In recent years the use of fiber reinforced polymers (FRP) and shape memory alloys (SMA) in RC structures has gained popularity due to their superior material properties and resistance to corrosion. The application of these smart materials could easily replace the traditional steel reinforcement in newly built concrete structures, or can be used for the retrofit of existing structures. BCJs are one of the most critical assemblies in the structural system. Furthermore, investigations on BCJs are more spar as compared to beams or columns. In the following sections BCJs reinforced with SMA, FRP and hybrid or composite SMA-FRP as internal and FRP as external reinforcement are discussed.
2.1. Beam-Column Joints Internally Reinforced with FRP Bars
Investigations on BCJs internally reinforced with FRP bars are limited as oppose to more common method of external FRP reinforcement. Most of these studies have commonly used glass fiber reinforced polymer (GFRP) material for reinforcement bars (
Table 1). Said and Nehdi [
7] tested two full-scale BCJs, one reinforced with GFRP grid and one was reinforced with steel bars and stirrups under cyclic loading. Although the 3% minimum drift ratio requirement of ductile frame was achieved in the GFRP reinforced joint, lower stiffness and energy dissipation were observed due to elastic behavior of the GFRP material as compared to the steel reinforced joint. Furthermore, the beam tip load was as high as that of the steel reinforced joint. Because of the low stiffness of GFRP bars, the joint failure was due to the brittle behavior that led to the rupture of two bottom GFRP longitudinal bars.
On the other hand, Saravanan and Kumaran [
8] conducted experiments on eighteen BCJs reinforced with GFRP stirrups and bars, followed by finite element analysis study. Variables considered were bar types (threaded, sand-coated, and grooved), beam and column reinforcement ratios, concrete strength, and joint aspect ratio. Furthermore, the influence of GFRP stirrups on the joint shear strength was investigated and a design equation to predict the joint shear strength was proposed. Their study showed that compared to the BCJs reinforced with traditional steel bars, the GFRP sand-coated reinforcement improved the load carrying capacity by almost 5%, but more importantly the deformation capacity increased by 30% to 50%. Moreover, the presence of stirrups in the joint area was able to move the failure from the joint core to the beam-column interface.
Two other experiments in 2011 were conducted by Mady et al. [
9] and Hassaballa et al. [
10] on full-scale concrete BCJs reinforced with GFRP bars and stirrups when subjected to seismic loading to explore the influence of GFRP reinforcement on the behavior of the joints. Mady et al. [
9] had the longitudinal and transverse reinforcement material types (steel and GFRP), and beam longitudinal reinforcement ratios as parameters. While, Hassaballa et al.’s [
10] variables of study were longitudinal and transverse reinforcement materials (steel and GFRP) and the beam’s longitudinal bar details (with hooks, straight, or straight with extension into a beam stub). Both experiments had one reference specimen (longitudinal and transversal steel reinforcement), one specimen reinforced with GFRP bars and steel stirrups, and the rest were reinforced with GFRP bars and stirrups. Mady et al. [
9] revealed that the increase in the GFRP longitudinal bars reinforcement ratio would result in higher energy dissipation in the joint. Their findings of using steel instead of GFRP stirrups would result in an increase of dissipated energy. In Hassaballa et al.’s study [
10], even though both GFRP reinforced and control joints failed in shear, the failure mode of GFRP reinforced joint with extended stub was observed by the formation of plastic hinge away from the column face which satisfies the design capacity concept (weak beam-strong column). Furthermore, the BCJs with GFRP bars as internal reinforcements of both experiments [
9,
10], sustained 4% storey drift ratio safely with no considerable damage or residual strains. Thus, the BCJ could retain its original shape upon removal of the seismic loads up to this drift ratio. Moreover, all joints exceeded their individual design capacity by an average of 9%. Based on Mady et al. [
9] and Hassaballa et al.’s [
10] investigations, one can conclude that the BCJs reinforced with longitudinal and transverse GFRP reinforcement, usually provide lower energy dissipation as compared to steel reinforced BCJ.
