Shear Behavior of Basalt Fiber Reinforced Concrete Beams with and without Basalt FRP Stirrups

The investigation included 12 BFRP RC-beams of which six $300\times 200\mathrm{mm}$ ($12\times 8\mathrm{in.}$) beams were nonshear reinforced (NSR) and six $200\times 300\mathrm{mm}$ ($8\times 12\mathrm{in.}$) beams were shear reinforced (SR) with 10 mm BFRP stirrups and #3 steel stirrups. The maximum span length for the beams was determined from Table 8.2 of ACI 440.1 R (ACI 2006). The span length for the NSR-beams was 2,440 mm (8 ft) and was selected to be less than or equal to 13 h for one way slabs owing to their orientation ($b>h$); the span length for the SR-beams was 3,050 mm (10 ft) and was selected to be less than or equal to 10 h for beams.

The RC-beams were reinforced with different areas of BFRP flexural reinforcements that included 10, 13, 16, and 25 mm diameter bars. Table 1 shows a summary of the BFRP flexural and shear reinforcements and the $a/d$ ratios under investigation. The beams were over-reinforced (${\rho }_f/{\rho }_{fb}>1$) to ensure a failure by crushing of the concrete on the top compression fiber as recommended by design codes [ACI 440.1R (ACI 2006); CSA S806 (CSA 2012)] rather than rupture of the FRP main reinforcements. The nonshear reinforced beams were designed with ${\rho }_f/{\rho }_{fb}$ ranging between 2.69 and 14.80 and were divided into two categories. The first category was tested under a span to depth ratio ($a/d$) of 5.65 with ${\rho }_f/{\rho }_{fb}$ ranging between 2.69 and 7.15 for beams 5-10N5, 5-13N5, and 5-16N5. The second category was tested under $a/d$ ratio of 7.0 with ${\rho }_f/{\rho }_{fb}$ ranging between 8.58 and 14.80 for beams 6-16N7, 3-25N7, and 4-25N7. The $a/d$ ratio was increased in the second category of beams for safety reasons and to avoid catastrophic failures in the beams attributable to their high ${\rho }_f/{\rho }_{fb}$. Beams with very high reinforcement ratios needed to be investigated for two reasons: (1) the lack of literature on the behavior of RC beams with very high ${\rho }_f/{\rho }_{fb}$; and (2) to check the validity of the analytical models for shear predictions when using higher ${\rho }_f/{\rho }_{fb}$. The shear reinforced beams were also divided into two categories. The first category was tested under an $a/d$ ratio of 1.5 with ${\rho }_f/{\rho }_{fb}$ ranging between 1.69 and 6.88 for beams 2-13SB1, 3-13SB1, 3-16SB1, and 2-25B1. The second category was tested under three different $a/d$ ratios of 1.5, 2.5, and 3.5 while using the same ${\rho }_f/{\rho }_{fb}$ of 4.03 (3-16SB1, 3-16B2, 3-16B3). The $a/d$ ratio was reduced in the SR-beams to prevent undesired flexural failure owing to the added shear capacity provided by the stirrups. In addition, SR-beams were reinforced with two #3 steel bars placed at the beam’s compression face to support the BFRP and steel stirrups. Fig. 1 shows the stirrups distribution and spacing for the SR-beams. Stirrups were spaced at 127 mm (5 in.) spacing within the shear span region and at 254 mm (10 in.) between the two loading points (constant moment region). Beams 2-13SB1, 3-13SB1, and 3-16SB1 were reinforced with 10 mm BFRP stirrups on one shear span and #3 steel stirrups on the other shear span.

The first number of the beam designation represents the number of flexural reinforcement followed by the size of the BFRP bar (i.e., 5-16 implies five of 16 mm BFRP bars). The letter $N$ indicates no shear reinforcement for NSR-beams. The letter SB represents beams that are half shear reinforced with #3 steel stirrups while their other half with 10 mm BFRP stirrups. The letter $B$ represents beams that are fully reinforced with BFRP stirrups within the shear span region. The last number designates the $a/d$ ratio of the tested specimens (i.e., numbers 5, 7, 1, 2, 3 represents $a/d$ ratios of 5.65, 7.0, 1.5, 2.5, and 3.5, respectively).

The 12 RC-beams were cast from the same concrete mix. The concrete was made with $371\mathrm{kg}/{\mathrm{m}}^3$ ($634\mathrm{lb}/{\mathrm{yd}}^3$) of total cementitious content and a water-to-cementitious material ratio of 0.36. The cementitious materials included Type I portland cement and Class C fly ash (19% partial replacement to cement). The coarse to fine aggregate ratio was $58/42$. The concrete compressive strength at 28 days was 35.9 MPa (5,200 psi) which is designated as the ultimate compressive strength.

