Behavior of Shear-Critical Recycled Aggregate Concrete Beams Containing BFRP Reinforcement

Abstract

The shear performance of recycled aggregates beams reinforced with basalt fiber-reinforced polymer (BFRP) bars is evaluated and compared with that of similar beams made with natural aggregates (NA). Six beams with a shear span-to-effective depth ratio (a/d) of 3.0 were tested to failure. Test variables consisted of the recycled concrete aggregate (RCA) replacement percentage (60 and 100%) and the presence of BFRP stirrups in the shear span. Experimental results showed that a RCA replacement of 60% marginally reduced (5%) the shear capacity. However, the reduction in the shear capacity was more pronounced (17%) for the specimen made with 100% RCA. The contribution of BFRP stirrups to the shear capacity decreased with an increase in the RCA replacement percentage. The width of the major shear crack at a given value of load was higher for the beams with RCA. The deflection values at the ultimate load were greater for beams made with RCA. A codified analytical approach as well as a model published in the literature were employed to predict the shear capacity of the tested beams. Predictions of the codified analytical approach were very conservative. The analytical model published in the literature provided a more reasonable prediction for the shear capacity of the tested beams than that of the codified analytical approach.

1. Introduction

The global production of concrete has reached approximately 25 billion tons annually [1]. Typically, a concrete mixture comprises 20–25% cementitious paste and 75–80% aggregates [2,3]. Coarse aggregates are obtained from the crushing of natural rocks [3]. The increasing need for coarse aggregates in concrete preparation is a threat to the depletion of its natural resources [4]. Therefore, there is a need for an alternative resource of coarse aggregates, such as recycled concrete aggregates (RCA) from construction and demolition waste (CDW) [5]. CDW can be defined as waste materials produced during the construction of new structures, repair and renovation works, partial demolition of buildings, and full demolition of buildings [6]. CDW comprises 36% of solid waste generation [6,7]. The mismanagement of CDW could result in the health deterioration of about 200 million workers and people living in the vicinity of demolition activities [6]. The extraction of RCA from CDW could be one of the potential resources of coarse aggregates. The benefits of using RCA include a reduction in the cost of construction material, preservation of natural resources, reductions in global harmful CO₂ emissions, preservation of landfills, and business opportunities for the recycling sector [8].

Generally, concrete mixtures prepared with RCA exhibit inferior mechanical and durability properties in comparison to mixtures made from NA [9,10,11,12,13]. However, the shear response of large-scale beams made with RCA is not fully understood. Some researchers observed inferior shear response of RCA-based beams compared to their counterpart NA-based beams [14,15,16], while others reported no noticeable difference between the shear response of NA- and RCA-based beams [17,18,19]. Concrete slabs made with RCA tended to exhibit a reduced punching shear resistance than that of similar slabs made with NA [20]. It is noteworthy that provision 26.4.1.2.1(c) in the ACI 318-19 Building Code [21] permits the use of recycled aggregate if approved by the licensed design professional and the building official based on documentation that demonstrates compliance with the mechanical properties and durability required in structural design. The ACI 318-19 Building Code [21] also requires experimental evidence for verification of the aggregate consistency in addition to adopting a quality control program to achieve consistency of properties of the concrete throughout the project duration.

Previous research was focused primarily on the shear response of recycled aggregates concrete beams reinforced with conventional steel reinforcement. Corrosion is a major issue associated with steel reinforcement bars, which results in reduced bond strength, reduction in effective cross-sectional area, and loss of serviceable life [22,23,24]. Steel reinforcement bars are more prone to corrosion damage in harsh conditions [25]. Therefore, the use of fiber-reinforced polymer (FRP) bars is a potential alternative to conventional steel reinforcement [26]. FRP bars are lightweight, non-magnetic, and corrosion-free [27,28]. There are different commercially available FRP bars, i.e., carbon (CFRP), glass (GFRP), aramid (AFRP), and basalt (BFRP). The BFRP bars are selected over commercially available FRP bars, owing to their unique characteristics. BFRP bars are made from natural basalt rocks and are eco-friendly [29]. They are lightweight, anti-corrosive, non-magnetic, and non-conductive. Additionally, they are more chemically stable than counterpart GFRP bars [29,30]. Codes and standards that regulate the use of FRP reinforcing bars in the construction industry have been developed and they are currently available for the engineering community [31,32].

The structural shear response of beams reinforced with FRP bars has been investigated. However, the shear behavior of beams made with RCA and reinforced with BFRP bars and BFRP stirrups is uncertain, since there is a scarcity of information available in the literature. Issa et al. [33] tested 12 beams to investigate the shear performance of concrete beams reinforced with BFRP bars and stirrups. Test variables comprised flexural reinforcement ratio and a/d ratio. The shear capacity of beams with and without stirrups increased with an increase in the amount of BFRP reinforcement for the same a/d ratio. Meanwhile, a reduction in the shear capacity of beams was recorded with an increase in the a/d ratio. The beams without stirrups exhibited a shear-tension mode of failure, while beams having a/d ≤ 2.5 and reinforced with stirrups experienced a shear-compression mode of failure. Said et al. [34] studied the shear behavior of beams reinforced with laboratory-produced GFRP bars and stirrups. Test variables consisted of concrete compressive strength and GFRP web reinforcement ratio. The use of stirrups at a spacing of 215 and 100 mm improved the shear capacity of beams with concrete compressive strength of 25 MPa, by 41 and 82%, respectively, relative to the beam without web reinforcement. The shear capacity of beams made with concrete of 45 and 70 MPa compressive strength and reinforced with stirrups at 100 mm spacing was 53 and 76% more than that of the beam made with concrete having a compressive strength of 25 MPa, respectively. The majority of the beams exhibited a diagonal tension mode of failure. Al-Hamrani et al. [35] studied the shear performance of beams reinforced with BFRP bars and stirrups. The test variables comprised a/d ratio, reinforcement ratio, and spacing between the stirrups. The beams with higher a/d ratio exhibited lesser shear capacity and increased deflection. In addition, a higher reinforcement ratio led to an increase in the shear capacity.

