Integrating CFRP U-Wrapping and Spike Anchors in the Applications of Flexural Strengthening of RC Beams

2.1. Steel and Concrete

To better represent the concrete type of most of the existing RC buildings, tested beams were fabricated using a concrete mix of designated compressive strength of 30 MPa. All concrete ingredients such as cement, coarse and fine aggregates had been individually tested and verified according to relevant specifications prior to use. On the other hand, 8 mm diameter steel bars had also been tested and verified as per the BS 4449-05 (BS 4449:2005+A3, 2016) to be used for both flexural and shear reinforcement. In addition to the material properties listed above, a total of six standard cubes of 150×150×150mm were casted and tested under compression as per the BS EN 12390-3 (BS EN 12390-3:, 2019) to estimate the actual compressive strength of concrete prior to beams casting. Table 1 shows some of the most important proprieties of the steel and concrete used.

2.2. Carbon Fiber and Epoxy

A 0.13 mm unidirectional woven Carbon Fiber (CF), commercially known as SikaWrap-230C, has been implemented for longitudinal strengthening and u-anchors. This fibre type has been applied to the concrete surface by using a paste-like epoxy resin designed for manual application of carbon fibre known as Sikadur-330. These materials are shown in Figure 2.

On the other hand, a 10 mm diameter (impregnated) unidirectional structural carbon fibre cord, known as SikaWrap Fx-50C, has been utilized as spike anchor which has been manufactured specifically for such applications. These anchors, as will be discussed later, had been impregnated first by using a low viscosity impregnation epoxy resin commercially known as Sikadur-52LP before inserting inside the predrilled anchors’ holes. Figure 2 also shows the anchors fabric and epoxy materials used in this research. In addition, some of the most important technical properties of these materials are provided in Table 2.

Figure 2:

2.3. Design of the Specimens

A total of nine geometrically identical RC beams were designed, fabricated and tested up to failure under four-points-bending scheme. All beams, including the strengthened ones, have been designed to fail in flexure considering to potential increase in the flexural capacity due to the strengthening process. Each beam was fabricated in dimensions of B × H × L equals to 120 × 170 × 1200mm with a clear span of 1000mm and a shear span-to-depth ratio (a/d) of 2.20. Flexural reinforcements consisted of 2D8 mm lower and 1D8 upper steel bars. The beams were strengthened against shear stresses by D8@60mm bars to avoid shear failure especially in strengthened beams as stated earlier. Figure 3 shows the geometry and reinforcement details of the tested beams.

In addition to the control beams (CB) of no strengthening or anchoring techniques incorporated, three groups of strengthened beams were tested. The first group (G1) comprised of three beams: one with only EBR along the bottom span of the beam, which also serves as a reference beam for strengthening benefits, a beam with EBR and additionally anchored through a single 100mm wide CFRP u-wrap at each end and a beam with EBR and additionally anchored by using an 80mm deep single vertical bottom spike anchor. This group was designed to investigate the potential strength gain, or behaviour improvement, when additionally anchoring the EBR layer through a single technique. Group G2, on the other hand, investigated EBR with a single end u-wrap combined with either a double lateral or single vertical bottom SA.

In an attempt to improve the end-anchors contribution through midspan strengthening, Group 3 adopted a configuration of double end u-wraps supplemented by a 100mm wide intermediate u-wrap, with anchorage provided by either single or double lateral spike anchors per sheet. Therefore, the term “double u-wraps” or “double end u-wrap” in this study denotes a configuration comprising three u-wraps: two applied at the beam ends and one positioned at midspan. Figure 4 and Table 3 show the strengthening and anchoring configurations details of all groups.

Figure 3:

2.4. Installation of CF Sheets and Anchors

At the end of concrete curing period, the beams have been removed from water and allowed to dry out before the CF sheets installation process begins. In accordance with the CFRP manufacturer’s guidelines and prior research recommendations, the concrete surface should be thoroughly prepared by removing any residue, debris, or contaminants such as grease, and ensuring it is as rough as possible before CFRP installation. Therefore, after marking and drilling the locations of spike anchors, bottom and side surfaces of the beams had been thoroughly cleansed by a steel grinder, roughened by mechanical steel brush, and thoroughly cleaned prior to the application of CFRP sheets and anchors as shown in Fig. 5 (a) and (b). It should also be noted that the sharp edges of the beams with u-wrapping have been smoothed (rounded) using a steel grinder as per prior research recommendations.