Table 1 shows additional details of the experiments and results.
Similarly, Hasaballa and El-Salakawy [
11] tested six full-scale GFRP-reinforced exterior BCJ prototypes under seismic loading. The parameters of their study were concrete strength, and shear stress level in the joint. Diagonal shear cracks were observed in some of the joint specimens. For the same shear stress level, similar strains were recorded in the joints stirrups at failure even though the failure occurred at different drift ratios. Furthermore, the energy dissipation and ductility were higher for the joints that had lower concrete strength as compared to those with higher concrete strength. Therefore, the joints with higher energy dissipation were able to regain their original shape after unloading. Further details and results are summarized in
Table 1.
Similar to the latter reference, Ghomi and El-Salakawy [
12] performed experiments on six full-scale BCJs reinforced internally with GFRP bars and stirrups under seismic loading. The joints had lateral beams on all four sides of the column. The variables considered were reinforcement materials (steel and GFRP), presence of lateral beams, joint shear stress level, and end anchorage of beam longitudinal bars (headed-end, and bent bars). Their findings revealed that in some cases, GFRP-RC BCJs confined with lateral beams provided nonlinear behavior and non-brittle failure, despite the expected linear behavior from FRP-RC structures. This was also due to high shear stress level in some joints, where at the same drift ratios they could dissipate more energy as compared to those joints with lower shear stress level. Furthermore, both methods of anchorages performed well with shear stress level of 1.1 √
fc′ or higher.
2.5. Beam-Column Joints Externally Reinforced with FRP Sheets or Straps
In the recent years, the FRP materials have been widely used to externally reinforce and retrofit the RC structures. The aim of using these materials is either to increase the structural load capacity or to repair the deteriorated structures along with the application of other materials such as mortars. The increase in the load capacity maybe necessary in order to address new code requirements, errors in the design or construction process, or to sustain an extra live load. In external reinforcement, the propagation of microcracks is a factor that plays an important role in affecting the bond strength. To that end, multiscale analytical models have been developed to predict either the delayed debonding of FRP external strengthening or the bond lifetime e.g., [
21,
22]. However, there are new methods to improve the bond of external reinforcement such as vacuum applications, stud shear connectors, and anchor spikes [
23,
24]. In this section, we shed the light particularly on some studies that have used FRP sheets or straps to strengthen the most critical structural element namely, the BCJ [
1,
5,
6,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34,
35,
36,
37,
38,
39,
40].
Few experimental investigations on BCJs have used steel anchorages for the FRP sheets [
25,
26,
37,
40]. Two of these studies were performed on the same size BCJs under the same loading conditions using different number of FRP layers and different steel anchorage configurations [
25,
26]. Ghobarah and Said [
25] did an experiment on four non-seismically designed RC BCJs (with no shear reinforcement in the joint core). Two of the specimens were considered as control joints and were subjected to quasi static cyclic loading. After this phase of the experiment, the two specimens were repaired using unidirectional and bidirectional U-shaped and X-shaped GFRP sheets on the columns at the joint core location only, and the other two undamaged joints were strengthened with the same retrofit configurations as the damaged ones. Then, the four specimens were subjected again to quasi static cyclic loading to find the effectiveness of rehabilitation. Steel plates and threaded steel rods were used as anchorages for the GFRP sheets. More details of GFRP sheets and material properties are provided in
Table 2. The comparison between the control and strengthened specimens showed that, the GFRP helped improve the shear capacity of the joints. Additionally, the ductility and the energy dissipation were increased by approximately 62% and 72%, respectively. The GFRP strengthening was successful in delaying the shear failure, and in some cases, the failure mode was transferred from shear failure to flexural hinge in the beam. Furthermore, El-Amoury and Ghobarah [
