Bond Behavior of Cleaned Corroded Lap Spliced Beams Repaired with Carbon Fiber Reinforced Polymer Sheets and Partial Depth Repairs

The present research investigated the bond behavior of a cleaned corroded reinforcing bar repaired with a partial depth concrete repair and a partial depth concrete repair followed by the application of fiber-reinforced polymer (FRP) sheets. Twelve lap splice beams were cast and tested under static loading. The test variables considered were a partial depth repair with prepackaged self-consolidating concrete (SCC) for six lap splice beams and additional confinement with carbon fiber reinforced polymer (CFRP) sheets for another six beams. The test results for the repaired lap splice beams were compared with those for a monolithic lap splice beam. This research found that the average bond strength increased as the bar mass loss increased for all bonded lengths. The lap splice beams repaired with partial depth were able to repair concrete with similar properties to those of the monolithic concrete. However, they had higher concrete strength than the monolithic beams which showed a higher average bond strength than the monolithic lap splice beams. The beams confined with FRP sheets showed a rise in the bond strength and the equivalent slip by 34–49%, and 56–260% as compared to the unconfined beams, respectively.


Introduction
The lap splice is an effective and simple method to connect reinforcing bars in order to transfer forces between them. However, lap spliced reinforced concrete (RC) members are commonly subject to bond problems. This is because the lap splice region of the bars is in a bond critical region, which may cause an RC member to exhibit a brittle bond failure. Bond failure can occur by splitting of the concrete cover between a spliced bar and the side face of the lap splice beam. The bond between the reinforcing bar and the surrounding concrete might fail through shearing off of the concrete keys between reinforcing bar ribs, resulting in bar slip which is called pull out bond failure [1][2][3]. Splitting bond failure is the most common bond failure, especially in the absence of confinement. Many parameters influence the bond strength between the reinforcing bar and the surrounding concrete, including bar diameter, anchorage/splice length, concrete strength, concrete cover and the presence of confinement such as transverse reinforcement of fiber-reinforced polymer (FRP) sheets [4][5][6]. Azizinamini et al. [7] consuming, raising the cost of the product. Consequently, approaches to decrease the curing time for composite systems are of great interest. In this regard, Antunes et al. [26] investigated the effect of seawater exposure at 80 • C for up to 28 days on filament-wound glass fiber-reinforced polymer (GFRP) cylinders partially cured by passing saturated steam through them just after winding seeking a faster curing route. The results of their study revealed that using GFRP as filament wound, significantly enhanced the mechanical and durability performance of composite components exposed to seawater.
Malumbela and Alexander [27] concluded that repairing a corroded RC member with an FRP sheet must be done after partial depth repair. FRP wrapping without the partial depth repair did not prevent further corrosion. However, applying only the partial depth repair protected the reinforcing rebar from further corrosion and restored the serviceability state of the beams. The partial depth repair method is widely used for rehabilitating corroded RC members. Before applying the partial depth repair, certain steps must be completed which include removing the damaged concrete, removing the rust and cleaning the reinforcing bar by sandblasting or water pressure and roughening the surface of the original concrete [28,29].
The current paper studied the bond behavior of lap splice beams with a corroded reinforcing bar that was cleaned and repaired with partial depth repair and/or FRP sheets. The partial depth repair used in this study was a commercial prepackaged self-compacting concrete (SCC) extended with 50% of coarse aggregate by mass with a 13 mm maximum aggregate size. This proportion of coarse aggregate was chosen to simulate the proportion of coarse aggregate to the monolithic beams that were cast with ready mix normal concrete (M). In this study, the FRP sheets were used to confine the lap splice zone, which was 300 mm in length. Two corrosion levels (7.5% and 15% mass loss) were studied and compared to non-corroded bars to assess the effect of corrosion severity on the bond behavior.
Many reinforced concrete structures with lap splice beams are vulnerable to corrosion, which is a serious issue that may deteriorate the bond amongst the surrounding concrete and reinforcing bars. However, a few researchers have studied the effect of rehabilitating a corroded RC member with FRP sheets on bond or flexural capacity. Hence, to the author's best knowledge, there is a lack of studies on the bond behavior of cleaned corroded rebar and repaired partial depth repair. Furthermore, repairing the corroded RC members to enhance and restore the bond efficiency with a combination of FRP sheets and partial depth repair has not been studied. To cover this gap, this research studies the effect of repairing the cleaned corroded rebar with partial depth repair and/or FRP sheets on the bond behavior of the lap splice beam.

