been conducted using ABAQUS, finite element analysis software, to better explain the cyclic behavior of repaired RC connections made with recycled concrete. The finite element model is verified using the experimental results of the test specimens. In addition, numerical and experimental comparisons are presented in order to evaluate the influence of the adopted CFRP sheets configuration on the cyclic behavior of repaired RC connections made with recycled concrete.

Experimental program

Materials and methods

This project experimentally and numerically investigates the structural behavior of repaired unconfined beam-column connections made with recycled concrete under cyclic loading. CFRP sheets are used to repair the specimens where sheet thickness is 0.1667 mm and its tensile strength and modulus of elasticity are 3800 MPa and 240 GPa, respectively. The sheets are epoxy bonded to the concrete surface. Furthermore, different waste materials including pulverized fuel ash, silica fume and iron fillings are added to the concrete mix. Cement in concrete mix is partially replaced by pulverized fuel ash, silica fume and iron fillings using two different percentages: 15% and 20%. The chemical composition of the implemented waste materials is shown in Table 1. Pulverized fuel ash is a by-product of pulverized fuel (normally coal) resulting from fired power stations. It consists of fine particles that are driven out of the boiler with the fuel gases. Silica Fume is an ultrafine powder collected as a by-product of the silicon and ferrosilicon alloy production. The main field of application is as pozzolanic material for high performance concrete. Iron Filings are mostly a by-product of the grinding, filing, or milling of finished iron products, so their history largely tracks the development of iron. For the most part, they have been a waste product.

Specimens details

Eight beam-to-column joint specimens made with recycled concrete are tested under quasi-static loading. All test specimens have the same details that are shown in Fig. 1 where all units are in mm. The adopted repair configuration using CFRP sheets is shown in Fig. 2 where joint panel is repaired along the diagonals only (angle= 60°) and parts of the beams and column are also wrapped using CFRP sheets. Specimens P-1 and P-2 are control specimens that made using ordinary concrete.

Cement, in concrete mixes, is partially replaced by 15% and 20% of silica fume, pulverised fuel ash and iron fillings in specimens S-15, S-20, A-15, A-20, I-15, and I-20, respectively. The mechanical properties of the recycled concrete mixes are illustrated in Tables 2 and 3 for compressive and tensile behaviors where the strength is taken as the average of two cylinder specimens for each concrete mix. The average longitudinal and transverse reinforcement yield strengths that used in the test specimens are 420 and 280 MPa, respectively. The specimens are instrumented to measure global response, applied loads, lateral displacement, and joint shear strains as shown in Figs. 3 and 4.

At the first stage, all joint specimens are partially loaded up to ±55 kN according to the cyclic loading pattern shown in Fig. 3(b). All specimens are then retrofitted using CFRP sheets according to the configuration shown in Fig. 2 and the repaired specimens are laterally loaded until failure.

Retrofit application

The surface of the partially damaged specimens is initially cleaned. Epoxy is then applied to the surface and CFRP sheets are then fixed to the concrete surface using epoxy. Another layer of epoxy is applied to the surface after fixing the CFRP sheets to the concrete surface as shown in Fig. 4. The joint panel is initially repaired using diagonal CFRP sheets followed by the rehabilitation of beams and column.

Test setup

The test setup consists of lateral loading system, and lateral restraint system as shown in Figs. 3 and 4. Beams are pin connected allowing horizontal translation at one end only while the column is pin connected from the bottom as shown in Fig. 4. Lateral loads are applied in two stages prior and after CFRP repair through a hydraulic actuator which is connected to the column top in order to apply lateral load by means of a loading collar. The unrepaired specimens are initially loaded to ±55 kN. The specimens are then repaired and loaded until failure. A quasi-static cyclic load is applied at the column top and measured using a load cell which is located between the hydraulic actuator and the loading collar. The actuator is pinned at the end to allow rotation during the test. At the first stage prior to CFRP repair, loads are applied as force-controlled steps beginning at ±5 kN load followed by load increments of 5 kN at each cycle up to a maximum load of 55 kN (80% of the control specimen capacity) at cycle 11 as shown in Fig. 3(b). The second stage after CFRP repair, lateral loads that illustrated in Fig. 6 are applied at loading rate of 12 kN/min. Figure 6(a) illustrates the lateral loading applied at the control specimen P-1 while Fig. 6(b) shows the lateral load that applied at the column top of all other specimens including the control specimen P-2.