26] conducted another experiment on the same size BCJs under the same loading conditions but with different number of GFRP layers and steel anchorage configurations. Besides GFRP column confinement, L-shaped GFRP sheets were added at the bottom face of the beam and column joints followed-up by two different anchorage systems. The first system was somewhat similar to that of Ghobarah and Said [
25] study using steel plates with threaded steel rods on the column at the joint core, except a steel angle was added at the bottom of BCJ core. In the second system, two U-shaped steel plates were applied to anchor the extended GFRP sheets on beams in order to avoid the bond slip of steel bars and the debonding of GFRP sheets. Following up with the latter experiments [
25,
26], Ghobarah and Al-Emoury [
37] had two groups of BCJ specimens. In the first group, same anchorage systems [
25,
26] were employed but the GFRP sheets were replaced with CFRP. In the second group, different configurations of anchorage systems were incorporated involving the use of additional steel rods and plates. In [
25,
26,
37], the brittle shear failure of the joints was eliminated, and load-carrying capacity, ductility and energy dissipation were improved. Specifically, the new anchorage system and GFRP configuration [
26] delayed the debonding of GFRP sheets, as well as the slippage of beam’s top reinforcement. Moreover, the use of rods [
37] led to an improvement in the anchorage system conditions where the tensile strength could be fully achieved (
Table 2).
Similarly, a more recent study was conducted on seismically deficient RC BCJs using steel plates with different shapes (U-shaped, L-shaped steel angles, and horizontal plates) in addition to threaded steel rods, to anchor the CFRP sheets (uniaxial, quadriaxial) in strengthening either damaged or undamaged joints [
40]. The joints were tested under cyclic loading to investigate the effectiveness of CFRP configurations (X-shaped, U-shaped, and horizontal), the anchorages methods using steel elements, and different internal steel reinforcement ratios. The results obtained from the experiment indicated that FRP-retrofitted members regained their strength and achieved a higher ductility than its control counterparts. It was also noted that the use of L-shaped steel angles on all corners of the joint provided confinement resulting in high level of displacement capacity up to 75% (
Table 2).
The same idea was applied by Al-Salloum and Almusallam [
28] in their experiment on RC BCJs designed under gravity load with pre-1970s deficient reinforcement details. Four half-scale joints, two controls and two strengthened with CFRP, were tested to study the efficiency of the CFRP sheets in upgrading the shear strength and ductility when subjected to seismic loads. Two different schemes were employed to strengthen the joints. In the first scheme, CFRP sheets were epoxy bonded to the BCJ regions. In the second scheme, sheets were epoxy bonded to the joint core only and mechanical anchorages were also provided to prevent any debonding. Furthermore, the damaged control specimens were repaired by filling the cracks with epoxy and were wrapped with CFRP sheets and tested again. Hence, a total of six specimens (two controls; two strengthened; and two repaired) were considered. It was observed that although in the first scheme, the CFRP reinforcement was extended from the joint to the beam and the column, the debonding happened due to the lack of anchorages. However, the provided anchorages in the second scheme were able to move the failure from the joint area to the beam. Furthermore, both schemes were successful in enhancing the strength and ductility as well as providing stiffness against shear distortion for the joints. Further results are provided in
Table 2.
Most experimental studies have been performed on scaled-down with few on full-scaled BCJs (e.g., [
1,
6,
29,
32,
38,
39]). Six full-scale non-seismically designed beam-column joints were strengthened with CFRP sheets and subjected to cyclic loading [
1]. The parameters of this study were the number of CFRP layers, CFRP sheets configurations, and the effects of lap splice and axial load on the column. Various CFRP configurations were applied to the joints. Based on the experimental results, it was noted that the increase in column’s axial load improved the stiffness and lateral load capacity in both control and CFRP-strengthened joints. However, the CFRP rehabilitation enhanced the overall structural performance of the joints. Increase in number of CFRP layers resulted in higher stiffness, lateral load, and energy dissipation capacities (
Table 3). Various anchorage systems stopped the debonding of CFRP sheets and the slippage of shortly embedded bottom bars in the beam was avoided.