Test Specimen
In this study, twelve lap splice beams were made and tested statically under four-point loading with 1800 mm clear span length, 600 mm constant moment region and 600 mm shear spans each. All the lap splice beams had the same dimensions, with a rectangular cross-section of 250 mm × 350 mm with a total length of 2200 mm. The lap splice length was designed to be 300 mm to lead to bar slippage to ensure that all beams would fail in a bond splitting mode before the reinforcing bar reached the yield strength (flexural failure). In addition, all of the beams were reinforced with two 20 M flexural reinforcing bars spliced at the mid-span region. All beams were also reinforced with two continuous 10 M steel bars at the top (compression zone) of the beam. To avoid a brittle shear failure, each beam was provided with 10 M transverse reinforcement in the shear spans spaced at 100 mm center-to-center ( Figure 1a).
All the beams were fabricated with two notches with dimensions of 100 mm × 50 mm at the bottom of the beam that exposed the flexural reinforcing bars. The two notches were placed at the two ends of the lap splice region to evaluate the bond stress-slip behavior. Five linear variable differential transformers (LVDTs) were used, two to measure the loaded end slip, two to measure the free end slip and one to measure the deflection of the beam, as revealed in Figure 1b. A total of 10 electrical resistance strain gauges were used on the spliced bars for each beam to measure the strain distribution along the spliced length. Two strain gauges were fixed on each bar at the lap splice zone at distances of 50 mm and 225 mm from the beginning of the splice zone. Moreover, one strain gauge was placed on each side of the splice zone on the exposed reinforcing bar at the notches to measure the bar stress in this region where the tensile force was taken only by the Six lap splice beams were unconfined and six were wrapped with carbon fiber reinforced polymer (CFRP) sheets. For the wrapped condition, the CFRP sheet was U-wrapped in a single layer of 950 mm length and 300 mm width to confine the lap splice zone. For each confinement case, three beams were monolithic at different mass loss levels (control, 7.5% and 15% mass loss) and three beams were repaired with a partial depth repair concrete at different mass loss levels (control, 7.5% and 15% mass loss).
For the beams that were repaired with a partial depth repair, a pocket was designed for the repair purposes, with dimensions of 300 mm in length, 250 mm in width and 70 mm in depth. The concrete cover for all beams was 25 mm. A 25 mm clearance was kept under the spliced bars to meet the minimum clearance reinforcement of 19 mm or 6 mm larger than maximum coarse aggregate size in the partial depth repair material in accordance with the specification of ACI 364-2014.
All the beams were fabricated with two notches with dimensions of 100 mm × 50 mm at the bottom of the beam that exposed the flexural reinforcing bars. The two notches were placed at the two ends of the lap splice region to evaluate the bond stress-slip behavior. Five linear variable differential transformers (LVDTs) were used, two to measure the loaded end slip, two to measure the free end slip and one to measure the deflection of the beam, as revealed in Figure 1b. A total of 10 electrical resistance strain gauges were used on the spliced bars for each beam to measure the strain distribution along the spliced length. Two strain gauges were fixed on each bar at the lap splice zone at distances of 50 mm and 225 mm from the beginning of the splice zone. Moreover, one strain gauge was placed on each side of the splice zone on the exposed reinforcing bar at the notches to measure the bar stress in this region where the tensile force was taken only by the bar in the beams. The increase in stress at other strain gauge locations toward the stress measured at this location as a test was used to describe the progress of the bar debonding.

Test Procedure
All of the lap splice beams were supported with a 1800 mm span length and tested statically under four-point loading until failure. The load was applied to the beam through a steel spreader beam connected to the actuator. The steel beam then transferred the load to two locations on the top of the beam to produce a constant moment in the central region of the lap splice beams. The test frame was servo-controlled with a capacity of 500 kN, and was run in displacement control at the rate of 0.3 mm/min. The displacement at the center of the lap splice beam was recorded as the load was applied until the failure occurred. Figure 2 shows the lap splice beam installed in the test frame. bar in the beams. The increase in stress at other strain gauge locations toward the stress measured at this location as a test was used to describe the progress of the bar debonding.