For the beams to be additionally anchored by using CF spike anchors, holes of 12 mm diameter (2 mm larger than the diameter of impregnated spike anchor to be used) and have been predrilled, cleaned by using air pump and special brush, as shown in Figure 5 (c) and then by water jetting to ensure best quality of boding with epoxy resin to be used. The depth of the bottom and side holes was kept as 90mm, which is 10mm deeper than the spike anchors to be inserted for better connectivity with epoxy resin.

Figure 4:

With respect to the CFRP sheets application, the epoxy material has been mixed by a percentage of 4:1 by weight and applied evenly to the surface of concrete by a suitable brush. After that, the pre-cut SikaWrap-230C sheets have been applied to the bottom surface of the beam (dry layup) and a special roller tube was used to increase the quality of glueing and to avoid any entrapped air underneath the CFRP sheets. Vertical u-wrapping installation was performed by following the same process and was done immediately after the bottom CFRP installation.

To prepare the anchors, a 6 mm dry diameter SikaWrap-Fx50 was used for this purpose. Each anchor was precut to length of 130 mm. From that length, 80 mm was impregnated first by Sikadur-52 LP, which is already prepared by mixing the two parts by the recommended percentages and then tightened from its end by using plastic ties to make it easy inserted inside prepared holes. The anchors were left in the room temperature for approximately 10 minutes before inserting them inside the predrilled holes. The diameter of the impregnated anchors was approximately 10 mm; which is 2 mm less than the diameter of the predrilled holes as per manufacturer recommendations. Details of spike anchors preparation and dimensions are shown in Figure 6.

Figure 5:

Figure 6:

After the applications of both bottom CF and u-wrapping sheets, the predrilled holes were filled by Sikadur 330 epoxy resin and then the prepared spike anchors were individually inserted up to the desired depth at each hole. The remaining 50 mm length of each spike anchor was then sprayed out to form a 100 mm diameter fan shape and glued over the preinstalled CF sheets by using Sikadur 330 and the same tools used in the installation of CF sheets. Same technique has been used previously by (Alhelal et al., 2026; del Rey Castillo, Dizhur, et al., 2019; Sun et al., 2020) and has been recommended by the most recent version of (ACI PRC 440.2, 2023). After the successful application of the hybrid anchoring system is done, the beam specimens have been left 7 days for curing at the room temperature. Fig. 7 shows the most important steps of the whole strengthening process.

2.5. Test Configuration

All beam specimens were simply supported with 1000mm clear span and tested under four-points flexural testing using a 200 kN machine. A stiff steel spreader beam was utilized to divide the applied load into two equal point loads with a clear spacing of 340 mm (shear span equals to 330 mm from each support) as shown in Figure 3. To determine the load-deflection relationships and the associated structural performance indices such as stiffness and ductility, a high-resolution electronic dial gauge of 0.01mm accuracy was installed underneath the central point of the beam to measure the deflection. The load was applied gradually with a constant rate of approximately 0.5 kN/s until the failure of beams. Prior to testing, the sides of the beam samples have been white painted to better monitor the initiation and propagation of developed cracks.

Figure 7:

3.1. Critical Loads and Deflections

The cracking, flexural steel yielding, and ultimate loads during test, and the related deflections have been carefully recorded. Table 8 shows the reported loads and deflections for all tested beams. On the hand, the percentage increase/decrease in loads and deflections have been calculated as compared to the control beam and listed inside parenthesis in Table 8 as well.

Since different materials such as steel, concrete and carbon fibre are incorporated in the tension side of the beam, no clearly defined yield point had resulted in almost all load-deflection curves produced. Therefore, the most realistic definition of yield point reported by (R. Park, 1988) has been implemented to calculate the yield load and the corresponding yield deflection of steel bars. The implemented definition considers the equivalent elasto-plastic behaviour with 75% secant stiffness (reduced stiffness) compared to the ultimate load. The same definition has been later adopted by many researchers such as (Mirmiran et al., 1999) and (Pam et al., 2001). Figure 8 shows the definition of the yield point, as well as cracking and ultimate loading states for the control (CB) beam. The same process has been followed to determine the yield load and deflection of each tested beam in this research.