Liu et al. [36] investigated the shear performance of recycled aggregates deep beams with varying a/d ratios and reinforced with BFRP bars. All of the tested beams exhibited a typical shear mode of failure due to the lack of stirrups. The splitting of the diagonal strut was the dominant mode of failure for the tested beams. The increase in the a/d ratio led to reduced ultimate strength and mid-span deflection of BFRP-reinforced recycled aggregates beams. The compressive strength of recycled aggregates concrete affected the shear strength of deep beams. However, the increment was not significant due to the increase in the RCA. Younis et al. [37] tested six beams with a/d = 2.7 to investigate the shear behavior of recycled aggregates concrete beams reinforced with GFRP stirrups. An average reduction of 12% in shear capacity was recorded for beams made with 100% RCA. The altering of the shear reinforcement type (steel versus GFRP) resulted in an insignificant effect on the shear capacity of beams. The beams without stirrups failed in shear, displaying diagonal shear crack. However, specimens reinforced with stirrups exhibited a combined shear/flexural failure mode.

Karimi Pour et al. [38] examined the effect of RCA replacement on the shear capacity of beams reinforced with longitudinal GFRP bars. For beams with a/d = 1.5 and reinforced with 8 mm diameter GFRP longitudinal bars, beams with 25 and 50 RCA exhibited 8 and 12% reductions in the shear capacity relative to the control beam made with NA. Their counterpart beams reinforced with 10 mm diameter GFRP bars experienced shear capacity reductions of 6 and 10%. However, for beams with a/d = 2.5 and reinforced with 8 mm GFRP bars, 22 and 35% reductions were recorded in the shear capacity of beams at RCA replacement of 25 and 50%, respectively, compared to the control beam made with NA. Their counterpart specimens reinforced with 10 mm GFRP bars exhibited 5 and 25% strength reductions. The researchers reported that the toughness of beams decreased due to incorporating RCA in the concrete mixtures. The lower toughness of RCA-based beams was attributed to the inferior properties of RCA.

A life cycle assessment study conducted by Hossain et al. [39] revealed that the replacement of NA with RCA could reduce greenhouse gas emissions by 65% and 58% of non-renewable energy resources could be conserved. In addition, Braga et al. [40] stated that the use of RCA could save 80% of the cost in comparison with natural granite aggregates. Another life cycle assessment study [41] indicates that the global warming potential of BFRP bars is 74% lower than steel and 44% lower than GFRP. In addition, a study conducted by Inman et al. [42] revealed that there is a reduction of 62% in anthropogenic emissions while using BFRP bars instead of conventional steel reinforcement. Due to the scarcity of freshwater worldwide, there is a global interest in using seawater-mixed concrete [43,44,45,46,47,48]. The use of seawater-mixed concrete necessitates employing non-metallic reinforcement to eliminate corrosion problems. Sbahieh et al. [43] assessed the environmental impact of steel-reinforced beams made with desalinated fresh water and GFRP bars made with seawater for concrete mixtures. The results indicate that GFRP-reinforced beams outperformed steel-reinforced beams in 9 out of 14 environmental impact categories. In another study, Sbahih et al. [44] concluded that GFRP- and CFRP-reinforced concrete beams made with seawater and sea sand have better environmental impact than those of steel-reinforced beams. Ebead et al. [45] mentioned that tensile strength and bond performance of GFRP bars in concrete prepared with seawater is similar to that of bars in concrete mixture made with fresh water. Younis et al. [46] conducted a life cycle cost (LCC) analysis study of GFRP-reinforced RCA-based concrete using seawater. The results indicated that for a 100 years study period and 0.7% discount rate, the LCC of concrete with seawater, RCA, and GFRP bars was 50% lower than that of traditional concrete mixture with black-steel reinforcement. Such promising data highlight the social and environmental benefits of using RCA and non-metallic reinforcement in the construction industry.

Little knowledge is available in the literature on the shear behavior of BFRP-reinforced concrete beams made with RCA. The aim of the paper is to study how the use of different percentages of RCA would influence the shear behavior of BFRP-reinforced concrete beams with and without stirrups. Such information is necessary before RCA can be routinely used along with non-metallic reinforcing bars in the construction industry. The research work comprised laboratory testing of large-scale concrete beams. The effect of partial or full replacement of natural aggregates with RCA on the shear behavior of concrete beams with and without BFRP stirrups was elucidated. The accuracy of published analytical models to predict the shear capacity of BFRP-reinforced concrete beams made with RCA was evaluated. Findings of this research are anticipated to contribute to the widespread use of recycled and non-ferrous construction materials, thus leading to sustainable steel-free buildings and structures.

2. Experimental Program

The experimental program consisted of testing six slender concrete beams with a/d of 3.0. Test parameters comprised RCA replacement percentage and the presence of BFRP stirrups in the shear span.

2.1. Test Matrix

The test matrix is listed in

Table 1. Test matrix.

Group	Specimen ID	Presence of Stirrups	RCA Replacement (%)
1	SB-R0-NS	-	0
	SB-R60-NS	-	60
	SB-R100-NS	-	100
2	SB-R0-S	√	0
	SB-R60-S	√	60
	SB-R100-S	√	100

. The specimens are identified as “SB-Rx-NS/S”. The notation “SB” denotes slender beam, “R” stands for RCA, “x” denotes RCA replacement percentage, “NS” means no BFRP stirrups in the shear span, and “S” indicates the presence of BFRP stirrups in the shear span. For instance, SB-R60-S denotes a beam specimen made with 60% RCA and reinforced with BFRP stirrups.

