# Bond performance of basalt fiber-reinforced polymer bars in conventional Portland cement concrete: a relative comparison with steel rebar using the hinged beam approach

Muhammet Seis and Ahmet Beycioğlu

# Abstract

This paper reports the results of an experimental investigation carried out to evaluate the bond stress behavior of basalt fiber-reinforced polymer (BFRP) bar and steel rebar (SR) in conventional C30-type concrete. The experimental program was conducted on testing 16 hinged beam specimens that were prepared according to BS 4449:2005+A2:2009 standard. Conventional C30-type concrete was used to produce hinged beams. In beam tests, bar diameter and embedment length were used as experimental variables. The bond performances of BFRP bar and SR were compared with the help of the load-slip results of hinged beam bending tests conducted on produced beam samples after 28 days of curing. Results showed that much higher bond stress values were obtained from BFRP bars compared to SR for both 12 and 8 mm bar diameters. Besides, the maximum bond stress values decreased with increasing bar diameter and embedment length for both BFRP bar and SR.

## 1 Introduction

Reinforced concrete (RC) structures, the most important and popular structural type all over the world, have been tremendously constructed worldwide in the past few decades (e.g. super-dams, long-span bridges, and skyscrapers) [1]. Steel rebars (SR) are widely used reinforcing bars in RC structures due to their efficiency and economic benefit. These bars are ductile and strong, which makes them appropriate for reinforcing concrete [2]. Unfortunately, compared with other factors, including freeze-thaw and the action of physical and chemical ambient, corrosion of reinforcing steel has been one of the most important factors that severely degrade the durability of RC structures [1, 3]. The corrosion of SR in RC structures leads to cracking and spalling of concrete, resulting in costly maintenance and repair. An innovative approach to solve the problems caused by SR can be provided using fiber-reinforced polymer (FRP) as an alternative [4].

Over the past two decades, laboratory tests have demonstrated that FRP bars can be used successfully and practically as internal reinforcement in concrete structures [5].

Nowadays, FRPs have been widely used in RC structures due to its superior material properties, such as high tensile strength, excellent electrochemical corrosion resistance, and cost effective fabrication, over traditional SRs [6]. Therefore, in the recent years, FRP bars have been introduced as a competent alternative to traditional reinforcing SRs for different RC structures subjected to severe environmental conditions, such as waste-water treatment and chemical plants, floating decks, sea walls, and water structures [7, 8]. An FRP bar consists of two main phases. These phases are matrices and fibers. The matrices commonly used are thermoset polymeric resins, and reinforcements are glass, carbon, aramid, and basalt fibers. An FRP bar is anisotropic and can be manufactured using different techniques, such as pultrusion, braiding, and weaving. Glass, carbon, aramid, and basalt fibers are commonly used to produce FRP bars. Glass fibers offer an economical balance between cost and specific strength properties; this makes them preferable to carbon and aramid in the applications of most RC structures [5].

Recently, basalt FRP (BFRP) bars have emerged as a promising alternative to conventional glass FRP (GFRP) materials. Few studies conducted to date involve BFRP bars. Altalmas et al. investigated the bond durability of sand-coated BFRP bars under direct tensile load after being exposed to accelerated conditioning environments (acid, saline, and alkaline). Results showed that conditioning reduced the bond strength of the BFRP bars by 14%–25% of their initial strength [9]. El Refai et al. investigated the effect of different accelerated environments on the bond stress-slip response, adhesion to concrete, and bond strength of two types of BFRP bars and one type of GFRP bar. The test results of the study showed that moister environments caused enhanced adhesion at the early loading stages for all FRP bars. At later stages, they found that such environments had a detrimental effect on the bond strength depending on the moisture absorption of the bar material. Another finding of the study is that the exposure to elevated temperatures caused insignificant effects in the bond strength of all tested specimens [10]. In another paper, El Refai et al. presented the test results of a study on the bond behavior of BFRP bars. In the experiments, direct pullout tests were conducted on the cylindrical samples. The results of the study demonstrate the promise of using the BFRP bars as an alternative to the GFRP bars in reinforcing concrete elements [11].