Test Procedure
All of the lap splice beams were supported with a 1800 mm span length and tested statically under four-point loading until failure. The load was applied to the beam through a steel spreader beam connected to the actuator. The steel beam then transferred the load to two locations on the top of the beam to produce a constant moment in the central region of the lap splice beams. The test frame was servo-controlled with a capacity of 500 kN, and was run in displacement control at the rate of 0.3 mm/min. The displacement at the center of the lap splice beam was recorded as the load was applied until the failure occurred. Figure 2 shows the lap splice beam installed in the test frame.

Specimen Fabrication
The lap spliced bars that needed to be corroded were cast first into small concrete prisms with dimensions of 100 mm in depth, 100 mm in width and 300 mm in length, which was the lap splice length. To accelerate the corrosion rate, the concrete mix used to cast those prisms contained 3.8% salt of the cement mass, which was equivalent to 2.3% chloride. The prisms were kept in a special chamber subject to a continuous combination of oxygen and moisture, which is essential to facilitate the corrosion operation. Power supplies were used to accelerate the corrosion through an impressed current. The prisms were divided into two groups: 7.5% and 15% mass loss, and each group that was to be corroded to the same level was connected in series to ensure a constant current. Every prism was cast with one reinforcing bar (anode) and one stainless steel bar (cathode). To reach the desired corrosion level (7.5% or 15% mass loss), the period of accelerated corrosion was estimated using Faraday's law [20]: (1) where mi is the mass loss (g); M is the atomic weight of the metal = 56 g for Fe; I is the corrosion current (A) = i × sa; i is the corrosion current density (μA/cm 2 ); sa is the surface area of the corroded steel (cm 2 ); T is the time (s); z is the valence of the corroding metal (2 for iron); F is Faraday's constant = 96,500 A·s.
From Faraday's law, the estimated time to achieve 7.5% and 15% mass loss was 75 and 150 days, respectively. Once the desired level of corrosion was achieved, the prisms were cut, and the reinforcing bars were removed and cleaned by sandblasting. Subsequently, the bars were weighed, and the mass loss of each bar was calculated as follows: Consequently, the lap splice beams were fabricated and divided into two groups: monolithic lap

Specimen Fabrication
The lap spliced bars that needed to be corroded were cast first into small concrete prisms with dimensions of 100 mm in depth, 100 mm in width and 300 mm in length, which was the lap splice length. To accelerate the corrosion rate, the concrete mix used to cast those prisms contained 3.8% salt of the cement mass, which was equivalent to 2.3% chloride. The prisms were kept in a special chamber subject to a continuous combination of oxygen and moisture, which is essential to facilitate the corrosion operation. Power supplies were used to accelerate the corrosion through an impressed current. The prisms were divided into two groups: 7.5% and 15% mass loss, and each group that was to be corroded to the same level was connected in series to ensure a constant current. Every prism was cast with one reinforcing bar (anode) and one stainless steel bar (cathode). To reach the desired corrosion level (7.5% or 15% mass loss), the period of accelerated corrosion was estimated using Faraday's law [20]: where m i is the mass loss (g); M is the atomic weight of the metal = 56 g for Fe; I is the corrosion current (A) = i × s a ; i is the corrosion current density (µA/cm 2 ); s a is the surface area of the corroded steel (cm 2 ); T is the time (s); z is the valence of the corroding metal (2 for iron); F is Faraday's constant = 96,500 A·s. From Faraday's law, the estimated time to achieve 7.5% and 15% mass loss was 75 and 150 days, respectively. Once the desired level of corrosion was achieved, the prisms were cut, and the reinforcing bars were removed and cleaned by sandblasting. Subsequently, the bars were weighed, and the mass loss of each bar was calculated as follows: Consequently, the lap splice beams were fabricated and divided into two groups: monolithic lap splice beams and repaired lap splice beams. For the monolithic lap splice beams, the reinforcing bars were placed and cast with ready mix concrete. However, for the lap splice beams to be repaired with partial depth repair, the splice zone was isolated with high-density foam to form the pocket of the partial depth repair region. After curing, the high-density foam was removed, and the partial depth repair concrete was installed.