2.2. Test Specimens

Details of tested slender beam specimens with and without BFRP stirrups in the shear span are shown in Figure 1. The length of the beam specimen was 2600 mm, and the cross-sectional dimensions were 150 × 300 mm (width × depth). The beam specimen was designed to ensure a shear failure mode without yielding the longitudinal BFRP bars. The BFRP reinforcement in the tension zone of the beam comprised four bars with a nominal diameter of 10 mm (4Φ10) placed at a distance of 250 mm from the face of the compression side and with a tension reinforcement ratio (ρ_f) of 0.0076. The BFRP reinforcement in the compression zone consisted of two bars with a nominal diameter of 6 mm (2Φ6) placed at a depth of 30 mm and with a compression reinforcement ratio (ρ′_f) of 0.0016. The shear reinforcement (if placed) comprised BFRP stirrups with a diameter of 6 mm placed at a spacing of 125 mm in the shear span. The placement of BFRP stirrups satisfies the minimum shear reinforcement requirement, as suggested in ACI 440.1R-15 [31]. Two steel stirrups were placed outside the left support to facilitate anchorage and hold the position of reinforcement bars during the casting of the beams. The beam outside the shear span was reinforced with steel stirrups of 8 mm diameter at a spacing of 75 mm.

2.3. Materials

Concrete mixtures were prepared in the laboratory. Dune sand (DS), acquired from Al Ain Municipality, United Arab Emirates (UAE), was used as the fine aggregate. The natural aggregates (NA), with a nominal maximum size (NMS) of 19 mm, were obtained from Stevin Rock, Ras Al Khaimah, UAE. The RCA were provided by Al Dhafra Recycling Plant, Abu Dhabi, UAE, with an NMS of 25 mm. The recycling plant acquires aggregates from crushing demolished buildings with an unspecified concrete strength. The RCA obtained from the recycling plant were used “as is” in the preparation of concrete mixtures. The inferior characteristics of RCA results in the strength degradation of RCA-based concrete relative to that of NA-based concrete. A study conducted by Xia et al. [49] revealed that compressive strength and splitting tensile strength of concrete mixtures made with RCA with an NMS of 19 mm were 24% and 11% higher than those of respective mixtures made with RCA having an NMS of 31.5 mm. ASTM Type-I [50] cement was used as cementitious material and was obtained from Emirates Cement Factory, Al Ain, UAE. The workability of the base concrete mixture was maintained in the range of 200–220 mm by adding a superplasticizer (SP) acquired from BASF, Abu Dhabi, UAE. The physical properties of the coarse and fine aggregates are listed in

Table 2. Physical properties of aggregates.

Property	Unit	ASTM Standard	DS	NA	RCA
Dry-rodded density	kg/m3	C29 [51]	1663.0	1635.0	1563.0
Los Angeles abrasion	%	C131 [52]	-	16.0	32.6
Absorption	%	C127 [53]	-	0.40	6.58
Specific gravity	-	C127 [53]	2.77	2.82	2.63
Fineness modulus	-	C136 [54]	1.45	6.82	7.44
Surface area	cm2/g	C136 [54]	116.8	2.49	2.50
Soundness (MgSO4)	%	C88 [55]	-	1.2	2.78

. The dry-rodded density of the RCA was 4.4% less than NA. Furthermore, mass loss in the abrasion test was 6.82 and 7.44% for NA and RCA, respectively. Additionally, the water absorption capacity of NA and RCA was 0.40 and 6.58%, respectively.

BFRP bars and BFRP stirrups were acquired from ReforceTech AS, Røyken, Norway [30]. Young’s modulus of BFRP bars was 43 GPa. The guaranteed minimum tensile strength of 6 and 10 mm BFRP longitudinal bars was 904 and 848 MPa, respectively. The ultimate strain of 6 and 10 mm BFRP bars was 2.1 and 2.0%, respectively [30]. BFRP stirrups of 6 mm diameters were used in the shear span. Steel stirrups of 8 mm diameter and a measured yield strength of 562 MPa were used. Photos of BFRP longitudinal bars are shown in Figure 2.

2.4. Mixture Proportioning

The benchmark mixture was prepared following the provisions of ACI 211.1 [56], with a design cylinder compressive strength (f′_c) of 30 MPa, which approximately corresponds to a cube compressive strength (f_cu) of 40 MPa.

Table 3. Mixture proportion of different concrete mixtures.

Mixture ID	Mass (kg/m3)
	Cement	DS	NA	RCA	Water	SP
M-R0	470	659	1080	0	230	0.47
M-R60	470	659	432	648	230	0.47
M-R100	470	659	0	1080	230	0.47

lists the proportions of concrete mixtures used in the current study. The concrete mixtures are identified as M-Ry, where “M” stands for concrete mix, “R” means recycled concrete aggregates, and “y” exhibits the percentage of RCA. For instance, M-R60 indicates a concrete mixture made with 60% RCA. The amount of cement, DS, water, and SP was kept the same in all concrete mixtures. The respective changes were made in the amount of NA and RCA.

2.5. Specimens Preparation

Reinforcement cages for beam specimens were prepared in the laboratory. Photos of reinforcement cages are shown in Figure 3. Concrete mixtures were prepared in the laboratory employing a mechanical mixer. The mixing process started with the addition of RCA and/or NA, as application, in the mixer. To account for the water absorption of RCA and NA needed to reach a saturated surface dry (SSD) condition, the required amount of water was added. Other dry mixture components including DS and cement were added, and mixing was performed for about three minutes. Finally, the specified amount of free water containing SP was added to the concrete mixture and mixing was performed for an additional three minutes. This mixing process led to a homogenous concrete mixture. Wooden molds, braced on the outside, were used to cast beam specimens. The reinforcement cage was placed at designated locations. The concrete mixture was placed inside a wooden mold, and compaction was performed using an electric vibrator. The specimens were water-cured, employing wet burlap for 28 days. Concrete cylinders and cubes were also cast to determine the mechanical properties of concrete mixtures. The cylinder and cube specimens were water-cured for 28 days. Figure 4 shows photos of concrete placement and the level surface of the beam specimen.