Another study on full-scale joints evaluated the capacity of RC damaged joints representing post-earthquake situation [
6]. The deficient joints were repaired with high strength concrete and then wrapped with CFRP sheets. It was found that the replacement of the concrete in the joint core with high strength concrete enhanced the shear strength of the joints up to 44% as compared to the control specimen. In addition, the CFRP strengthening was adequate to develop the full plastic capacity of the joint and increased the strength up to 69% (
Table 3).
Additionally, two full-scale RC BCJs subjected to cyclic loading on the beam tip were tested by Vatani-Oskouei [
29]. The two specimens were then repaired and strengthened with CFRP sheets along with steel angles added to the top and bottom faces of the beam. The results indicated that adding the CFRP sheets to the damaged joints helped to increase the load-carrying capacity, the amount of energy dissipated by the joints and the ductility. Furthermore, the failure location moved from the beam-column interface to the beam. However, the failure was due to the rupture and debonding of CFRP sheets. It is worth noting that injecting epoxy into the cracks did not enhance the overall behavior of the specimens. Details and results are noted in
Table 3.
Another full-scale test was performed on a non-seismic BCJ subjected to reverse cyclic loading which was repaired and strengthened by CFRP sheets around the joint core and tested again [
32]. The specimen was built using low compressive strength of concrete (8.5 MPa) and plain reinforcement bars. Upon strengthening, the failure mode was changed from the joint shear failure to the rupture of CFRP sheets. It was noted that increasing the number of CFRP diagonal sheets around the joint, might delay the CFRP rupture which in turns it would result in higher drift ratios and horizontal load capacities (see
Table 3 for more details).
In another study [
38], six full-scale corner BCJs including slab were tested under cyclic loading to study the influence of several parameters on their behavior. Variables considered were the presence of CFRP joint strengthening below the slab level, fibers orientation, CFRP thickness, column confinement, and concrete compressive strength. The results obtained indicated that, the use of a suitable amount of CFRP reinforcement helped to increase the strength of the joint while avoiding the full debonding of CFRP ends. Higher concrete strength resulted in higher peak joint strength. Moreover, various combined modes of failure were observed depending on the strengthening configurations, i.e., CFRP debonding accompanied with joint shear failure, CFRP debonding accompanied with column flexural hinging, and CFRP rupture with column flexural hinging (
Table 3).
A new idea was employed by Di Ludovico et al. to test a full-scale RC structure in order to understand the global behavior of RC frame when BCJs were retrofitted by GFRP laminates [
39]. The variables in this study were, the level of PGA (0.2, 0.3), the GFRP configuration (uniaxial, quadriaxial), and the column cross-section dimensions and confinement. It was noted that, although the confinement of the columns by GFRP laminates did not significantly increase the strength; however, the rotational plastic hinge capacity extremely improved. It was concluded that the use of GFRP for retrofitting RC joints would considerably improve the seismic performance of the frame structure. Further details are provided in
Table 3.