According to Bi et al. the experimental results indicate that the average bonding strength decreased, whereas the bonding length increased, and it became weaker, whereas the diameter of the BFRP was bigger [12]. In Mahroug et al.’s study, the test results and code predictions of four continuously supported and two simply supported concrete slabs reinforced with BFRP bars were presented. One continuously supported steel RC slab was also tested for comparison purposes. Different combinations of under and over BFRP reinforcement at the top and bottom layers of slabs were investigated. The continuously supported BFRP RC slabs exhibited larger deflections and wider cracks than the counterpart reinforced with steel. Furthermore, the over reinforced BFRP RC slab at the top and bottom layers showed the highest load capacity and the least deflection of all BFRP slabs tested [13].

In this study, the bond performance of BFRP bars in conventional concrete was investigated using the hinged beam method.

## 2 Materials, hinged beam method, and sample preparations

### 2.1 Materials

For this research, a C30-type conventional concrete mix design was prepared and used to build the hinged beam specimens. The materials selected to produce concrete were readily available in the market. The cement used as binder in concrete production was a normal CEM I 42.5R-type Portland cement (PC), conforming to EN 197-1 [14]. On the contrary, low-lime fly ash [LLFA; fly ash having lime content of 1.81% (CaO<10%)] was employed to increase the flowability of concrete. The chemical and physical properties of PC and LLFA are given in Table 1.

### Table 1

Chemical composition and physical properties of the PC and LLFA.

Chemical composition (%) PC LLFA
SiO2 18.95 58.56
Fe2O3 4.07 6.51
TiO2 1.21
Al2O3 5.32 23.39
CaO 64.72 1.81
MgO 1.35 2.02
Na2O 0.16 0.53
K2O 0.51 4.13
SO3 2.9 0.0013
P2O5 0.14
Loss on ignition 3.83 1.25
Physical properties
Specific gravity 3.18 2.09

Fine and coarse aggregates of crushed limestone were used to produce the concrete samples. The maximum aggregate size was 12.5 mm (fine 0–4.75 mm and coarse 4.75–12.5 mm). The coarse and fine aggregates had specific gravities of 2.72 and 2.70, and water absorptions of aggregates were 0.55% and 2.3%, respectively. To find the mechanical properties (compressive and splitting tensile strength) of concrete, cube (150×150×150 mm) and cylinder (150 mm diameter and 300 mm height) samples were prepared.

The mix design of concrete is given in Table 2. To ensure that concrete would fit into molds easily, an S4 slump-class (160–210 mm) mix design was prepared as specified in the BS EN 12350-2 standard [15]. LLFA was used in the mixtures to improve the flowability of concrete to produce S4 consistency class. LLFA can improve the flowability of concrete by the help of its smooth, spherical particles. The scanning electron microscopy image (SEM) given in Figure 1 shows the particle structure of LLFA.

### Table 2

Materials mixed in 1 m3 concrete.

Materials Volume (dm³) Specific weight Quantity (kg/m³)
Cement 95 3.17 300
LLFA 23 2.2 50
Water 196 1 196
Crushed sand (0–5) 403 2.66 1073
Crushed stone (5–15) 269 2.68 721
Superplasticizer 4.25 1.07 4.55

### Figure 1:

SEM image of LLFA used in this research.

The surface forms of BFRP bar and SR used in this research are shown in Figure 2. The physical, mechanical, and thermal properties of BFRP bar, provided by the manufacturer, are presented in Table 3.

### Figure 2:

Surface forms of BFRP bar and SR.

### Table 3

Physical, mechanical, and thermal properties of BFRP bar provided by the manufacturer.

Properties Values
Tensile strength (MPa) 1100
Modulus of elasticity (MPa) 70,000
Density (g/cm3) 1.9
Thermal conductivity coefficient (W/mK) 0.35–0.59
Linear expansion coefficient 1
Elongation (%) 2.2
Corrosion resistance No rusting

### 2.2 Hinged beam method and sample preparations

The hinged beam test is one of the bending tests for determining the bond behavior of concrete and reinforcement. The principle of the test method is to load a test beam by simple flexure until a complete bond failure of the reinforcing steel occurs in one of the half-beams or until the reinforcing steel itself fails. During loading, the slip of the two ends of the reinforcing bar is measured. The beam used for the test consists of two parallelepiped RC blocks interconnected at the bottom by the reinforcing bar, of which the bond is to be tested, and at the top by a steel hinge [16]. Steel hinge is placed in the middle of the beam to ascertain the tensile forces more adequately. Figure 3 shows the test set-up used in this study.