Material Properties
The partial depth repair material of SCC was used and extended with 50% of 13 mm coarse aggregate. Table 1 shows the proportions for the monolithic concrete. Table 2 shows the hardened properties such as compressive strength, splitting strength and fracture energy of the monolithic concrete (M) and self-consolidating concrete (SCC).  The 20 M deformed reinforcing bar had a 479 MPa yield strength and 612 MPa ultimate tensile strength, and the supplier provided this data. The repair material had a rapid strength gain, a fast turnover of repair area and a flowable consistency. The CFRP wrapping sheet was SikaWrap Hex 103C with a weight of 610 g/m 2 . Two types of epoxies were used for the CFRP sheets installation: Sikadur 330 and Sikadur 300. Prior to placing the CFRP, the surface of the concrete was sealed with Sikadur 330, and the CFRP sheets were then saturated manually with Sikadur 300. The properties of CFRP sheets and the two epoxies are shown in Tables 3 and 4, respectively.

Mode of Failure and Cracking Pattern
All of the lap splice beams failed by splitting bond failure (as intended). The first flexural crack in the beams occurred at the upper corners of the notches at the end of the lap spliced region, still within the constant moment region. As the applied load increased, more flexural cracks developed in the spliced zone together with longitudinal cracks along the lap spliced bars. It can be seen from Figure 3a that the longitudinal splitting cracks were developed on the bottom face of the beams from both ends of the splice regions. The unconfined beams had a sudden brittle failure associated with a loud sound resulting from the splitting of the final length of the concrete cover. The bottom face cracks in the unconfined beams formed a V shape (Figure 3b). Sizable chunks of concrete were formed and in some cases spalled off from the concrete cover at failure because of the absence of confinement. The mode of failure was different for the unconfined beams than for the beams confined with FRP sheets. The confined beams had a more ductile mode of failure related to the unconfined beams in which the confinement delayed the splitting of the concrete cover. Unlike the unconfined beams, the concrete cover did not spall off in the beams confined with the FRP sheet. The FRP sheets arrested the concrete cover in place after splitting occurred. The confined beams experienced an increase in their ultimate load, corresponding deflection at failure and flexural crack width compared to the unconfined beams. For the beams confined with the FRP sheet, the crack pattern could not be monitored during testing since the cracks did not penetrate the FRP sheet. At failure, the unconfined beam formed horizontal splitting cracks on the side of the spliced zone along the lap spliced bars. After the test, when the FRP sheets were removed for inspection, it was found that the failure produced smaller pieces of concrete compared to unconfined beams, as illustrated in Figure 4. The mode of failure was different for the unconfined beams than for the beams confined with FRP sheets. The confined beams had a more ductile mode of failure related to the unconfined beams in which the confinement delayed the splitting of the concrete cover. Unlike the unconfined beams, the concrete cover did not spall off in the beams confined with the FRP sheet. The FRP sheets arrested the concrete cover in place after splitting occurred. The confined beams experienced an increase in their ultimate load, corresponding deflection at failure and flexural crack width compared to the unconfined beams. For the beams confined with the FRP sheet, the crack pattern could not be monitored during testing since the cracks did not penetrate the FRP sheet. At failure, the unconfined beam formed horizontal splitting cracks on the side of the spliced zone along the lap spliced bars. After the test, when the FRP sheets were removed for inspection, it was found that the failure produced smaller pieces of concrete compared to unconfined beams, as illustrated in Figure 4. Crystals 2020, 10, x 8 of 14  Table 5 shows a summary of all lap splice beam test results, including the load at failure, the average bond strength at failure, the loaded end slip and the failure mode. The lap splice beams were labeled as follows: the first part (LS) refers to a lap splice and the second part refers to the type of concrete for monolithic concrete (M) and SCC for commercial prepackaged self-consolidating concrete extended with 50% of 13 mm coarse aggregate. The third part refers to the confinement condition (UN for the unconfined beams and F for the beams confined with FRP sheets). The last part represents the corrosion level (C for the control specimens (non-corroded), 7.5 for the specimens with a 7.5% mass loss level and 15 for the specimens with a 15% mass loss level). The splice length and the concrete cover for all beams were fixed as 300 mm and 25 mm, respectively.

Lap Splice Beam
Based on the splice length, the concrete cover, the mass loss level and the confinement condition, all lap splice beams were designed to fail by a splitting failure. The average bond strength was calculated using Equation (3).

