Compressive behavior of circular and square concrete column externally confined by different types of basalt fiber

Abstract

This article studies the compressive behavior of concrete columns confined by different basalt fiber–reinforced polymers. A total of 30 columns were divided into 10 groups according to section shapes (circular and square), basalt fiber–reinforced polymer types (unidirectional basalt fiber–reinforced polymer, bidirectional basalt fiber–reinforced polymer, and hybrid basalt fiber–reinforced polymer/carbon fiber–reinforced polymers), and number of layers (0, 1, and 2). The test results showed that the compressive strengths of confined specimens increased by 20%–71% for circular columns and by 23%–41% for square columns. Similarly, the ultimate strains improved by 49%–296% for circular specimens and by 45%–145% for square specimens. The two-layer basalt fiber–reinforced polymer jacket had the best confinement effect, whereas the confining effect of bidirectional basalt fiber–reinforced polymer wrapping was relatively lower than that of unidirectional basalt fiber–reinforced polymer wrapping. Moreover, both the strength and ultimate strain of confined concrete improved with increasing number of basalt fiber–reinforced polymer layers. Finite element numerical models were also developed and verified by experimental results, and then the stress distributions of basalt fiber–reinforced polymer jackets and cross-sectional concrete were presented. Based on the test results and experimental data from several existing studies, modified strength and ultimate strain models were further developed for basalt fiber–reinforced polymer-confined circular and square columns.

Keywords

BFRP compressive behavior confinement finite element method strength model

Introduction

Fiber-reinforced polymer (FRP) sheets have been widely used for strengthening concrete structures due to their light weight, high tensile properties, good corrosion resistance, and construction convenience. Numerous studies on this topic have been reported in recent decades, but most have investigated columns externally with conventional FRPs including carbon FRP (CFRP), aramid FRP (AFRP), and glass FRP (GFRP) (Aslani and Kohnehpooshi, 2018; Baky et al., 2010; Lam and Teng, 2004; Rashid and Aboutaha, 2014; Zhang et al., 2019). Recently, basalt FRP (BFRP) has received increasing attention. Basalt fibers are made from natural melting basalt rock, and the manufacturing process of basalt fibers is environmentally friendly and saves energy at low cost. The price of basalt fibers is approximately 60%–80% less than that of carbon fibers. Furthermore, basalt fibers also possess higher tensile strength than glass fibers and larger ultimate strain than carbon fibers. Thus, BFRPs are a good alternative to CFRPs and GFRPs in the field of strengthening structures (Dhand et al., 2015; Monaldo et al., 2019).

Ludovico et al. (2010) studied the compressive behavior of concrete cylinders confined by bidirectional BFRP laminates compared with that of uniaxial GFRP laminates. The results showed that the former had superior compressive strength of concrete, whereas the latter had better ultimate axial strain. Campione et al. (2015) investigated a compression test of concrete cylinders confined by bidirectional BFRP and unidirectional CFRP sheets. The stress–strain curves of BFRP-confined concrete indicated strain-softening behavior, while the curves of CFRP-wrapped concrete showed strain-hardening behavior. Wei et al. (2018) reported that the CFRP-confined concrete cylinders under compression tests exhibited larger post-peak stiffness than the BFRP-confined columns with the same layers, whereas the BFRP showed higher ultimate strain compared to the CFRP for confined specimens with similar confining pressure. Li et al. (2017) focused on the compression behavior of concrete cylinders wrapped with different FRP types. BFRP-confined specimens exhibited the highest hoop strain of FRP rupture in comparison with those achieved by GFRP and CFRP, but the strength and ultimate strain were only at intermediate levels with the same number of FRP layers. For hybrid FRP types, the confining efficiency of BFRP/CFRP was significantly better than that of BFRP/GFRP, which was mainly affected by confinement stiffness. Less research was conducted on confined rectangular columns compared with cylinders. Deng et al. (2012) compared the compressive behavior of BFRP-confined circular concrete specimens and non-circular specimens. Moreover, Liu et al. (2015) found that the effectiveness of BFRP confinement in non-circular specimens increased with decreasing sectional aspect ratio, and Suon et al. (2019) also found that this effectiveness increased with corner radius.