2.6. Properties of Concrete Mixtures

The slump of freshly made concrete was measured using a slump cone test following the provisions of ASTM C143 [57]. The compressive strength of concrete mixtures was determined from compression tests on cube specimens of size 150 × 150 × 150 mm and cylinder specimens of size 150 × 300 mm (diameter and height) following BS EN 12390-3:2019 [58] and ASTM C39 [59], respectively. The splitting tensile strength was determined using cylinder specimens of size 150 × 300 mm (diameter and height) following the provisions of ASTM C496 [60]. The modulus of elasticity of concrete was determined using a cylinder specimen of size 100 × 200 mm (diameter and height) following ASTM C469 [61]. Three replicate specimens were tested for each property. The slump and average values of cube compressive strength (f_cu), cylinder compressive strength (f′_c), modulus of elasticity of concrete (E_c), and splitting tensile strength (f_sp) are listed in

Table 4. Workability and mechanical properties of concrete mixtures.

Mixture ID	Slump (mm)	fcu (MPa)	f′c (MPa)	Ec (GPa)	fsp (MPa)
M-R0	210	44.9 ± 2.83	32.2 ± 4.33	30.5 ± 1.70	2.38 ± 0.27
M-R60	200	35.9 ± 0.78	26.5 ± 3.57	23.3 ± 0.02	2.00 ± 0.04
M-R100	205	31.2 ± 0.56	24.2 ± 3.78	17.4 ± 0.48	1.82 ± 0.05

.

2.7. Test Setup and Instrumentation

All beams were tested to failure. The beam specimens were placed on steel pedestal supports that were 2250 mm apart at the center. The shear span was divided into three segments of equal longitudinal dimensions of 250 mm. The load was applied on the top surface at a distance of 750 mm from the center of the left pedestal support employing an MTS 201.45 hydraulic actuator, made by MTS Systems Corporation, Eden Prairie, MN, USA. A displacement-controlled loading was applied at a rate of 1.5 mm/min until failure of the beam. The applied load was recorded through a 500 kN load capacity load cell made by Tokyo Measuring Instruments Laboratory, Japan. A linear variable differential transducer (LVDT) manufactured by Tokyo Measuring Instruments Laboratory, Japan, was placed under the loading point at the bottom soffit of the beam to record the deflection of the slender beam. A 5 mm long strain gauge (SG), manufactured by Tokyo Measuring Instruments Laboratory, Japan, was bonded on the bottom surface of tensile BFRP reinforcement at a point aligned with the applied load. One SG to each leg of the BFRP stirrup was bonded at a vertical distance d/2 = 125 mm from the face of the compression side. A strain gauge of 60 mm in length was placed just below the loading point at the top surface to measure the compression concrete strain. Crack displacement transducers (K-5A) manufactured by Tokyo Measuring Instruments Laboratory, Japan, were placed diagonally at each segment of the shear span (see Figure 5) to capture the diagonal displacement across cracks. The load cell, LVDT, crack displacement transducers, and SGs were connected with a data acquisition system to record data. The crack development during testing was marked and respective load values were recorded. The schematic diagram of the test setup is displayed in Figure 5, while a photo of the test in progress is shown in Figure 6.

3. Experimental Results and Discussion

3.1. Crack Pattern and Failure Mode

The crack pattern of tested beams with and without BFRP stirrups in the shear span is shown in Figure 7. Beam specimens in Group 1 (without BFRP stirrups) exhibited a shear-tension mode of failure, which involved the formation of a diagonal crack connected to a longitudinal splitting crack at the level of the flexural reinforcement. This mode of failure typically occurs in beams without stirrups [62]. It is worth noting that flexural cracks appeared first in the long shear span. The failure mode of the control beam (SB-R0-NS) is shown in Figure 7a. The diagonal shear crack in SB-R0-NS was initiated in segment 3 of the shear span at a shear load value of 17.4 kN and propagated toward the loading point with an increase in applied load. A longitudinal crack at the level of tension BFRP reinforcement was developed close to the failure load and propagated toward segment 1 of the shear span. Figure 7b shows the crack pattern and failure mode of SB-R60-NS. A diagonal shear crack appeared in segment 3 of SB-R60-NS at a shear load value of 7.5 kN and propagated toward the loading point and toward the bottom face of the beam with an increase in load. At a shear load value of 23.5 kN, a longitudinal splitting crack originated at the level of tension reinforcement and propagated toward the major diagonal shear crack developed earlier and toward the support. Specimen SB-R100-NS experienced the development of a diagonal shear crack at segment 3 of the shear span at a shear load value of 7.4 kN. The width of the crack increased with an increase in applied load (Figure 7c). At a shear load value of 22 kN, a longitudinal splitting crack appeared at the tension reinforcement and propagated toward the support and diagonal shear crack.

The presence of stirrups in beam specimens SB-R0-S and SB-R60-S of Group 2 prevented the development of the longitudinal splitting crack, and hence, changed their mode of failure from shear-tension to diagonal tension. The diagonal tension mode of failure involves initiation of flexural-shear cracks in the shear span, which, in the presence of light shear reinforcement, increase in length and width, causing a loss of shear integrity and failure of the beam [62]. Figure 7d,e show the crack pattern of specimens SB-R0-S and SB-R60-S, respectively. The major diagonal shear crack appeared in segment 2 of both beam specimens at shear load values of 27.7 and 19.1 kN, respectively. Figure 7f shows the crack pattern and failure mode of specimen SB-R100-S. The major diagonal crack initiated in segment 2 of SB-R100-S at a shear load of 14.7 kN and propagated toward the loading point and bottom face of the beam. Due to the weak properties of the concrete of SB-R100-S, a local crushing occurred at the tip of the major shear crack at the onset of failure, rendering a shear-compression mode of failure. A fine longitudinal crack at the level of tension reinforcement was also observed at failure of SB-R100-S.