The BFRP shear and flexural reinforcements are shown in Fig. 2. The bars and stirrups were pultruded by impregnating them with vinyl epoxy resin followed by surface painting. The flexural bars were then sand-coated to improve their bonding characteristics with concrete. The ultimate tensile strength, strain at rupture, and modulus of elasticity were tested according to ASTM D7205 (ASTM 2006) and ACI 440.3R (ACI 2004) for the BFRP bars, and the results are shown in Table 2.

The beams were tested under a four-point bending load on a simply supported steel rigid frame test set up (Fig. 3). The beams were supported on two steel rollers. The two-point load setup was achieved using a wide flange spread beam placed on two steel plates covering the entire width of the beam. The load was generated by a hydraulic jack located in the center of the beam and applied to the wide flange. The steel plates had steel rollers to allow for their rotation with the beam bending.

Strain gauges were mounted on the concrete compression surface and on the BFRP flexural reinforcemnt at the midspan of the NSR- and SR-beams. In addition, three strain gauges were mounted at the third, fourth, and fifth stirrup from the left and the right end of the SR-beams, all being located in the shear span region as shown in Fig. 3.

The midspan deflections were monitored using a linear variable differential transformer (LVDT) placed at the beam’s midspan. The applied load was measured using a load cell attached to the hydraulic jack. The data from all the instrumented equipment were collected using a data logger system.

The crack growth and the mode of failure were reported for the NSR- and SR-beams.

The load deflection behaviors for the NSR-beams are presented in Fig. 7 and for the SR-beams in Fig. 8. The load deflection behavior is divided into three stages: precracking, transition from the precracking to postcracking, and postcracking.

The load versus flexural strain in concrete and BFRP are presented in Figs. 9 and 10 for NSR- and SR- beams, respectively. The relationship between the load and the flexural strain followed a trend similar to the load-deflection behavior. The flexural strain in the BFRP was negligible prior to the initiation of the first crack. Once the first crack initiated, some beams experienced a sudden jump in the flexural strain without showing an increase in the applied load. This behavior was less pronounced in beams with higher ${\rho }_f/{\rho }_{fb}$ ratios [Figs. 9 and 10(a)]. This strain behavior was also observed in Tomlinson and Fam (2014) and Brik (2003). In addition, as the $a/d$ ratio increased for SR-beams with same ${\rho }_f/{\rho }_{fb}$ [Fig. 10(b)], the BFRP strain in the transition zone decreased.

It was observed that, under the same applied load, the flexural strain in the BFRP bars and concrete decreased when the ${\rho }_f/{\rho }_{fb}$ ratio increased and/or the $a/d$ ratio decreased. The concrete and BFRP strains at failure in NSR-beams decreased as the ${\rho }_f/{\rho }_{fb}$ ratio increased. This was also observed in SR-beams tested under the same $a/d$ ratio except for beam 3-13SB1 which failed in flexure mode by crushing of concrete in the top compression zone. The strain in the BFRP in NSR-beams ranged between 9,600 με for beam 5-10N5 and 4,200 με for beam 4-25N7, whereas the strain in SR-beams ranged between 13,800 με for beam 2-13SB1 and 6,000 με for beam 2-25B1. The concrete strain in NSR-beams ranged between 2,300 με for beam 3-25N7 and 1,600 με for beam 5-13N5, whereas the strain in SR-beams ranged between 2,300 με for beam 2-13SB1 and 1,400 με for beam 2-25B1. Accordingly, none of the beams reached the theoretical maximum compressive strain of 3,000 με at ultimate, because the beams failed in a shear failure mode. In addition, as shown in Fig. 10(b), under the same ${\rho }_f/{\rho }_{fb}$ ratio, the change in the $a/d$ ratio did not show noticeable effect on the ultimate strain at failure.

The tensile strain of the stirrups located within the right and left shear span region of each SR-beam was monitored. Fig. 11 shows the strain deformation of the third and fourth stirrups from the end of beams 2-13SB1, 3-13SB1, and 3-16SB1, and the strain deformation at the fourth and fifth stirrups from the end of beams 2-25B1, 3-16B2, and 3-16B3. It is observed that, under the same applied load the tensile strain in each SR-beam increased in the stirrups located farther from the end supports.

The strain deformations for the stirrups in the right shear span were not consistent with the left shear span located at the same distance from the end supports. This difference was not observed to be driven by the location of the shear failure or the type of stirrups used. However, this difference might be attributed to the location of the shear crack with respect to the stirrup and strain gauge. The strain gauges were mounted at middepth of the stirrup; accordingly the strain readings could have been affected depending on whether the shear crack crossed the stirrup near the location of the strain gauge or slightly above or below it. Beam 2-13SB1 failed in the shear span region reinforced with BFRP stirrups but showed less strain deformation in the BFRP compared with steel stirrups. On the contrary, beam 3-16SB1 experienced higher deformation in the BFRP stirrup but failed in the shear span reinforced with steel stirrups.