### Figure 3:

Hinged beam test set-up used in this study [21].

Hinged beam enables to measure the slips of the reinforcing bars by loading the beam at midpoint. This test method was chosen for this study because beam test certainly provides a better estimation of the bond strength due to its similarities with the real structural elements (especially flexural members) [17]. Figure 3 shows the details of the experimental test set-up used in this research. During the application of the test, the load cell and the potentiometric rulers were used to measure the vertically applied loads and slips by loading the beam at midpoint as seen in Figure 4. Here, it should be expressed that it is important to complete the experiments without shear cracks. Stirrups were used to ensure the shear resistance of beam specimens. The reinforcement arrangement of the hinged beams blocks is given in Figure 5.

### Figure 4:

Details of the experimental test set-up used in this research.

### Figure 5:

Reinforcement arrangement of the hinged beam blocks used in this research [20].

In the production of hinged beam samples, plastic sheaths were placed in appropriate places of reinforcements using hot silicon to make embedment lengths of 10 and 20 Φ (Figure 6A). The reinforcements having limited embedment lengths were passed through stirrups and placed into special molds together with stirrups (Figure 6A–C). The hinged beam samples ready for bending test after curing are shown in Figure 7.

### Figure 6:

Preparation phases of the hinged beam samples: (A) limiting process of the embedment length, (B) prepared stirrup reinforcement, and (C) beam samples ready for the concrete casting.

### Figure 7:

A view of hinged beams prepared for the bond test.

## 3 Results and discussions

The slump test of fresh concrete tests was carried out to determine the conformity of C30-type concrete prepared for the production of hinged beams to the requirements of the S4 slump class. Additionally, the compressive strength, splitting tensile strength, and water absorption tests were carried out to determine the mechanical and physical characteristics of hardened concrete. The mean results of all tests applied on concrete are summarized in Table 4.

### Table 4

Properties of C30-type conventional concrete produced in this research.

Concrete properties Results
Slump (mm) 190
fccube (MPa), 7 days 25.91
fccube (MPa), 28 days 37.9
Splitting tensile strength (MPa), 28 days 3.84
Water absorption (%) 3.41

As can be seen in Table 4, the average slump value was found to be 190 mm in the produced concretes. This result is within the values (160–210 mm) specified in the BS EN 12350-2 standard for the S4 slump class. Moreover, it was found that the 28 days’ compressive strength of the cube sample of the concrete is 37.9 MPa, the splitting tensile strength of the cylinder sample of the concrete is 3.84 MPa, and the water absorption of concrete is 3.41% (37.9 MPa cube sample means 30 MPa cylinder sample (C30) according to the BS EN 206-1 [18]).

The yield stress, ultimate stress, and rupture stress of SR were found to be 490, 570, and 478 MPa, respectively, by uniaxial tensile testing. For BFRP bar, rupture stress was found to be 1021 MPa. Because BFRP bar is a brittle material, yield stress could not obtained. These results mean that the tensile strength of the basalt reinforcement, which was found experimentally in the scope of this research, is 2.08 times greater than the yield strength of SR and 1.79 times greater than the tensile strength of SR.

In the hinged beam test, to determine the bond stress of reinforcement bars using beam specimens, the loads in the reinforcement bars were first determined. Based on the geometry of the beam specimens and the locations of the applied loads and supports (Figure 3), the loads in the reinforcement bars (P) were determined by the following equation:

(1)$P=1.25F (1)$

After determining the loads in the reinforcement bars, the bond stress between reinforcement bars and concrete were calculated. According to the RILEM Technical Recommendation, the bond stress (τu) between reinforcement bars and concrete can be calculated using Equation (2). In this equation, P is the applied ultimate load (UL; N), d is the bar diameter (mm), and ld is the embedment length (mm) [19].