Most of the stress and strain expressions of confined circular concrete columns were based on the model by Mander et al. (1988) (Ozbakkaloglu et al., 2013; Wu and Wei, 2015). For non-circular columns, modified models varied in format considering different shape factors (ACI 440 2R-08, 2002; Jing and Cao, 2005; Lam and Teng, 2003; Wu et al., 2007). However, the models mentioned above were based on the existing experimental data of CFRP-, AFRP-, and GFRP-confined concrete specimens, and few studies have focused on BFRP confinement.

Regression analysis was commonly used to modify the expressions of BFRP-confined cylinders based on the model (Lam and Teng, 2004; Teng et al., 2009) with tested data (Campione et al., 2015; Ludovico et al., 2010; Ma et al., 2018; Xia et al., 2014). There were still controversies about the BFRP confinement effect. Xia et al. (2014) pointed out that the strength confinement coefficient $k_{1}$ of BFRP was calculated to be 4.4, which was larger than the value of 3.3 of CFRP (Lam and Teng, 2004). But Ma et al. (2018) indicated this value was 2.8, which was less than that of CFRP. Then the modified compressive strength expressions of non-circular columns were discussed (Deng et al., 2012; Liu et al., 2015). But an obvious limitation of these analyses was that data from other experiments were not included.

This study emphasizes the compressive behavior of plain concrete columns externally confined with BFRP fabrics. The column section shapes, BFRP types, and number of layers are investigated by experiments and numerical study. Based on existing test data, modified stress–strain models for BFRP-confined concrete are proposed. This study can provide reliable support for the application of BFRPs as strengthening materials.

Experimental program

Specimen design

An axial compressive test was presented. A total of 30 concrete columns with a height of 515 mm were manufactured, in which 15 specimens had circular cross section with a diameter of 112.8 mm and 15 specimens had square section with a length of 100 mm and a corner radius of 20 mm. Three considered parameters were the section shapes (circular and square), BFRP types (unidirectional BFRP, bidirectional BFRP, and hybrid BFRP/CFRP), and number of layers (0, 1, and 2). Then 10 series of specimens are shown in Table 1. For each series, three identical samples were prepared.

Table 1.

Test specimens.

Group	Series ID	Cross section	Dimensions (mm)	FRP types	Number of FRP layers
C	C-0	Circular	112.8 × 515	–	0
	C-B1			BFRP	1
	C-B2			BFRP	2
	C-B1C1			BFRP/CFRP	2
	C-2B1			BFRP (bidirectional)	1
S	S-0	Square	100.0 × 515	–	0
	S-B1			BFRP	1
	S-B2			BFRP	2
	S-B1C1			BFRP/CFRP	2
	S-2B1			BFRP (bidirectional)	1

FRP: fiber-reinforced polymer; BFRP: basalt fiber–reinforced polymer; CFRP: carbon fiber–reinforced polymer.

The FRP types without a specification in parentheses represent unidirectional FRP.

Identity tags such as C-B2 and S-2B1 were used for the specimens. The first letter C or S before the dash denoted the circular or square cross section, respectively; the second letter B or C after the dash distinguished BFRP or CFRP, respectively. The numeral 2 before the second letter B stood for bidirectional BFRP, while the default for unidirectional BFRP. The numeral following the second letter B represented the number of FRP layers.

Material properties

The concrete mix composition by weight was 0.41:1:1.59:2.49 for water, cement, sand, and coarse aggregate. The maximum size of the coarse aggregate was 30 mm. The average compressive strength of cubes with a side length of 150 mm was 44.4 MPa at 28 days. The FRP fabrics used in this experimental program have the same areal mass of 300 g/m², and the tensile properties of the FRPs are summarized in Table 2 from the flat coupon test according to ASTM D3039/D3039M-08 (2008). A two-component epoxy resin with a mix proportion of 3:1 by weight was used as the matrix for impregnation of the fabric. The tensile strength, tensile elastic modulus, ultimate tensile strain, and normal tensile bond strength of the interface between the concrete and FRP was 44 MPa, 2771 MPa, 1.7%, and 5.7 MPa, respectively, which were provided by the manufacturer. For FRP wrapping, each jacket sheet had an additional overlap of 100 mm at the end to prevent debonding failure.

Table 2.

Mechanical properties of FRP fabrics.

Type of fibers	Tensile strength $f_{frp}$ (MPa)	Elastic modulus $E_{frp}$ (GPa)	Elongation $ε_{frp}$ (%)	Thicknesst (mm)
Unidirectional BFRP	1673	65.7	2.5	0.107
Bidirectional BFRP	940	57.2	1.6	0.112
CFRP	1390	96.5	1.4	0.168

FRP: fiber-reinforced polymer; BFRP: basalt fiber–reinforced polymer; CFRP: carbon fiber–reinforced polymer.