3.2. Shear Load-Deflection Response

The shear load-deflection responses of the tested beams are shown in Figure 8. The shear load represents the reaction value of the left support of beam. The deflection was recorded under the loading point at 750 mm from the left support. All of the tested beams exhibited a linear load-deflection response before the flexural cracking load was reached. Additionally, the deflection of all of the tested beam specimens at flexural cracking load was almost the same irrespective of the RCA replacement percentage or the presence of BFRP stirrups in the shear span. After the initiation of flexural cracking, the deflection of the beams increased until the peak load was reached. At peak load, the deflection of the beams made with RCA was higher than that of the corresponding control beams made with NA. These observations are consistent with findings reported by Younis et al. [37], where the deflection of a GFRP-reinforced beam prepared with 100% RCA at ultimate load was two times that of its counterpart made with NA.

The results of the tests are listed in

Table 5. Summary of test results.

Group	Specimen ID	Shear Cracking Load	Shear Capacity	Shear by BFRP Stirrups	Deflection Capacity	Failure Mode
		Vcr (kN)	Vu,exp (kN)	Vf,exp (kN)	Δpeak (mm)
1	SB-R0-NS	17.4	38.9	-	23.9	Shear-tension
	SB-R60-NS	7.5	36.9	-	31.8	Shear-tension
	SB-R100-NS	7.4	32.1	-	25.6	Shear-tension
2	SB-R0-S	24.7	49.9	11.0	32.6	Diagonal tension
	SB-R60-S	19.1	45.4	8.5	39.8	Diagonal tension
	SB-R100-S	14.7	39.6	7.5	39.0	Shear-compression

. The shear cracking load of beams SB-R60-NS and SB-R100-NS with RCA was 43% of that of SB-R0-NS made with NA. For the beams without BFRP stirrups, increasing the RCA replacement percentage from 60 to 100% had no effect on the shear cracking load. The presence of stirrups delayed the appearance of the first shear crack. The shear cracking load of beams SB-R60-S and SB-R100-S with RCA was on average 68% of that of SB-R0-S made with NA. For the beams with BFRP stirrups, increasing the RCA replacement percentage from 60 to 100% reduced the shear cracking load by 23%. The beams with RCA exhibited a reduced stiffness and a higher deflection capacity than those of the control beams made with NA. Beam specimens without stirrups and made with 60 and 100% RCA exhibited 33 and 7% higher deflection at shear capacity compared to that of the control specimen (SB-R0-NS). Additionally, deflection increments at peak load of 22 and 20% were recorded for beams made with 60 and 100% RCA and reinforced with BFRP stirrups in comparison to that of the NA-based beam (SB-R0-S), respectively.

The RCA replacement of 60 and 100% reduced the shear capacity of beams without stirrups by 5 and 17%, respectively, relative to that of SB-R0-NS. However, in the presence of BFRP stirrups in the shear span, the shear capacity of beams made with 60 and 100% RCA was reduced by 9 and 21% relative to the strength of their counterpart, SB-R0-S. Thus, the presence of stirrups did not affect the percentage reduction in the shear capacity of beams caused by RCA replacement. The reduction in shear capacity with RCA replacement is attributable to the high porosity of RCA and the weak interfacial transition zone between old adhered mortar with RCA and new mortar [37,63]. The reduction in shear capacity with RCA replacement is consistent with findings reported by Younis et al. [37], where RCA replacement of 100% led to a 12% reduction in shear capacity on average. In the current study, the presence of BFRP stirrups enhanced the shear capacity of the NA-based beam SB-R0-S by 28% compared to that of its counterpart SB-R0-NS without stirrups. An improvement of 23% was recorded in the shear capacity of the beams made with 60 and 100% RCA due to the presence of BFRP stirrups relative to that of their counterparts without stirrups. It is noteworthy that the gain in shear strength experienced by the beams with RCA caused by the addition of BFRP stirrups (23%) was comparable to that of the beams with NA (28%). These findings are consistent with results reported by Younis et al. [37], where the presence of GFRP stirrups improved the shear capacity of NA- and RCA-based (100% RCA replacement) concrete beams by 34 and 40%, respectively. Table 5 shows that the shear contribution of BFRP stirrups to the shear capacity, V_f,exp, tended to decrease with an increase in the RCA replacement percentage. The shear strength of BFRP-reinforced concrete beams is reached when the shear crack width reaches a limiting value that causes loss of shear integrity [31,32]. Since the major shear crack initiated earlier in RCA-based beams due to their reduced mechanical properties, the shear crack width increased at a higher rate and the loss of shear integrity occurred earlier in RCA-based beams, causing a reduction in the contribution of BFRP stirrups to the shear capacity.

3.3. BFRP Reinforcement Strain

Figure 9 shows the relationship between shear load and strain in tension BFRP reinforcement of tested beams. The left and right BFRP tension reinforcement bars exhibited similar strain versus shear load responses. The strain values increased insignificantly at the uncracked stage. The strains in tension reinforcement increased at a higher rate in the post-cracking stage. For a given load value, beams made with 60 and 100% RCA exhibited higher strain values (Figure 9a) than those of the control NA-based beam (SB-R0-NS). The maximum strain values in the tension reinforcement at peak load for beams without stirrups were in the range of 9221–9928 µƐ (i.e., 46–50% of the rupture strain of BFRP bars).

A similar trend was observed for the beams with BFRP stirrups (Figure 9b). Due to their higher failure load, the strain values in the tension reinforcement of beams with BFRP stirrups at peak load were higher than those of their counterparts without BFRP stirrups. The maximum strain values in the tension reinforcement at peak load for beams with stirrups were in the range of 13,753–14,779 µƐ (i.e., 69–74% of the rupture strain of BFRP bars). The shear load versus tension reinforcement strain response observed in the current study is consistent with findings reported by Issa et al. [33], where negligible strains were recorded in BFRP tension reinforcement before the initiation of flexural cracks followed by an increase in the rate of strain after cracking.