Three high-accuracy displacement transducers are used where two of them are fixed along the joint face diagonals to measure joint shear strain while the third one is attached at the column top to measure the lateral displacement, as shown in Fig. 7. Test results as shown in Table 4.

Finite element modeling

An experimental program is conducted to explain the cyclic behavior of repaired RC connections made with recycled concrete. Furthermore, numerical modeling is performed to better understand their behavior at lower costs. Numerical finite element modeling is carried out using the nonlinear finite element software package, ABAQUS. Sensitivity mesh analysis is a mesh convergence analysis that has been done to find out element mesh and element size to obtain stable results that do not change with final mesh. This means the rate of change between two different mesh sizes are too small. The rate of change is measured in term of cracks pattern and lateral displacement where mesh size is selected at the instant the cracks pattern and the lateral displacement of the specimens have not changed by changing the mesh size. After performing a sensitivity analysis, the adopted finite element mesh of the joint model is shown in Fig. 8 where the mesh size is 4 cm. An implicit nonlinear dynamic analysis is performed in ABAQUS to simulate the response of the RC beam-column connection. Beam-column connections are modeled with full geometry in 3 dimensions. Concrete and reinforcement are modeled using an eight node-3D solid element C3D8R which is a general purpose linear brick element, with reduced integration (1 integration point) [²³]. Perfect bond between concrete and reinforcement is assumed by embedding steel bars in concrete with the same degree of freedom because all simulated specimens did not experience bond slip during the test however major shear cracks were extended at column neck.

Material constitutive behavior

The adopted material constitutive models of concrete, reinforcement and CFRP sheets are presented in this section.

Concrete model

Concrete Damage Plasticity model (CDP) is adopted in this research to simulate the constitutive behavior of concrete. CDP can be applied to simulate concrete behavior in RC elements subjected to monotonic, cyclic or dynamic loading. CDP is based on the models proposed by Lubliner et al. [²⁴] and Lee and Fenves [²⁵]. Scalar damage elasticity variables control the stress-strain relations. Stress-strain curves of CDP model which describes compressive and tensile response of concrete are shown in Figs. 9(a) and 9(b), respectively. According to CDP, the uniaxial compressive stress-strain response of concrete is characterized by three stages. The first stage assumes concrete behavior is linear until the initial yield point σ_co then concrete enters the plastic region in the second stage which characterized by stress hardening followed by the last stage which characterized by concrete softening beyond the ultimate stress σ_cu as shown in shown in Fig. 9(a).

Figure 9(b) describes concrete response under uniaxial tension where the behavior is considered linear-elastic until the maximum stress σ_co which corresponds to the onset of micro-cracking in concrete material. The second stage is characterized by softening in the stress-strain curve when cracks propagate in concrete.

The unloading response of concrete, when concrete specimen is unloaded from any point on the strain softening branch of the stress-strain curves, is observed to be weakened. Concrete elastic stiffness appears to be damaged (or degraded) and is characterized by two damage variables, d_t and d_c. These damage variables take values from zero to one. Zero represents the undamaged material where one represents total loss of strength [²³]. The model also allows for stiffness recovery during load reversals.

Different models are proposed in the literature to define damage in concrete. Rabczuk et al. [²⁶] have proposed model for treating crack growth by particle methods where the crack is considered as a collection of cracked particles but particles are split and crack segments are introduced. Furthermore, a dynamic cohesive law for concrete is proposed that considers the change of energy. The model has also a strain-rate dependent damage-plasticity model in the bulk. Rabczuk and Belytschko [²⁷] have also proposed an alternative method for large deformation applications where various cracking criteria have been used as rate dependent and rate independent constitutive and cohesive models. The method is established using Lagrangian-Eulerian kernel formulation so that the Lagrangian kernel allows that material fracture occurs due to physical condition while the switch to the Eulerian kernel assures that the method is stable for extremely large deformations.

Rabczuk and Belytschko [²⁸] have developed a new method for modeling discrete cracks in mesh free methods. The crack can randomly oriented and the representation of the crack’s topology is not required because the crack’s growth is represented discretely by activation of crack surfaces at individual particles. Rabczuk et al. [²⁹] have also proposed another method based on mesh-free method for modeling crack in RC structure based on a partition of unity concept and the method is formulated for geometrically nonlinear problems. Vu-Bac et al. [³⁰] and Hamdia et al. [³¹–³²] have implemented probabilistic sensitivity analysis for modeling damage and material failure. This method can be used for computationally expensive models. The method estimates the sensitivity indices for the model with correlated parameters.