4
( 3) where τb is the average bond strength (MPa); db is the bar diameter account for actual mass loss (mm); fs is the steel stress at failure (MPa); and Ls is the splice length (mm) = 300 mm.   Table 5 shows a summary of all lap splice beam test results, including the load at failure, the average bond strength at failure, the loaded end slip and the failure mode. The lap splice beams were labeled as follows: the first part (LS) refers to a lap splice and the second part refers to the type of concrete for monolithic concrete (M) and SCC for commercial prepackaged self-consolidating concrete extended with 50% of 13 mm coarse aggregate. The third part refers to the confinement condition (UN for the unconfined beams and F for the beams confined with FRP sheets). The last part represents the corrosion level (C for the control specimens (non-corroded), 7.5 for the specimens with a 7.5% mass loss level and 15 for the specimens with a 15% mass loss level). The splice length and the concrete cover for all beams were fixed as 300 mm and 25 mm, respectively. Based on the splice length, the concrete cover, the mass loss level and the confinement condition, all lap splice beams were designed to fail by a splitting failure. The average bond strength was calculated using Equation (3). where τ b is the average bond strength (MPa); d b is the bar diameter account for actual mass loss (mm); f s is the steel stress at failure (MPa); and L s is the splice length (mm) = 300 mm.

Effect of Partial Depth Repair with and without FRP Wrapping
The load-deflection curves for the lap splice beams (monolithic specimens (M) and partial depth repaired specimens (SCC) with and without FRP wrapping (F)) are revealed in Figure 5a-c for the control beam sample and the 7.5% rebar mass loss and 15% rebar mass loss beams, respectively. The beams confined with FRP sheets experienced an increase in the ultimate load and the corresponding deflection at failure compared to the unconfined beams. Moreover, the maximum load and corresponding deflection were recorded for the beams wrapped with FRP sheets by 55% and 191%, respectively.
The load-deflection curves increased linearly after initial flexural cracking until the maximum load was achieved in coincidence with bond splitting failure. After the splitting failure, the load-deflection behavior of the unconfined beams was different from the beams confined with FRP sheets. The splitting failure of unconfined beams was accompanied by a sudden drop in the applied load and rise in deflection. Nevertheless, the beams confined with FRP sheets revealed a ductile post-failure behavior, in which the load dropped gradually as the deflection increased. This could be attributed to the FRP sheets limiting the bond crack widths and preventing spalling of the concrete cover. Besides, the partial depth repair was able to restore and increase the capacity of the repaired beams, which had a higher maximum load at failure than that of the monolithic beams.
The LS-SCC-15-F beam failed by bar yielding followed by a splitting failure. It was observed that the stress on the bar exceeded the manufactured reported yield strength of 479 MPa before the splitting failure occurred. After the bar yielded, the deflection of the beam increased with a slight rise in the applied load before failure occurred. Except for this beam, all the other beams failed by splitting bond failure. The re-bond state (constant or slight increase in bond after failure) in the unconfined beams could be from the residual friction between the reinforcing bar and the surrounding concrete where the concrete keys did not shear off, unlike the confined beams. It was also observed that beams confined with FRC sheets sustained a higher load than unconfined beams. The higher yield strength of confined beams could be attributed to the existence of FRP sheets, which prevented the sudden failure of the beams and therefore, sustained higher loads.