3.2. Structural Indices

The behaviour of structural elements such as beams can be quantified by using some structural indices such as ductility, stiffness and toughness. These indices can define the overall behaviour of the tested element since they utilize realistic data driven from the tested objects. In this section, these indices will be defined and then calculated for each tested beam.

Figure 8:

Ductility is defined as the ability of a structural member to absorb energy before sudden failure. This index utilizes the maximum and the yield displacements and can be calculated as the ratio of the maximum-to-yield displacement (Hussein et al., 2012; Pam et al., 2001). Stiffness index, on the other hand, can be defined as the amount of elastic deformation the element can undergo under the applied loads (Hajighasemali et al., 2014). This definition is like the slope of the elastic part of the load-deflection curve and therefore, can be directly calculated from the load-deflection data. Toughness is another important performance index which defines the amount of load it can be absorbed by the member before failure. This index can therefore be defined as the area under the load-deflection curves (Hajighasemali et al., 2014).

The calculated structural indices and the percentages of increase/decrease of such indices compared to the control beam are summarized in Table 9. For better readability, Figure 9 can also be used to show the load carrying capacity of all tested beams as well as the percentages of increase (or decrease) of the structural indices discussed. As will be discussed later, careful reviewing of Table 9 and Figure 9, the reader can clearly see that although the critical loads have increased, the ductility of all tested beams decreased due to the decrease in the ultimate deflection due to the application of CFRP strengthening.

3.3. Load-Deflection Curves

The load-deflection curves have been constructed for all tested beams. As stated earlier, the load was applied gradually from zero up to the failure load with constant rate of 0.5 kN/s. In all beams, the data of load, and the corresponding deflection, were collected for each 2 kN increment for better construction of the curves. Figures 10–12 illustrates the load-deflection curves of the three tested groups, where the reader can clearly notice the strength gain due to the implementation of either a single (as in Figure 10) or the hybrid (Figure 11 and 12) anchoring system for the EBR strengthening. Nevertheless, the results of all tested beams will be individually discussed. Please notice is that the samples CB and EBR were included in all of these three figures since both can serve as reference specimen in the discussion that will be provided in the next sections.

Figure 9:

3.4. Modes of Failure

The failure mode represents a critical indicator of how a beam ultimately behaves under applied loads. Despite its qualitative nature, failure mode carries greater significance than other structural indices previously discussed. Therefore, the authors aim to qualitatively describe the observed failure mechanisms of each tested beam. A review of Figure 13, which illustrates the failure modes of all tested beams, reveals that the incorporation of anchorage techniques such as u-wrapping or spike anchors effectively altered the failure mechanism. Whereas the externally bonded reinforcement (EBR) beam failed by premature end debonding, the anchored beams exhibited a more favourable mode of failure, namely CFRP rupture/tearing. Alongside the notable increase in load-carrying capacity, this shift in failure mode indicates a more efficient utilization of the CFRP material, ensuring that it performs closer to its intended structural purpose.

The rupture load of the CFRP can be directly related to its tensile properties. For the CFRP sheet used in this study (thickness = 0.13 mm, width = 100 mm, tensile strength = 4300 MPa), the theoretical tensile capacity is approximately 56 kN. This value represents the maximum axial force the CFRP can sustain before rupture. All tested beams exhibited ultimate loads exceeding this theoretical CFRP capacity, especially with the further implementation of spike anchors and u-wraps. The final failure mode was CFRP rupture, confirming that the CFRP material properties governed the ultimate load. This demonstrates that the reported rupture loads are physically justified by the intrinsic tensile strength and modulus of the CFRP.