(2)$τu=P/(πdld) (2)$

In this study, a coding style was used to name hinged beam specimens. In the coding, beams were coded in the form of B-X-Y-Z or S-X-Y-Z, where B represents basalt, S represents steel, “X” represents embedment length (10 means 10Ø and 20 means 20Ø), “Y” represents bar diameters (08 means Ø8 and 12 means Ø12), and “Z” represents the sample number.

All results of bending tests found in this research are given in Table 5. These results include the UL, ultimate stress on rebar (USR), ultimate bond stress (τu), slip at UL (SUL), average slip at UL (ASUL), and failure mode (FM). Normally, two slip values are taken from each of the hinged beam samples by loading the beams (from potentiometric rulers in the right and left sides of the beam). The values given in Table 5 are the maximum slip values of each loaded beam.

### Table 5

Bond test results under hinged beam condition.

Sample code UL (kN) USR (MPa) τu (MPa) τu average SUL (mm) ASUL (mm) FM
B-20-12-1 113.92 1007.77 12.60 12.75 0.129042 0.119665 Bar rupture
B-20-12-2 116.65 1031.96 12.90 0.110288 Bar rupture
B-20-08-1 56.35 1121.56 14.02 13.74 0.833958 0.889427 Bar rupture
B-20-08-2 54.06 1075.88 13.45 0.9448952 Bar rupture
S-20-12-1 64.8 573.27 7.17 7.31 0.249281 0.250858 Bar rupture
S-20-12-2 67.25 594.9 7.44 0.252435 Bar rupture
S-20-08-1 31.17 620.49 7.76 7.75 0.548395 0.597364 Bar rupture
S-20-08-2 31.11 619.23 7.74 0.646332 Bar rupture
B-10-12-1 83.84 741.7 18.54 18.19 2.74662 2.680145 Pull-out
B-10-12-2 80.62 713.17 17.83 2.61367 Pull-out
B-10-08-1 44.92 894.02 22.35 22.15 2.91334 2.740845 Pull-out
B-10-08-2 44.11 877.92 21.95 2.56835 Pull-out
S-10-12-1 60.13 531.94 13.30 13.13 1.68856 1.50903 Pull-out
S-10-12-2 58.59 518.34 12.96 1.3295 Pull-out
S-10-08-1 27.49 547.27 13.68 13.51 1.19529 1.518475 Pull-out
S-10-08-2 26.81 533.71 13.34 1.84166 Pull-out

According to the results in Table 5, when the average τu values obtained from samples with 20 Φ embedment length and 12 mm bar diameter were examined, the average τu values obtained in SR as 7.31 MPa (S-20-12-average) increased by 73.9% in BFRP reinforcing bar up to 12.75 MPa (B-20-12-average). Similarly, in samples with 20 Φ embedment length and 8 mm bar diameter, the average τu value obtained in SR as 7.75 MPa (S-20-08-average) increased by 77.3% in BFRP reinforcing bar up to 13.74 MPa (B-20-08-average). In results obtained from samples with 10 Φ embedment length, similar to samples with 20 Φ embedment length, the average τu values of the BFRP reinforcing bar were found to be higher than SR. (The B-10-12-average value was 38.5% higher than the S-10-12-average value and the B-10-08 average value was 63% higher than the S-10-08-average value.) One of the important results in Table 5 was that the FM in all series with 20 Φ embedment length were in bar rupture type and that all ASUL values were below 1 mm. These results show that both SR and BFRP reinforcing bars do not lose adherence until a break-in series with 20 Φ embedment length. However, as mentioned above, much higher τu values were obtained from BFRP reinforcing bars compared to SR for both 12 and 8 mm bar diameters.

According to the obtained findings, if a general evaluation is made about the effect of bar diameter on bond stress, it can be said that bond stress values decrease by increasing bar diameter. A similar conclusion has already been demonstrated in previous studies [17, 2022].