In this study, the implemented damage parameters for concrete compressive and tensile behavior are shown in Table 5 where a parametric study is conducted and hence these parameters are selected to best fit the experimental results. Concrete tensile and compressive behavior is defined based on the experimental stress-strain curves defined in Tables 2 and 3. Sümer and Aktaş [³³] and other researchers [²⁴,³⁴] proposed parameters for defining the CDP model in ABAQUS. A parametric study is also performed in this research to find the plasticity parameters which best fit the experimental results. Based on the parametric study, the input plasticity parameters are taken as follows: dilation angle 30°, eccentricity 0.1, K 0.667, and f_b0/f_c0 1.16.

Reinforcement model

The stress-strain response of reinforcement bars and stirrups is assumed to be elastic perfectly plastic. This model assumes that reinforcement do not harden after yielding. The behavior is considered linear elastic until the yield stress σ_y then it is assumed that strains keep increasing without stress increment.

CFRP sheets model

An elastic-perfectly plastic isotropic model is used to model CFRP material property in this research. The elastic modulus of CFRP sheets is 240 GPa and the ultimate tensile strength is 3800 MPa while their Poisson’s ratio is 0.3. The behavior of CFRP sheets is simulated using shell elements (S4R), which is a 4-noded general purpose element. Tie elements are used to connect concrete and CFRP sheets.

Load simulation before and after repair

Different steps are defined to simulate the load before and after repair. At the first step, the un-repaired specimens are initially partially loaded up to ±55 kN according to the loading history shown in Fig. 3(b). CFRP sheets are then connected to concrete surface using tie elements after the termination of the initial lateral loads at the first loading stage (±55 kN). CFRP sheets are then activated at the end of the first step and the repaired specimens are laterally loaded according to the loading history shown in Fig. 6 until failure.

Model geometry and boundary conditions

A finite element model is constructed in ABAQUS to simulate the cyclic response of the repaired RC beam-column connection made with recycled concrete. The geometry is the same for all test specimens as shown in Fig. 1. The boundary conditions are constructed in ABAQUS to simulate the experimental lateral restraint system that shown in Fig. 5.

Model verification

The numerical finite element model is verified using the experimental results of the test specimens. The experimental concrete stress-strain curves of each recycled material are implemented to simulate concrete constitutive behavior. The compressive and tensile behaviors of the recycled materials are shown in Tables 2 and 3, respectively. The results are validated based on the load-displacement curves and cracks pattern. The finite element model can reasonably predict the experimental response of the test specimens as shown in Fig. 10.

Results and discussions

The experimental and numerical hysteresis load-displacement responses, cracks pattern and envelope curves of the repaired test specimens are investigated.

The hysteresis load-displacement response

Control specimens P-1 and P-2

Two control specimens are tested P-1 and P-2. Both specimens are typical but the applied lateral loads are not the same where the lateral loads applied at specimen P-1 and P-2 are shown in Figs. 6(a) and 6(b), respectively. The lateral load that imposed to specimen P-2 has two repetitive cycles at each peak load and hence the load carrying capacity of specimen P-2 (110 kN) that shown in Fig. 10(b) is less than that measured in specimen P-1 (140 kN) as expected. The ultimate lateral drift ratio of specimen P-1 and P-2 is 4.4% and 3.7%, respectively. The influence of CFRP sheets is significant on the load carrying capacity, ultimate deflection and ductility of the test specimens. The load carrying capacity of the repaired specimens is almost duplicated compared to the un-repaired specimens (55 kN). Furthermore, the failure mode of the repaired specimens has been changed from brittle joint shear to a ductile failure mode after repair where column hinge is depicted at the column neck at failure as shown in Figs. 11(a) and 11(c) for specimen P-1 and P-2, respectively. The experimental and numerical hysteresis responses are shown in Figs. 10 (a) and 11(b) for specimen P-1 and P-2, respectively. The experimental and numerical hysteresis responses are close which indicates that the finite element model can reasonably predict the cyclic behavior of the repaired RC connection.