3.4. Stirrup Strains

Stirrup strains were recorded for specimens SB-R0-S, SB-R60-S, and SB-R100-S containing BFRP stirrups in the shear span. Figure 10 demonstrates the strains recorded in the two vertical legs of BFRP stirrups that were intersected by a major diagonal shear crack. At the pre-crack initiation stage, all beams exhibited a negligible increase in the stirrup strains with the applied load. The stirrups started to exhibit strains after initiation of the shear crack. The stirrups in the control beam made with NA (SB-R0-S) started to strain at a higher load value relative to that of the beams made with 60 and 100% RCA. The shear loads at which beams SB-R0-S, SB-R60-S, and SB-R100-S started to strain at a higher rate were approximately 25, 19, and 15 kN, respectively. The strain in the BFRP stirrups increased in a quasi-linear trend for specimen SB-R0-S and almost linearly for SB-R60-S and SB-R100-S. The stirrup strain in SB-R100-S made with 100% RCA increased at a rate higher than that of its counterpart SB-R60-S made with 60% RCA. The higher rate of stirrup strain implies a lower concrete contribution to the shear resistance. The SG in one of the stirrup legs of SB-R0-S was able to record the strains up to the peak load. However, the SGs attached to the stirrups of SB-R60-S and SB-R100-S failed at approximately 83 and 92% of the ultimate load. The maximum strain values recorded in the stirrups were in the range of 3903–5857 µƐ, which corresponded to 19–28% of the rupture strain of BFRP stirrups. It is noteworthy that values of the strains recorded in the BFRP stirrups prior to failure were almost equal to or higher than the limiting value of 0.004 recommended by the ACI 440.1R-15 [31].

3.5. Concrete Strain

The longitudinal concrete strain was measured just below the loading point.

Table 6. Concrete strains at shear capacity.

Group	Specimen ID	Longitudinal Concrete Strain (µƐ)
1	SB-R0-NS	-
	SB-R60-NS	1828
	SB-R100-NS	1754
2	SB-R0-S	1957
	SB-R60-S	2417
	SB-R100-S	2105

lists the concrete strain values of specimens. The concrete strain at peak load for the beams without stirrups was approximately 60% of the concrete crushing strain (3000 µƐ) [21]. Beams with BFRP stirrups failed at a higher load than that of their counterparts without BFRP stirrups, and hence, exhibited higher concrete strains at peak load. The concrete strain at peak load for the beams with stirrups was 65–81% of the concrete crushing strain (3000 µƐ) [21].

3.6. Diagonal Displacement across Shear Cracks

The values of maximum crack displacement at peak load in segments 1, 2, and 3 are listed in

Table 7. Displacement across shear cracks at shear capacity.

Group	Specimen ID	Displacement across Shear Cracks (mm) *
		Segment 1	Segment 2	Segment 3
1	SB-R0-NS	0.00	0.00	3.37
	SB-R60-NS	0.00	0.00	5.13
	SB-R100-NS	0.00	0.00	3.79
2	SB-R0-S	0.00	3.16	0.01
	SB-R60-S	0.00	3.42	0.51
	SB-R100-S	0.00	0.00	0.54

* Crack displacement at shear capacity.

. It is noteworthy that the crack displacement values were not recorded when shear cracks happened outside the range of displacement transducers. The beams in Group 1 (without BFRP stirrups) exhibited maximum crack width in segment 3, since the major diagonal shear crack occurred in this segment. Additionally, the beams in Group 2 (with BFRP stirrups) experienced maximum crack width in segment 2, since the major shear crack happened in this segment. It is noteworthy that for beam SB-R100-S, the major diagonal shear crack appeared outside the range of the transducer. Specimens SB-R60-NS and SB-R100-NS with RCA exhibited a greater crack width at the shear capacity than those of SB-R0-NS with NA. The displacement across shear cracks at peak load for BFRP-reinforced beams was less than those of beam specimens without BFRP stirrups, verifying the confinement effect of BFRP stirrups.

The relationships between the shear load and displacement across shear cracks for the beams without and with BFRP stirrups are illustrated in Figure 11a,b, respectively. The beams in Group 1 exhibited a trilinear response (Figure 11a). The first part of the curve corresponds to the uncracked stage. The beams made with 60 and 100% RCA exhibited lower cracking load relative to that of the control beam (SB-R0-NS). Following the shear crack initiation, the beams exhibited a similar rate of increase in the crack width with the applied load up to a crack width of approximately 2.3 mm. Afterward, beams SB-R0-NS and SB-R60-NS exhibited a higher rate of increase in the shear crack with the applied load. However, the specimen SB-R100-NS displayed a plastic response. The beams reinforced with BFRP stirrups exhibited an almost bi-linear response. The shear cracking load of the beams made with 60% RCA was less than that of the respective control beam with NA (SB-R0-S). During the post-cracking stage, the shear crack width of specimens SB-R0-S and SB-R60-S continued to increase quasi-linearly. The presence of BFRP stirrups provided a confinement effect that restricted the development of the diagonal shear cracks relative to that which occurred in the beams without BFRP stirrups. These observations are consistent with findings reported by Al-Hamrani et al. [35], where the presence of BFRP stirrups was effective in reducing the width of the diagonal shear crack.

4. Analytical Procedure

The nominal shear capacity (V_n) of slender beams can be expressed in terms of Equation (1), where V_c indicates the concrete contribution to shear strength and V_s shows shear resistance offered by stirrups. The concrete’s contribution to shear strength is affected by the mechanical properties of the concrete mixture, a/d ratio, and flexural reinforcement ratio. Additionally, the shear contribution of stirrups is influenced by the type of material, the cross-sectional area, and the number of shear reinforcement bars intersected by shear cracks.

$${\mathrm{V}}_{\mathrm{n}}={\mathrm{V}}_{\mathrm{c}}+{\mathrm{V}}_{\mathrm{s}}$$

(1)

ACI 440.1R-15 [31] proposed an analytical approach to determine the concrete contribution to the shear capacity of beams reinforced with FRP bars as shown in Equation (2). It is worth mentioning that the ACI 440.1R-15 [31] equation does not account for the shear span to effective depth ratio of beams.