4.1. Control Beam

The non-strengthened RC control beams, CB, has been designed to fail in flexure. As per the most current ACI 318 Code (ACI Committee 318, 2025), the theoretical design load it can carry is about 49 kN. Test results showed a normal load-deflection behaviour where the deflection has increased gradually as the applied load was increased from zero up to the failure load of 56 kN. The failure mode of this beam sample was a pure flexural failure duo to the propagation of flexural cracks near the midspan of the beam. The ductility index of this beam was 2.0, which is larger than other tested beams in this experimental work.

4.2. Group G1

This group, as discussed previously, consisted of three beams (EBR, BA1 and U1) and has been designed to investigate the effects of individual anchorage system that can be additionally implemented to overcome the debonding failure of EBR strengthened beams. All the beams in this group showed linear load-deflection relationships up to the failure loadings are reached. Beam EBR has showed a greater load-carrying capacity (≈22%) than the control beam. However, it failed suddenly by CFRP debonding from one end of the beam when the ultimate load reached 68 kN. Flexural cracks began to appear at the beam’s midspan once the applied load reached the cracking threshold of approximately 30 kN. With further loading, these cracks propagated deeper into the section and widened progressively, accompanied by the initiation of additional flexural and shear cracks as the applied load increased. The ultimate deflection of the beam sample was lower than the control beam by 6mm, which has resulted in a significant reduction of the ductility as shown in both Figure 9 and Table 9.

On the other hand, BA1 and U1 which are additionally anchored by using either a bottom spike anchor or a single u-wrap at each end, respectively, have supported greater load as compared to EBR sample and both samples have failed by CFRP rupture roughly at midspan of the beam. The failure mode of BA1 and U1 emphasizes the role of anchors in providing extra capacity for the anchored beams and therefore altering the mode of failure from debonding to rupture. As can be seen in Figure 9 and Table 9, that although the ductility of both samples has been reduced compared to the CB sample, the overall performance of BA1 was much better than U1 sample in terms of stiffness and toughness. For example, BA1 showed percentages increase in ultimate load, stiffness and toughness by 81, 42 and 35%, respectively, compared to the control beam.

Figure 10:

Figure 11:

Figure 12:

4.1. Group G2

To investigate the hybrid contribution of both u-wrapping and spike anchors, Group G2, which consisted of two beam sample was designed. Both samples were anchored by using a single u-wrap at each end, however, sample U1SA2 was additionally anchored by two side spike anchors while sample U1BA1 was anchored by a single bottom vertical spike anchor. As shown in Table 8, comparing the ultimate load capacity of both beams to CB specimen, both beams have provided superior ultimate load capacity by more than 84% increase. Stiffness and toughness of both beams, as shown in Fig. 9, have also been respectively increased by more than 22 and 63% for beam U1SA2, and by more than 30 and 23% for beam U1BA1. Compared to each other, on the other hand, the performance of both samples was relatively similar in terms of the determined structural performance indicators and mode of failure. Although both samples surpassed either CB or EBR, the performance of these two beams was almost indistinguishable from beam BA1 tested in group G1. The mode of failure of both samples in this group, was CFRP rupture at midspan triggered by the development of wide and extensive intermediate cracks (IC). In fact, such mode of failure can be encountered due to the higher flexural stresses at midspan compared to the stresses at ends of the beams. Consequently, the anchorage modifications applied at the ends in Group G2 did not yield significant improvements over Group G1. To address this limitation, all beams in Group G3 were additionally anchored at midspan using an extra u-wrap, aiming to reinforce the region of maximum stress and redistribute the demand toward the beam ends for potentially enhanced anchorage efficiency.

4.2. Group G3

In addition to the double u-wraps incorporated at each end of the beams of this group, they have been additionally anchored at midspan by an extra 10mm wide u-wrap. Beam U2 was tested to assess potential improvements in structural behaviour relative to sample U1. Beams U2SA1 and U2SA2 have been additionally anchored by single and double side spike anchors, respectively. The results of this group, in general, have not resulted in any significant improvements compared to the beams of Group G2, except sample U2SA2 which provided percentage increase in ultimate load, stiffness and toughness by 97, 41 and 27%, respectively compared to CB beam. On the other hands, the mode of failure of this group was CFRP rupture at the unanchored area between beams midspan and one of its ends as shown in Figure 13.