Specimens A-15 and A-20

Cement is partially replaced by 15% and 20% of pulverised fuel in specimen A-15 and A-20, respectively. Pulverised fuel ash has increased the load carrying capacity of specimen A-15 about 18% compared to the control specimen P-2. The load carrying capacities of specimen A-15 and A-20 are 130 and 110 kN, respectively, where increasing the replacement percentage of cement by pulverised fuel ash from 15% to 20% has reduced the load carrying capacity of the specimen. The load carrying capacities of specimen A-20 and the control specimen P-2 is the same (110 kN). The ultimate drift ratio of specimen A-15 and A-20 at failure is 4.2% and 4%, respectively. Pulverised fuel ash has increased the lateral drift ratio about 14% and 8% in specimen A-15 and A-20, respectively, compared to the control specimen P-2. Increasing the replacement percentage of cement by pulverized fuel ash from 15% to 20% has reduced the joint carrying capacity and ultimate drift by 18% and 5%, respectively.

CFRP sheets with the adopted configuration have increased the load carrying capacity of the specimens significantly compared to the un-repaired specimens. The repaired specimens A-15 and A-20 have regained almost double the capacity of the un-repaired specimens. In addition, CFRP has changed the failure mode from brittle joint shear before repair to a ductile column hinge after repair as shown in Figs. 11(e) and 11(g) for specimen A-15 and A-20, respectively.

The experimental and numerical load-deflection curves of specimen A-15 and A-20 are shown in Figs. 10(c) and 10(d), respectively. The experimental and numerical cyclic responses are close which indicates that the finite element model can predict the cyclic response of the repaired RC connection with reasonable accuracy.

Specimens I-15 and I-20

Specimen I-15 and I-20 contain 15% and 20% iron fillings, respectively, as cement partial replacement. Lateral load carrying capacity of specimen I-15 and I-20 is 100 and 90 kN, respectively, and the ultimate drift ratio of specimen I-15 and I-20 is 2.5% and 3.2%, respectively. Replacing cement by 15% and 20% iron fillings has decreased the load carrying capacity of specimen I-15 and I-20 about 10% and 22%, respectively, compared to the control specimen P-2. Furthermore, iron fillings has decreased the ultimate drift ratio of specimen I-15 and I-20 by 48% and 16% compared to the control specimen P-2. Increasing the replacement percentage of iron fillings from 15% to 20% has reduced the joint carrying capacity by 10% but it has increased the ultimate drift ratio by 28%.

Adopting the proposed repair configuration using CFRP sheets has increased the load carrying capacity and drift ratio of specimen I-15 and I-20. Furthermore, it has changed the failure mode from brittle joint shear to ductile mode where column hinges are depicted in Figs. 11(i) and 11(k) for specimen I-15 and I-20, respectively.

The experimental and numerical load-deflection curves are very close. The experimental and numerical response of specimen I-15 and I-20 is shown in Figs. 10(e) and 11(f), respectively. It can be seen that the experimental and numerical cyclic responses are close which indicates that the finite element model is capable to predict the cyclic response of the repaired RC connection with reasonable accuracy.

Specimens S-15 and S-20

Specimen S-15 and S-20 contain 15% and 20% silica-fume, respectively, as cement partial replacement. Specimen S-15 has the largest load carrying capacity (130 kN) and drift ratio (4.9%) among all other test specimens. Silica fume concrete has increased the load carrying capacity of specimen S-15 by 18% compared to the control specimen P-2. The load carrying capacities of specimen S-20, A-20 and the control specimen P-2 is the same (110kN).

Silica fume has increased the lateral drift ratio significantly about 32% and 8% in specimen S-15 and S-20, respectively, compared to the control specimen P-2. Increasing the replacement percentage of cement by silica-fume from 15% to 20% has reduced the joint carrying capacity and ultimate drift by 18% and 23%, respectively.

CFRP sheets have significantly increased the load carrying capacity and drift ratio of the repaired specimens S-15 and S-20 (almost double the capacity of the un-repaired specimen). Moreover, CFRP sheets have changed the type of failure of specimen S-15 and S-20 from brittle before repair to ductile failure after repair where column hinge and beam hinge have formed in specimen S-15 at failure whereas column hinge is depicted in specimen S-20 at failure as shown in Figs. 11(m) and 11(o), respectively.

The finite element model can reasonably predict the experimental response of the repaired specimen S-15 and S-20. Figures 10(g) and 10(h) illustrate the experimental and numerical load-deflection curves of specimen S-15 and S-20, respectively. The experimental and numerical cyclic responses are close.