Effect of Partial Depth Repair with and without FRP Wrapping
The load-deflection curves for the lap splice beams (monolithic specimens (M) and partial depth repaired specimens (SCC) with and without FRP wrapping (F)) are revealed in Figure 5a-c for the control beam sample and the 7.5% rebar mass loss and 15% rebar mass loss beams, respectively. The beams confined with FRP sheets experienced an increase in the ultimate load and the corresponding deflection at failure compared to the unconfined beams. Moreover, the maximum load and corresponding deflection were recorded for the beams wrapped with FRP sheets by 55% and 191%, respectively.
The load-deflection curves increased linearly after initial flexural cracking until the maximum load was achieved in coincidence with bond splitting failure. After the splitting failure, the loaddeflection behavior of the unconfined beams was different from the beams confined with FRP sheets. The splitting failure of unconfined beams was accompanied by a sudden drop in the applied load and rise in deflection. Nevertheless, the beams confined with FRP sheets revealed a ductile postfailure behavior, in which the load dropped gradually as the deflection increased. This could be attributed to the FRP sheets limiting the bond crack widths and preventing spalling of the concrete cover. Besides, the partial depth repair was able to restore and increase the capacity of the repaired beams, which had a higher maximum load at failure than that of the monolithic beams.
The LS-SCC-15-F beam failed by bar yielding followed by a splitting failure. It was observed that the stress on the bar exceeded the manufactured reported yield strength of 479 MPa before the splitting failure occurred. After the bar yielded, the deflection of the beam increased with a slight rise in the applied load before failure occurred. Except for this beam, all the other beams failed by splitting bond failure. The re-bond state (constant or slight increase in bond after failure) in the unconfined beams could be from the residual friction between the reinforcing bar and the surrounding concrete where the concrete keys did not shear off, unlike the confined beams. It was also observed that beams confined with FRC sheets sustained a higher load than unconfined beams. The higher yield strength of confined beams could be attributed to the existence of FRP sheets, which prevented the sudden failure of the beams and therefore, sustained higher loads.  Figure 6 shows typical load-deflection curves of beams with different levels of mass loss. The beams in the same mass loss category had almost the same stiffness. However, the stiffness decreased slightly with increased reinforcing bar mass loss. As the bar mass loss increased, the total crosssectional area of the bars decreased, reducing the effective stiffness of the beam, thereby increasing the mid-span deflection. Despite the minor differences in load at failure, the average bond strength increased as the mass loss increased, as shown in Figure 7 due to improved bond strength between the concrete and reinforced bars. This occurred because as corrosion increased, the rebar surface became rougher, which improved both the mechanical bonds and friction properties.  Figure 6 shows typical load-deflection curves of beams with different levels of mass loss. The beams in the same mass loss category had almost the same stiffness. However, the stiffness decreased slightly with increased reinforcing bar mass loss. As the bar mass loss increased, the total cross-sectional area of the bars decreased, reducing the effective stiffness of the beam, thereby increasing the mid-span deflection. Despite the minor differences in load at failure, the average bond strength increased as the mass loss increased, as shown in Figure 7 due to improved bond strength between the concrete and reinforced bars. This occurred because as corrosion increased, the rebar surface became rougher, which improved both the mechanical bonds and friction properties.

Bond Stress versus Slip Response
The typical bond stress versus slip response for the control monolithic beams (M) and partial depth repaired specimens (SCC) with and without FRP wrapping (F) is shown in Figure 8a. The beams confined with FRP sheets had higher values of bond strength and slipped at maximum load than the unconfined beams. The confined beams experienced a later start and slower growth of splitting cracks. The beams confined with FRP sheets revealed a rise in the bond strength values and the equivalent slip values by about 34-49%, and 56-260% compared to the unconfined beams, respectively. Regardless of the condition of confinement, the partial depth repair with the SCC had a higher bond strength than the monolithic specimens.
The typical bond stress-slip curve of beams confined with FRP sheets for the control beam and the 7.5% and 15% mass loss beams are shown in Figure 8b. The bond strength and the corresponding slip increased with increasing rebar mass loss. This is because the cross sectional area of the bars decreased with increasing corrosion levels; thus, the force on the bar at bond failure decreased more slowly than the area of the bar [30,31].

Bond Stress versus Slip Response
The typical bond stress versus slip response for the control monolithic beams (M) and partial depth repaired specimens (SCC) with and without FRP wrapping (F) is shown in Figure 8a. The beams confined with FRP sheets had higher values of bond strength and slipped at maximum load than the unconfined beams. The confined beams experienced a later start and slower growth of splitting cracks. The beams confined with FRP sheets revealed a rise in the bond strength values and the equivalent slip values by about 34-49%, and 56-260% compared to the unconfined beams, respectively. Regardless of the condition of confinement, the partial depth repair with the SCC had a higher bond strength than the monolithic specimens.
The typical bond stress-slip curve of beams confined with FRP sheets for the control beam and the 7.5% and 15% mass loss beams are shown in Figure 8b. The bond strength and the corresponding slip increased with increasing rebar mass loss. This is because the cross sectional area of the bars decreased with increasing corrosion levels; thus, the force on the bar at bond failure decreased more slowly than the area of the bar [30,31].

Bond Stress versus Slip Response
The typical bond stress versus slip response for the control monolithic beams (M) and partial depth repaired specimens (SCC) with and without FRP wrapping (F) is shown in Figure 8a. The beams confined with FRP sheets had higher values of bond strength and slipped at maximum load than the unconfined beams. The confined beams experienced a later start and slower growth of splitting cracks. The beams confined with FRP sheets revealed a rise in the bond strength values and the equivalent slip values by about 34-49%, and 56-260% compared to the unconfined beams, respectively. Regardless of the condition of confinement, the partial depth repair with the SCC had a higher bond strength than the monolithic specimens.