In addition to large-scale testing, due to budgetary constraints, and laboratory space and equipment limitations, many researchers performed experiments on scaled-down BCJs, (e.g., [
31,
33,
36]. In one experiment test was carried on one seismic (S) and four non-seismic (NS) 2/3-scale RC BCJs [
31]. The NS joints were subjected to reversed cyclic loading then they were repaired and strengthened with various CFRP laminates shapes (U-shaped, L-shaped, sheets, and wraps) and tested again. The results of the four NS strengthened damaged joints were compared to the seismically designed joint. The repaired and strengthened specimens could achieve load-carrying capacity equal to or higher than that of the seismic one. However, the initial stiffness of the strengthened specimens was lower than the seismically designed joint due to the existing cracks. Based on the damage index assessment achievement, it was noted that the use of CFRP laminates is effective up to the repair-ability performance level in order to improve the joints seismic behavior. More Details about the specimens and results are provided in
Table 4. In another experiment, Le-Trung et al. [
33] tested RC BCJs strengthened with FRP sheets under cyclic loading. The experiment, performed on 1/3-scale, exterior joints consisted of one non-seismically (NS) and one seismically (SD) designed specimens and six non-seismically designed specimens (RNS) retrofitted with CFRP sheets. The parameters of the study were CFRP sheet thickness and configurations (T, L, X-shaped and strip combinations). It was observed that the NS BCJ failed in brittle manner in shear with significant damage, while the SD joint showed more of a flexural failure in the beam indicating a ductile behavior. Furthermore, the use of CFRP sheets for the non-seismic joints resulted in the lateral strength and ductility improvement. The X-shaped, and the combination of T-shaped/L-shaped/column strips configurations outperformed other strengthened joints in terms of strength and ductility. The failure mode changed from shear in the joint core to flexural failure in the beam (
Table 4).
Another experimental study was conducted on eighteen 2/3-scale RC BCJs to examine the effectiveness of large number of parameters including presence of mechanical anchorages, area fraction of FRP, distribution of FRP between the beam and the column, column axial load, internal joint steel reinforcement, initial damage, carbon versus glass fibers, sheets versus strips, number of FRP sheets or strips, and the effect of transverse beams [
36]. The results obtained from the experiment indicated an increase in the effectiveness of both strips and sheets when using mechanical anchorages, an increase of 30% was achieved in terms of strength, while the energy increased by 40%. The most effective axial load on the column was 2.5 higher than the initial loading. The strength and energy dissipation increased with higher number of FRP layers. Additionally, it was observed that increasing the area fraction of FRP in both the columns and beams led to an increase in the strength and the energy dissipation in a comparable amount to those where only the area fraction of the beam was increased. It was also noticed that, the use of GFRP was slightly better than CFRP in increasing the strength, while it achieved higher increase in terms of energy dissipation when compared to CFRP. The presence of transverse beam diminished the dependency on FRP sheets in terms of strength and energy dissipation, more results are illustrated in
Table 4.
In the experimental studies discussed so far, only Antonopoulos and Triantafillou [
36] had the FRP material type as a variable. However, other experiments were conducted on exterior BCJs strengthened with CFRP or GFRP materials [
27], and interior BCJs strengthened with the combination of CFRP and GFRP sheets [
5]. Mukherjee and Joshi [
27] investigated two different types of joints, non-seismic (non-ductile), and seismic (ductile). Two FRP-strengthening schemes were proposed. Both schemes have seismic and non-seismic joints. For the first scheme, the CFRP or GFRP sheets were used on the transverse beams and the column around the joint core to strengthen the joints. In the second scheme, CFRP plates were installed on the transverse beams and wrapped with CFRP sheets to improve the bending stiffness. Additional joint details and material properties are shown in