Crack pattern

The proposed finite element model can reasonably predict the crack pattern of the test specimens. Figure 11 compares between the experimental and numerical crack patterns of the test specimens. The repaired specimens remain coherent after failure. The un-repaired specimens have experienced major diagonal shear cracks in their joint panels and some other shear cracks in their upper columns’ necks [²¹]. After repair, column hinge is formed at the column neck of all test specimens at failure. Beam hinge and column hinge are formed in specimen S-15 only. Specimen S-15 has the largest drift ratio compared to all test specimens which indicate that specimen S-15 is relatively ductile specimen. The red then the green colors in the numerical response indicate to the points with severe stresses (major cracks) and the light blue illustrates minor cracks.

Envelope curves

The envelope curve is the backbone curve of the hysteresis response. Figure 12 compares between the numerical and experimental envelope curves of all test specimens. The experimental and numerical response is close which indicates that the finite element model is capable to predict the experimental behavior of the repaired specimens with acceptable accuracy.

Recycled concrete effect

Eight beam-to-column sub-assemblages made with recycled concrete are tested where cement in concrete mixes is partially replaced by pulverised fuel ash, silica fume and iron fillings using two different percentages 15% and 20%. Test results have shown that regardless the type of the implemented recycled concrete material; the load carrying capacity of joints made with 15% waste material (silica fume, pulverized fuel ash and iron fillings) is greater than that measured with 20% waste materials as shown in Table 2.

The joint carrying capacity has been reduced about 18% by increasing cement replacement percentage of pulverized fuel ash concrete and silica fume concrete from 15% to 20%. Furthermore, the ultimate drift ratio has been reduced by 5% and 23% by increasing the replacement percentage of pulverized fuel ash concrete and silica fume concrete from 15% to 20%, respectively. However, increasing cement replacement percentage of iron filling concrete from 15% to 20% has reduced the joint carrying capacity by 10% but it increased the ultimate drift by 28%.

Lateral load carrying capacity and ultimate drift ratio of beam-column joints made with 15% pulverised fuel ash and that made with 15% silica fume concrete are the greatest. Load carrying capacity and ultimate drift ratio of specimens made with iron filling concrete are relatively less than all other test specimens.

Pulverised fuel ash and silica fume have increased the load carrying capacity of specimen A-15 and S-15 about 18% compared to the control specimen P-2. Furthermore, pulverised fuel ash and silica fume have increased the ultimate drift ratio of specimen A-15 and S-15 about 14% and 32%, respectively, compared to the control specimen P-2.

Beam hinge and column hinge are depicted in the repaired specimen S-15 that made with 15% silica fume as cement partial replacement while column hinge is depicted in all other repaired specimens at failure.

CFRP sheets effect

CFRP sheets are epoxy bonded to the test specimens along the joint diagonals and parts of columns and beams. The adopted repair configuration, shown in Fig. 2, has enhanced the behavior of RC joints significantly where CFRP sheets have duplicated the load carrying capacity of the repaired specimens compared to the un-repaired specimens. Furthermore, they have increased the ultimate drift ratio and hence the ductility of the test specimens.

CFRP sheets have also changed the mode of failure from brittle joint shear failure before repair to a ductile failure mode after repair where column hinge is depicted at the column neck of the test specimens at failure. Repaired specimens remain coherent at failure.

Conclusions

An experimental and numerical program is carried out in this research to investigate the influence of CFRP sheets on the cyclic behavior of unconfined connections made with recycled concrete. Each specimen is partially loaded at the first stage and then specimens are repaired using CFRP sheets that are epoxy bonded along the joint diagonals and parts of columns and beams. The repaired specimens are then laterally loaded until failure. Different types of recycled concrete are used where cement is partially replaced by silica fume, iron filling and pulverised fuel ash using two different percentages 15% and 20%. Furthermore, a finite element model is built in ABAQUS to clearly study the behavior of the retrofitted specimens under cyclic loading. The proposed finite element model can reasonably predict the experimental cyclic behavior of the repaired connections made with recycled concrete. The following points can summarize the research outcomes.

1) The adopted repair configuration using CFRP sheets is recommended for retrofitting seismically vulnerable RC connections. The results have shown CFRP sheets have significantly increased the strength and ultimate drift ratio of test specimens.

2) The repaired specimens have regained almost double the capacity of the un-repaired specimens.

3) The retrofitted specimens have remained coherent at failure and CFRP sheets has changed the mode of failure from brittle joint shear before repair to a ductile failure mode with plastic hinges at column neck after repair.