It was observed that, under the same applied load, the strain in the fourth stirrup decreased with increasing the ${\rho }_f/{\rho }_{fb}$ ratio for beams with same $a/d$ ratio (1.5), and the strain at ultimate failure was reduced except for beam 3-13SB1 attributable to its unexpected flexure failure mode. In addition, under the same applied load, as the $a/d$ ratio increased from 1.5 to 3.5 the strain in the BFRP stirrups was increased.

The shear capacity ($V_{\text{ult}}$) in the NSR-beams is represented by the shear strength of concrete, $V_c$, whereas the shear strength of the SR-beams is represented by the contribution of concrete ($V_c$) and shear stirrups ($V_s$). The shear strength, the normalized shear strength, and the mode of failure for the NSR- and SR-beams are presented in Table 3. For beams without web reinforcement (NSR-beams), the shear strength is identified when the critical inclined crack is formed followed by a sudden drop in the applied load. Tureyen and Frosch (2002) suggested that for nonshear reinforced beams with an $a/d$ ratio greater than 2.5, the shear strength at the critical inclined crack and the ultimate shear capacity are very close. Therefore, the shear strength for NSR-beams is represented by their shear capacity, $V_{\text{ult}}$.

As shown in Fig. 12, the $V_{\text{ult}}$ for the NSR-beams increased as the ${\rho }_f/{\rho }_{fb}$ ratio increased under the same $a/d$ ratio. This is expected because increasing the flexural reinforcement ratio decreases the deflection under the same load and thereby improves the mechanism of shear transfer by reducing the depth and width of the shear cracks. This was observed in the crack patterns [Fig. 4(a)] where the depth of crack decreased as the ${\rho }_f/{\rho }_{fb}$ ratio increased, thereby allowing more cracks to initiate prior to failure. A similar trend was observed in the SR-beams with an $a/d$ ratio of 1.5 except for beam 2-25B1 [Fig. 12(b)]. The reduced shear strength in 2-25B1 was influenced by the rupture of the BFRP stirrup at the location of failure, whereas beam 2-16SB1 that was made with both BFRP and steel stirrups failed by rupturing of the BFRP main reinforcement at the location of the steel stirrup. It is expected that beam 2-25B1 would achieve the highest shear strength if the failure took place in the main reinforcement. However, the 25 mm BFRP bar provided a higher transverse shear strength which caused a premature failure in the BFRP stirrup leading to a reduction in the ultimate shear strength of the beam. In addition, the strain of the failed BFRP stirrup for the 2-25B1 beam was less than 5,000 με, which indicates that the stirrup failed at a much lower stress than its ultimate tensile strength. According to Ahmed et al. (2014), the bend zone in the FRP stirrup reduces its tensile strength resulting from the combination of both the fiber weakness in the transverse direction and the kinking of fibers at the curvature. The effect of the $a/d$ ratio was also apparent in the NSR- and SR-beams. As shown in Fig. 12(b), under the same ${\rho }_f/{\rho }_{fb}$ ratio, the shear strength decreased linearly with increasing the $a/d$ ratio. It is noted that the decreasing trend in the shear strength would not be linear if higher $a/d$ ratios were considered.

The normalized shear versus deflection behavior of the NSR- and SR-beams that have similar ${\rho }_f/{\rho }_{fb}$ ratios are shown in Fig. 13. The normalized shear is calculated in the form ($V_{\exp }/\sqrt{f_c^{\prime }}bd$) and multiplied by the ratio of the beam’s shear span $a$ to the shear span of beam 3-13SB1. Based on the following assumption, the shear deflection behavior shows an excellent curve fitting between the NSR- and SR-beams with equivalent ${\rho }_f/{\rho }_{fb}$ ratios. It is revealed that the shear stirrups had significantly improved the shear strength in the concrete beams. The added stirrups in beams 3-13SB1, 3-16SB1, and 2-25B1 had an additional shear capacity equivalent to 1.7, 1.2, and 0.7 times the shear capacity of their counterpart NSR-beams (5-10N5, 5-13N5, and 5-16N5), respectively. This indicates that the shear resistance in the stirrups was less pronounced when the ${\rho }_f/{\rho }_{fb}$ ratio increased. As it was mentioned earlier, the shear resistance in beam 2-25B1 was also reduced owing to the failure in the BFRP stirrup rather than the main reinforcement.