As seen in Table 5, in all cases, the slip at τu is decreasing with increasing bar diameter, as already demonstrated in previous studies of Pop et al. [17] and Sonebi and Bartos [23]. The evolution of the maximum average bond stress with the decrease in embedment length is shown in Figure 8. As observed in Figure 8, if BFRP bar and SR were used both in 8 and 12 mm bar diameters, bond stress values decreased by increasing embedment length. In addition to the results given in Table 5, the load-slippage figures were created using the obtained data from bending tests (Figures 912). In these figures, the load required to achieve the yield stress of conventional steel reinforcement (LRYS-S-Conventional), the load required to achieve the yield stress of steel reinforcement used in this research (LRYS-This research), the load required to achieve the tensile stress of steel reinforcement used in this research (LRTS-This research), and the load required to achieve the tensile stress of basalt reinforcement used in this research [LRTS-B-VM (for Value obtained from Manufacturer)] were given to help clearly compare the bond performances of BFRP bar and SR.

### Figure 8:

Maximum average bond stress values corresponding to the embedment length.

### Figure 9:

Load and free end slip relationship of 12 mm reinforcement bars and concrete (20 Φ embedment length).

### Figure 10:

Load and free end slip relationship of 8 mm reinforcement bars and concrete (20 Φ embedment length).

### Figure 11:

Load and free end slip relationship of 12 mm reinforcement bars and concrete (10 Φ embedment length).

### Figure 12:

Load and free end slip relationship of 8 mm reinforcement bars and concrete (10 Φ embedment length).

Using the results shown in Figures 912, the slip values of BFRP bars corresponding to loads required to reach the yield and tensile strength of SR used in this study are given in Table 6 as briefly to interpret the bond performance of BFRP bar.

### Table 6

Slip values of BFRP bars at loads required to reach the yield and tensile strength of SR.

Sample code Slip values of BFRP bars when the yield strength of steel is reached (mm) Slip values of BFRP bars when the tensile strength of steel is reached (mm)
B-20-12-1 0 0
B-20-12-2 0 0
B-20-08-1 0 0
B-20-08-2 0 0
S-20-12-1 0.111799 0.24777
S-20-12-2 0.159184 0.234537
S-20-08-1 0.186332 0.300535
S-20-08-2 0.045924 0.446332
B-10-12-1 0.587755 0.916483
B-10-12-2 0.539354 1.13158
B-10-08-1 0.597174 0.83642
B-10-08-2 0.397339 0.587699
S-10-12-1 0.602615 PO before the tensile strength of steel is reached
S-10-12-2 1.49115 PO before the tensile strength of steel is reached
S-10-08-1 0.527472 PO before the tensile strength of steel is reached
S-10-08-2 0.317267 PO before the tensile strength of steel is reached

The most significant result in Table 6 is that there are not any slip values obtained from BFRP bars at the loads corresponding to the yield strength and tensile strength of SR. Another significant result in Table 6 is that, in the series where the embedment length is 10 Φ, although BFRP bar preserves its adherence until the load corresponding to the tensile strength of SR, it loses its adherence before reaching the tensile strength in the series where a reinforcing bar with a diameter of both 8 and 12 mm is used and demonstrates PO-type adherence failure.

## 4 Conclusions

The bond stress behavior of BFRP bar and SR in conventional C30-type concrete has been investigated experimentally on hinged beam test specimens. Based on the obtained results, the following conclusions can be written:

• The tensile strength of BFRP bar is 2.08 times greater than the yield strength of SR and 1.79 times greater than the tensile strength of SR.

• The maximum bond stress values have a tendency to decrease when the embedment length increased for both BFRP bar and SR.

• The maximum bond stress values decrease with increasing bar diameter for both BFRP bar and SR.

• In a series with 20 Φ embedment length and 12 mm bar diameter, the maximum bond stress of BFRP bar was 73.9% higher compared to SR, and when bar diameter was 8 mm, the maximum bond stress of BFRP bar was 77.3% higher compared to SR.

• Considering the results, in case of necessity, if BFRP bar is used in a project, the embedment length requirement can be reduced in comparison to the SR.

• When the load values corresponding to the yield strength of SR with a diameter of 12 mm were reached, the slip value occurring on BFRP bar with a diameter of 8 mm was obtained as 0.6174232 mm. When the load values corresponding to 72% of the yield strength of SR with a diameter of 12 mm were reached, no slip values occurred on BFRP bar with a diameter of 8 mm. These results show that the studies about the usability of BFRP bars with a smaller diameter than SR in projects containing C30-type or more resistant concrete and 20 Φ embedment length will be useful. Besides, this subject can be evaluated in future studies in terms of cost.