4) The cyclic behavior of RC connections is significantly influenced by the nature of the applied lateral loads. The specimen that exposed to lateral loads with two repetitive cycles at each peak load has relatively less load carrying capacity compared to the specimen that exposed to lateral loads that have one cycle at each peak load.

5) Increasing cement replacement percentage by silica fume, fuel ash or iron filling from 15% to 20% has reduced joint carrying capacity and weakened the joint.

6) It is recommended using 15% pulverised fuel ash or silica fume as cement partial replacement to enhance the strength and ultimate drift of beam-column joints under cyclic loading. Iron filling concrete is also recommended but the enhancement is relatively less than that found with pulverised fuel ash concrete and silica fume concrete.

7) Using 15% of pulverised fuel ash or silica fume as cement partial replacement has increased the load carrying capacity of the specimen about 18% compared to the control specimen P-2. Furthermore, it has increased the ultimate drift ratio about 14% and 32%, respectively.

8) Increasing cement replacement percentage of pulverized fuel ash concrete and silica fume concrete from 15% and 20% has reduced the joint carrying capacity by 18% and it has reduced the ultimate drift ratio by 5% and 23%, respectively. However, increasing cement replacement percentage of iron fillings from 15% to 20% has reduced the joint carrying capacity by 10% but it has increased the ultimate drift by 28%.

9) Specimen made with 15% silica fume has the largest drift ratio where beam hinge and column hinge are depicted on the repaired specimen at failure while column hinge is depicted in all other repaired specimens.

10) The proposed finite element model can reasonably predict the cyclic behavior of RC connections made with recycled concrete at lower costs.

References

[1] Murad Y Z. Analytical and numerical assessment of seismically vulnerable corner connections under bidirectional loading in RC framed structures. Dissertation for the Doctoral Degree. London: Imperial College London, 2016

[2] Karayannis C G, Sirkelis G M. Strengthening and rehabilitation of RC beam-column joints using carbon-FRP jacketing and epoxy resin injection. Earthquake Engineering & Structural Dynamics, 2008, 37(5): 769–790

[3] Kalogeropoulos G I, Tsonos A D G, Konstandinidis D, Tsetines S. Pre-earthquake and post-earthquake retrofitting of poorly detailed exterior RC beam-to-column joints. Engineering Structures, 2016, 109: 1–15

[4] Faleschini F, Gonzalez-Libreros J, Zanini M A, Hofer L, Sneed L, Pellegrino C. Repair of severely-damaged RC exterior beam-column joints with FRP and FRCM composites. Composite Structures, 2019, 207: 352–363

[5] Le-Trung K, Lee K, Lee J, Lee D H, Woo S. Experimental study of RC beam–column joints strengthened using CFRP composites. Composites. Part B, Engineering, 2010, 41(1): 76–85

[6] Garcia R, Hajirasouliha I, Pilakoutas K. Seismic behaviour of deficient RC frames strengthened with CFRP composites. Engineering Structures, 2010, 32(10): 3075–3085

[7] Sasmal S, Ramanjaneyulu K, Novák B, Srinivas V, Saravana Kumar K, Korkowski C, Roehm C, Lakshmanan N, Iyer N R. Seismic retrofitting of nonductile beam-column sub-assemblage using FRP wrapping and steel plate jacketing. Construction & Building Materials, 2011, 25(1): 175–182

[8] Sharma R, Bansal P P. Behavior of RC exterior beam column joint retrofitted using UHP-HFRC. Construction & Building Materials, 2019, 195: 376–389

[9] Beydokhty E Z, Shariatmadar H. Behavior of damaged exterior RC beam-column joints strengthened by CFRP composites. Latin American Journal of Solids and Structures, 2016, 13(5): 880–896

[10] Obaidat Y T, Abu-Farsakh G A F R, Ashteyat A M. Retrofitting of partially damaged reinforced concrete beam-column joints using various plate-configurations of CFRP under cyclic loading. Construction & Building Materials, 2019, 198: 313–322

[11] Abu Tahnat Y B, Dwaikat M M S, Samaaneh M A. Effect of using CFRP wraps on the strength and ductility behaviors of exterior reinforced concrete joint. Composite Structures, 2018, 201: 721–739

[12] Rodopoulos C A, Pilakoutas K, Gdoutos E E. Failure Analysis of Industrial Composite Materials. McGraw-Hill Professional Engineering, 2000

[13] Murad Y. An experimental study on flexural strengthening of RC beams using CFRP sheets. International Journal of Engineering & Technology, 2018, 7(4): 2075–2080