Table 5. The joints were subjected to cyclic loading and then damaged specimens were repaired by replacing the loose concrete with epoxy and strengthened with CFRP or GFRP sheets. The results showed that GFRP and CFRP strengthening enhanced the lateral load capacity, ductility, and energy dissipation. The increase in ultimate deflection was an average of 33% as compared to the control counterpart specimen; an indicative of higher ductility. Also, the increase in energy dissipation in the non-seismic strengthened joints, compared to control counterpart, was 216%, 104.3%, 111.5%, and 60.6% for the BCJs reinforced with double GFRP sheets, single CFRP sheets, double CFRP sheets, and CFRP plates, respectively. For the seismic joints, the increase in energy dissipation was 76.4%, 97.5%, 151.48%, and 119.8% for the BCJs reinforced with double GFRP sheets, single GFRP sheets, double CFRP sheets, and CFRP plates, respectively compared to control counterpart. Furthermore, using two layers of CFRP or GFRP led to higher load capacity for both seismic and non-seismic joints. The pullout of the CFRP and steel reinforcement was the mode of failure for the joints strengthened with CFRP plates; however, the yield load was enhanced considerably. Using the CFRP as oppose to GFRP sheets helped the joints to exhibit higher stiffness (
Table 5). For the interior joints experiment, four RC non-seismically interior beam-wide column joints subjected to cyclic loading were strengthened with two schemes [
5]. The joints were repaired by injecting epoxy in the cracks and then strengthened with GFRP sheets (scheme 1) and CFRP and GFRP sheets (scheme 2) and tested again. Fiber anchors were also used to prevent the debonding. The four specimens were divided into two series based on the dimensions of the joints (details in
Table 5). The results of the damaged specimens were compared with those of control counterparts. It was observed that increasing the axial load applied on the column helped to close the cracks which occurred on the side face of the column in some of the specimens. Although the anchors were provided to prevent delamination of FRP sheets, the debonding occurred in all specimens. Moreover, scheme 1 could recover the original stiffness and the strength capacity and the energy dissipation was increased by an average of 74% as compared to the control counterpart. Similar improvement was observed in scheme 2. The energy dissipation capacity was enhanced by 61% to 92% depending on the specimen size. The latter result indicated that the use of CFRP sheets was better in confining the joint than GFRP although the failure mode for both schemes was debonding of FRP sheets (
Table 5).
On the other hand, Analytical models have been developed by researchers on CFRP strengthening BCJs based on experimental results of their own or others [
30,
34,
35].
Based on experimental studies [
29,
30,
31,
32,
33,
34,
35,
36], Hadi and Tran [
30] performed experimental and analytical studies on four RC non-seismic BCJs subjected to reversed cyclic loading. One joint was kept as a control specimen, while the other three were strengthened externally with CFRP wraps. The shape of the columns was modified from square to circular sections by using concrete cover followed by CFRP wrapping. Experimental results showed that the columns shape modification helped the confinement effectiveness of the CFRP wraps (
Table 3). As a result, the shear capacity and the overall performance of the joints were improved and debonding and/or bulging of FRP wrap from the concrete surface were eliminated. Due to the increase in CFRP ratios, the failure changed from joint core shear to flexure in the beam, and the plastic hinge moved further away from the joint to the beam (
Table 3). To predict the shear capacity an analytical model was developed and validated with the present CFRP-strengthened joints results [
33] and 32 other joints in the existing literature. The analytical model proved to be suitable for practical applications. Additionally, the shear capacity of RC members has also been predicted in the case of RC member subjected to concentrated load [
41].
Another analytical model was developed by Le-Trung et al. [
34] for the joints with the same CFRP configurations of Le-Trung et al. [
33], using the DRAIN-2DX program. The analytical model could accurately predict the CFRP strengthened joint behavior by taking into account joint shear behavior, bond slip of longitudinal beam reinforcement, and the effects of anchorage at the ends of the attached CFRP sheets.
Although few studies have examined analytical models of FRP-strengthened RC BCJs [
30,
33,
34], there is a lack of simple and generalized formulations to predict the shear capacity. Del-Vecchio et al. [
35] collected experimental data from large-scale experiments performed on RC joints [
1,
27,
28,
30,
36,
37,
38,
39,
40]. Their study examines a new and simplified model to predict the shear capacity and study the effectiveness of the FRP in strengthening deficient corner BCJs subjected to severe seismic loading. The parameters included in the model were defined based on the effect of externally bonded FRP systems used in large number of data set obtained from existing experimental work in the literature. The proposed formulation had good agreement with the experimental results in terms of average effective strain, and could predict the shear strength of corner beam-column joint strengthened with FRP while taking into consideration the effect of all parameters of experimental studies used in their analytical model. Furthermore, the model was able to avoid the brittle shear failure of the joint by knowing the amount of FRP reinforcement needed.