Corresponding author: Ahmet Beycioğlu, Technology Faculty Civil Engineering Department, Düzce University, 81620 Central-Duzce, Turkey, e-mail:

### References

[1] Qiao G, Hong Y, Ou J. Measurement 2015, 67, 78–83.10.1016/j.measurement.2014.12.018Search in Google Scholar

[2] Won J, Park C, Kim H, Lee S, Jang C. Compos. Pt. B Eng. 2008, 39, 747–755.10.1016/j.compositesb.2007.11.005Search in Google Scholar

[3] Qiao G, Hong Y, Ou J, Guan X. Measurement 2015, 67, 84–91.10.1016/j.measurement.2014.12.048Search in Google Scholar

[4] Metwally IM. HBRC J. In press. http://dx.doi.org/10.1016/j.hbrcj.2015.02.006.10.1016/j.hbrcj.2015.02.006Search in Google Scholar

[5] Nanni A, De Luca A, Zadeh HJ. Reinforced Concrete with FRP Bars: Mechanics and Design. CRC Press, Taylor & Francis Group, 2014.10.1201/b16669Search in Google Scholar

[6] Lin X, Zhang YX. Compos. Struct. 2014, 107, 131–141.10.1016/j.compstruct.2013.07.037Search in Google Scholar

[7] Islam S, Afefy HM, Sennah K, Azimi H. Constr. Build. Mater. 2015, 83, 283–298.10.1016/j.conbuildmat.2015.03.025Search in Google Scholar

[8] Alves J, El-Ragaby A, El-Salakawy E. J. Compos. Constr. 2011, 15, 249–62.10.1061/(ASCE)CC.1943-5614.0000161Search in Google Scholar

[9] Altalmas A, El Refai A, Abed F. Constr. Build. Mater. 2015, 81, 162–171.10.1016/j.conbuildmat.2015.02.036Search in Google Scholar

[10] El Refai A, Abed F, Altalmas A. J. Compos. Constr. 2015, 19, 5, 04014078-1-11.Search in Google Scholar

[11] El Refai A, Ammar M, Masmoudi R. J. Compos. Constr. 2015, 19, 3, 04014050-1-12.Search in Google Scholar

[12] Bi Q, Wang Q, Wang H. Adv. Mater. Res. 2011, 163, 1251–1256.Search in Google Scholar

[13] Mahroug MEM, Ashour AF, Lam D. Compos. Struct. 2014, 107, 664–674.10.1016/j.compstruct.2013.08.029Search in Google Scholar

[14] British Standards Institution. BS EN 197-1:2000, Cement. Part 1: Composition, Specifications and Conformity Criteria for Common Cements.Search in Google Scholar

[15] British Standards Institution. BS EN 12350-2:2009, Testing Fresh Concrete. Slump Test.Search in Google Scholar

[16] British Standards Institution. BS 4449:2005+A2:2009.Search in Google Scholar

[17] Pop I, De Schutter G, Desnerck P, Onet T. Constr. Build. Mater. 2013, 41, 824–833.10.1016/j.conbuildmat.2012.12.029Search in Google Scholar

[18] British Standards Institution. BS EN 206:2013.Search in Google Scholar

[19] RILEM. Technical Recommendations for the Testing and Use of Construction Materials. RILEM, 1992.Search in Google Scholar

[20] Sonebi M, Davidson R, Cleland D. 12DBMC. International Conference on Durability of Building Materials and Components, Portugal, 2011.Search in Google Scholar

[21] Desnerck P, De Schutter G, Taerwe L. Mater. Struct. 2010, 43, 53–62.10.1617/s11527-010-9596-6Search in Google Scholar

[22] Bamonte P, Coronelli D, Gambarova PG. Fracture Mechanics of Concrete and Concrete Structures, FraMCos 5, Vail, Colorado, USA, 2004.Search in Google Scholar

[23] Sonebi M, Bartos PJM. In Proceedings of the RILEM Symposium on Self-Compacting Concrete, Skarendahl A, Petersson O, Eds., Stockholm, 1999, 275–289.Search in Google Scholar