$${\mathrm{V}}_{\mathrm{c}}=\frac{2}{5}\mathrm{k}\sqrt{{\mathrm{f}’}_{\mathrm{c}}}\mathrm{bd}$$

(2)

where V_c = shear contribution by concrete; k = ratio of the depth of neutral axis to reinforcement depth; f′_c = cylinder compressive strength; b = width of the beam; and d = effective depth of the beam.

$$\mathrm{k}=\sqrt{{\left({\mathrm{n}}_{\mathrm{f}}{\mathsf{\rho }}_{\mathrm{f}}\right)}^2+2{\mathrm{n}}_{\mathrm{f}}{\mathsf{\rho }}_{\mathrm{f}}} -{\mathrm{n}}_{\mathrm{f}}{\mathsf{\rho }}_{\mathrm{f}}$$

(3)

where n_f = ratio of modulus of elasticity of FRP bars to modulus of elasticity of concrete and ${\mathsf{\rho }}_{\mathrm{f}}$ = flexural reinforcement ratio. ACI 440.1R-15 [31] recommends Equation (4) to determine the shear contribution of FRP stirrups.

$${\mathrm{V}}_{\mathrm{f}}=\frac{{\mathrm{A}}_{\mathrm{fv}}{\mathrm{f}}_{\mathrm{fv}}\mathrm{d}}{\mathrm{s}}$$

(4)

where V_f = shear contribution by FRP stirrups; A_fv = cross-sectional area of FRP stirrup; d = effective depth of beam; s = spacing between stirrups; f_fv = tensile strength of FRP taken as the minimum of design tensile strength f_fu, the strength of the bent portion of FRP stirrups f_fb, and stress corresponding to 0.004E_f.

$${\mathrm{f}}_{\mathrm{fb}}=[0.05 \left(\frac{{\mathrm{r}}_{\mathrm{b}}}{{\mathrm{d}}_{\mathrm{b}}}\right)+0.3]{\mathrm{f}}_{\mathrm{fu}}$$

(5)

where r_b = radius of the bent of FRP stirrup and d_b = diameter of FRP stirrup.

Jang et al. [64] proposed Equation (6) to determine the nominal shear capacity of FRP-reinforced concrete beams. The symbols b and d denote the width and effective depth of the beam, respectively. The value of constant β₁ can be determined from expressions. Since a/d ≥ 2.5 in the current study, Equation (8) was used to determine the value of β₁. It is noteworthy that Jang’s approach does not account for the shear contribution by stirrups in the shear span. The shear capacity contribution of BFRP stirrups was calculated using the ACI 440 1R-15 [31] approach in the current study.

$${\mathrm{V}}_{\mathrm{c}}={\mathsf{\beta }}_1\frac{1}{6}\sqrt{{\mathrm{f}’}_{\mathrm{c}}} \mathrm{bd}$$

(6)

$$\mathrm{For}\frac{\mathrm{a}}{\mathrm{d}}\le 2.5: {\mathsf{\beta }}_1=3.944+0.256\frac{{\mathrm{E}}_{\mathrm{f}}}{{\mathrm{E}}_{\mathrm{s}}}-1.472\frac{\mathrm{a}}{\mathrm{d}}+73.886{\mathsf{\rho }}_{\mathrm{f}}$$

(7)

$$\mathrm{For}\frac{\mathrm{a}}{\mathrm{d}}\ge 2.5: {\mathsf{\beta }}_1=0.716+0.466\frac{{\mathrm{E}}_{\mathrm{f}}}{{\mathrm{E}}_{\mathrm{s}}}-0.095\frac{\mathrm{a}}{\mathrm{d}}+32.101{\mathsf{\rho }}_{\mathrm{f}}$$

(8)

The results of analytical predictions are compared with those obtained from laboratory tests in

Table 8. Comparison of analytical predictions with experimental results.

Specimen	Vu,exp (kN)	ACI 440.1R-15 [31]		Jang et al. [64]
		Vpred (kN)	Vpred/Vu,exp	Vpred (kN)	Vpred/Vu,exp
SB-R0-NS	38.9	11.6	0.30	27.5	0.71
SB-R60-NS	36.9	11.9	0.32	24.9	0.67
SB-R100-NS	32.1	12.9	0.40	23.8	0.74
SB-R0-S	49.9	33.4	0.67	49.3	0.99
SB-R60-S	45.4	33.7	0.74	46.7	1.03
SB-R100-S	39.6	34.7	0.88	45.6	1.15
Average	-	-	0.55	-	0.88
St Dev	-	-	0.24	-	0.20
COV (%)	-	-	44	-	23

. The objective is to evaluate the validity of analytical models to predict the shear capacity of beams made with RCA. The analytical approach suggested by ACI 440.1R-15 [31] provided conservative shear capacity results of beams and such a trend was more obvious for beams without BFRP stirrups. The value of the ratio (V_pred/V_u,exp) of beams without and with BFRP stirrups was in the range of 0.30–0.40 and 0.67–0.88, respectively. On average, the value of the ratio V_pred/V_u,exp was 0.55 for tested beams in the current study. The ACI 440.1R-15 [31] approach does not account for a/d which is a crucial parameter. This could be the possible reason for the very conservative prediction results. Issa et al. [33] confirmed that the shear capacity of NA-based concrete beams with and without BFRP stirrups was conservatively predicted by the ACI analytical approach with an average V_pred/V_u,exp value of 0.60. Similarly, Al-Hamrian et al. [35] found that the ACI 440 1R-15 [31] approach conservatively predicted the shear capacity of NA-based concrete beams reinforced with BFRP bars and BFRP stirrups with an average V_pred/V_u,exp = 0.54. In the current study, the shear capacity of RCA-based beams was conservatively predicted by ACI 440.1R-15 [31], especially for beams without BFRP stirrups in the shear span. However, Younis et al. [37] reported that the ACI approach overestimated the shear capacity of the beam without stirrups and made with 100% RCA. This could be ascribed to an insignificant reduction (4%) in the compressive strength of concrete when NA was replaced by 100% RCA. Additionally, the negligible decrease in the compressive strength of 100% RCA-based concrete is attributable to an increased amount of cement and slag relative to that of the NA-based mixture.