[14] Murad Y. The influence of CFRP orientation angle on the shear strength of RC beams. The Open Construction & Building Technology Journal, 2018, 12: 269–281

[15] Bukhari I A, Vollum R L, Ahmad S, Sagaseta J. Shear strengthening of reinforced concrete beams with CFRP. Magazine of Concrete Research, 2010, 62(1): 65–77

[16] Norris T, Saadatmanesh H, Ehsani M R. Shear and flexural strengthening of R/C beams with carbon fiber sheets. Journal of Structural Engineering, 1997, 123(7): 903–911

[17] Noori K M G, Ibrahim H H. Mechanical properties of concrete using iron waste as a partial replacement of sand. Eurasian Journal of Science & Engineering, 2018, 3(3): 75–82

[18] Zhang Z, Hsu C T T. Shear strengthening of reinforced concrete beams using carbon-fiber-reinforced polymer laminates. Journal of Composites for Construction, 2005, 9(2): 158–169

[19] Khedr S A, Abou-Zeid M N. Characteristics of silica-fume concrete. Journal of Materials in Civil Engineering, 1994, 6(3): 357–375

[20] Dhir R K, Munday J, Ong L T. Investigations of the engineering properties of OPC/pulverised fuel ash concrete: Strength development and maturity. Proceedings of the Institution of Civil Engineers, 1984, 77: 239–254

[21] Murad Y, Abu-Haniyi Y, Alkaraki A, Hamadeh Z. An experimental study on cyclic behaviour of RC connections using waste materials as cement partial replacement. Canadian Journal of Civil Engineering, 2019, 46(6): 522–533

[22] Murad Y, AL-Bodour W, Abu-Hajar H. Cyclic behavior of RC beam-column joints made with sustainable concrete. International Review of Civil Engineering (IRECE), 2019, 10(6): 301

[23] Smith M. ABAQUS/Standard User’s Manual, Version 6.9. Providence, RI: Dassault Systèmes Simulia Corp, 2009

[24] Lubliner J, Oliver J, Oller S, Oñate E. A plastic-damage model for concrete. International Journal of Solids and Structures, 1989, 25(3): 299–326

[25] Lee J, Fenves G L. Plastic-damage model for cyclic loading of concrete structures. Journal of Engineering Mechanics, 1998, 124(8): 892–900

[26] Rabczuk T, Zi G, Bordas S, Nguyen-Xuan H. A simple and robust three-dimensional cracking-particle method without enrichment. Computer Methods in Applied Mechanics and Engineering, 2010, 199(37–40): 2437–2455

[27] Rabczuk T, Belytschko T. A three-dimensional large deformation meshfree method for arbitrary evolving cracks. Computer Methods in Applied Mechanics and Engineering, 2007, 196(29–30): 2777–2799

[28] Rabczuk T, Belytschko T. Cracking particles: A simplified meshfree method for arbitrary evolving cracks. International Journal for Numerical Methods in Engineering, 2004, 61(13): 2316–2343

[29] Rabczuk T, Zi G, Bordas S, Nguyen-Xuan H. A geometrically non-linear three-dimensional cohesive crack method for reinforced concrete structures. Engineering Fracture Mechanics, 2008, 75(16): 4740–4758

[30] Vu-Bac N, Lahmer T, Zhuang X, Nguyen-Thoi T, Rabczuk T. A software framework for probabilistic sensitivity analysis for computationally expensive models. Advances in Engineering Software, 2016, 100: 19–31

[31] Hamdia K M, Silani M, Zhuang X, He P, Rabczuk T. Stochastic analysis of the fracture toughness of polymeric nanoparticle composites using polynomial chaos expansions. International Journal of Fracture, 2017, 206(2): 215–227

[32] Hamdia K M, Msekh M A, Silani M, Thai T Q, Budarapu P R, Rabczuk T. Assessment of computational fracture models using Bayesian method. Engineering Fracture Mechanics, 2019, 205: 387–398

[33] Sümer Y, Aktaş M. Defining parameters for concrete damage plasticity model. Challenge Journal of Structural Mechanics, 2015, 1(3): 149–155

[34] Alfarah B, López-Almansa F, Oller S. New methodology for calculating damage variables evolution in Plastic Damage Model for RC structures. Engineering Structures, 2017, 132: 70–86

RIGHTS & PERMISSIONS

Higher Education Press