The analytical approach suggested by Jang et al. [64] predicted the shear capacity of beams without stirrups conservatively. The average value of the ratio V_pred/V_u,exp was 0.88. There was a gradual reduction in the predicted shear capacity values for beams without and with stirrups. The value of the ratio (V_pred/V_u,exp) of beams without and with BFRP stirrups was in the range of 0.67–0.74 and 0.99–1.15, respectively. The analytical model accurately predicted the shear capacity of SB-R0-S and SB-R60-S. However, the analytical model overestimated the shear capacity of the beam with 100% RCA (SB-R100-S) by 15%. Overall, the Jang et al. [64] approach provided better results in shear capacity prediction compared to the ACI 440.1R-15 [31] approach.

The contribution of the concrete to the shear resistance in FRP-reinforced concrete beams is typically lower than that of similar beams with conventional steel reinforcing bars [31,32]. As such, it would be noteworthy to compare the contribution of concrete to the shear strength of BFRP-reinforced concrete beams tested in the present study to that of similar beams with conventional steel bars at the same flexural reinforcement ratio. Equation (9), derived by Zsutty [65], is adopted to predict the contribution of concrete to the shear resistance of concrete beams with conventional steel bars. This equation accounts for the effect of flexural steel reinforcement ratio (${\mathsf{\rho }}_{\mathrm{s}}$) and a/d on the contribution of concrete to the shear resistance of steel-reinforced concrete beams. It is also included in standard textbooks [62]. The analytical predictions provided in

Table 9. Comparison of predicted concrete contribution to shear resistance of beams with different types of flexural reinforcement.

Aggregate Type	Predicted Contribution of Concrete to Shear Resistance		Ratio
	BFRP-Reinforced Beams 1 Vc (kN)	Steel-Reinforced Beams 2 Vc,s (kN)	Vc/Vc,s
Natural Aggregate	27.5	35.8	0.77
60% RCA	24.9	33.5	0.74
100% RCA	23.8	32.5	0.73

¹ The prediction is based on Jang et al. [64] analytical formula. ² The prediction is based on Zsutty [65] analytical formula.

verify the reduced contribution of concrete to the shear resistance of BFRP-reinforced beams relative to that of their steel-reinforced counterparts. The ratio of V_c/V_c,s was on average 0.75. Such a reduction in the contribution of concrete to the shear resistance can be attributed to the increased crack width of BFRP-reinforced concrete beams relative to that which could happen in steel-reinforced beams [31,32]. It is noteworthy that it is not possible to compare the contribution of stirrups to the shear resistance of beams with different types of reinforcement because conventional steel could have different values of yield strength in the range of 280 to 690 MPa. As such, the comparison was limited to the contribution of concrete to the shear resistance.

$${\mathrm{V}}_{\mathrm{c},\mathrm{s}}=2.2{\left({\mathrm{f}’}_{\mathrm{c}}{\mathsf{\rho }}_{\mathrm{s}}\frac{\mathrm{d}}{\mathrm{a}}\right)}^{1/3}\mathrm{bd}$$

(9)

5. Conclusions

This research investigated the shear response of slender beams made with RCA and reinforced with BFRP bars. The effect of partial and full replacement of NA with RCA on the shear response of slender beams with and without BFRP stirrups in the shear span was studied. The usefulness of BFRP stirrups to improve the shear capacity of RCA-based slender beams was revealed. A comparison study has been made to investigate the validity and accuracy of two different analytical models to predict the shear capacity of tested slender beams. The main conclusions of the study are the following:

• The shear cracking load of the beams made with RCA without BFRP stirrups was 43% of that of the corresponding control beam made with NA. For the beams without BFRP stirrups, increasing the RCA replacement percentage from 60 to 100% had no effect on the shear cracking load.
• In the presence of BFRP stirrups, the shear cracking load of the beams with RCA was on average 68% of that of the corresponding control beam made with NA. Increasing the RCA replacement percentage from 60 to 100% reduced the shear cracking load of the beams with BFRP stirrups by 23%.
• The RCA replacement of 60% reduced the shear capacity of BFRP-reinforced beams by 5–9% in comparison to that of their respective counterpart beams made with NA. The reduction in the shear capacity increased to 17–21% for the beams with 100% RCA.
• The contribution of BFRP stirrups to the shear capacity of the beams tended to decrease with an increase in the RCA replacement percentage. The contribution of BFRP stirrups to the shear capacity of the beam made with NA was 11 kN. The respective values for the beams with 60 and 100% RCA were 8.5 and 7.5 kN, respectively.
• The width of the major shear crack at a given load increased due to the partial or full replacement of NA by RCA. The BFRP stirrups were effective in delaying the appearance of the major shear crack in the shear span and restricting its rate of growth.
• The analytical approach recommended by ACI 440.1R-15 [31] significantly underestimated the shear capacity of the beams without and with BFRP stirrups by up to 70 and 33%, respectively. The analytical approach by Jang et al. [64] provided more accurate prediction for the shear capacity of the tested beams. The percentage difference between the experimental shear capacity and that predicted by the Jang et al. [64] analytical approach was in the range of −26 to −33% for the beams without BFRP stirrups. The respective margin of error for the predicted shear capacity of the beams with BFRP stirrups was −1 to +15%.

Valuable information on the influence of using RCA rather than NA on the shear behavior of BFRP-reinforced concrete beams with and without stirrups is furnished in this study. More research is needed to examine the influence of varying the BFRP shear reinforcement ratio and RCA replacement percentages on the shear behavior of BFRP-reinforced concrete beams. Future research shall consider the development of numerical simulation models capable of predicting the shear behavior of RCA-based concrete beams reinforced with BFRP with a wider range of parameters. Verified numerical simulation models can be subsequently used to introduce generalized analytical formulations for the engineering community.