Test and instrumentation

As shown in Figure 1(a) and (b), the specimens are tested using a hydraulic axial compression machine with a maximum load capacity of 5000 kN. The steel bearing plates guaranteed a uniform compressive load. The specimen was subjected to a monotonic static axial load in step. Each load step was 20 kN at a rate of 1 kN/s, and each load level was kept constant for 2 min to record crack propagation. One YHD-50 displacement transducer was fixed at the column bottom to measure vertical displacement. The strain gauges of concrete and FRP jackets are depicted in Figure 1(c), which are pasted horizontally and vertically at the mid-height.

Figure 1.

Test setup: (a) photo, (b) schematic diagram, and (c) position of instrumentation.

Experimental results and discussion

Failure modes

The typical failure modes of the specimens are shown in Figure 2. The unconfined concrete columns crushed as depicted in Figure 2(a) and (f), wide longitudinal cracks, and small concrete fragments are observed at the ultimate state. The confined concrete columns all failed by FRP jackets rupturing around the mid-height, as shown in Figure 2(b) to (e) and (g) to (j). During the load course, some BFRP sheets first became shallow color with white matrix particles, which indicated the cracking of the resin and concrete inside. Then the load continued to increase and the fibers started to rupture near the mid-height with evident concrete dilation. Finally, the column failed suddenly by FRP jacket fracturing with an explosive sound. For square specimens, the FRP jackets always fractured at the section corners.

Figure 2.

Typical failure modes of specimens: (a) C-0, (b) C-B1, (c) C-B2, (d) C-B1C1, (e) C-2B1, (f) S-0, (g) S-B1, (h) S-B2, (i) S-B1C1, and (j) S-2B1.

Compressive strength and ultimate strain

The test results of compressive concrete strength and axial strain are listed in Tables 3 and 4, where $f'_{co}$ and $ε_{co}$ are the peak stress and the corresponding strain of unconfined concrete, respectively; $ε_{cu}$ is the ultimate strain of unconfined concrete; $f'_{cc}$ and $ε_{cc}$ are the peak stress and the corresponding strain of FRP-confined concrete with strain-softening curves, respectively; $f'_{cu}$ and $ε_{ccu}$ are the ultimate stress and the corresponding strain of FRP-confined concrete with strain-hardening and strain-softening curves, respectively; and the notation with the overline represents the mean test value of three specimens in each group. The detailed locations of the notations in the typical stress–strain curves of FRP-confined concrete columns are shown in Figure 3.

Table 3.

Mechanical characteristics of unconfined specimens.

ID	$f'_{co}$ (MPa)	$\bar{f}'_{co}$ (MPa)	$ε_{co}$ (%)	${\bar{ε}}_{co}$ (%)	$ε_{cu}$ (%)	${\bar{ε}}_{cu}$ (%)
C0-1	26.11	28.29	0.13	0.16	0.25	0.23
C0-2	27.69		0.18		0.23
C0-3	31.06		0.16		0.20
S0-1	33.31	32.57	0.12	0.13	0.20	0.22
S0-2	31.12		0.12		0.24
S0-3	33.27		0.16		0.21

The last digit of ID indicates the numbering of specimens in each group.

Table 4.

Mechanical characteristics of confined specimens.

ID	$f'_{cc}$ (MPa)	$f'_{cu}$ (MPa)	$ε_{cc}$ (%)	$ε_{ccu}$ (%)	$f'_{cc} / \bar{f}'_{co}$	$f'_{cu} / \bar{f}'_{co}$	$ε_{cc} / {\bar{ε}}_{co}$	$ε_{ccu} / {\bar{ε}}_{cu}$
C-B1-1	–	36.56	–	0.46	–	1.29	–	2.00
C-B1-2	–	37.84	–	0.43	–	1.34	–	1.87
C-B1-3	–	39.23	–	0.42	–	1.39	–	1.83
C-B2-1	–	51.70	–	1.00	–	1.82	–	4.35
C-B2-2	–	48.25	–	0.82	–	1.71	–	3.57
C-B2-3	–	45.47	–	0.92	–	1.61	–	4.00
C-B1C1-1	–	47.59	–	0.50	–	1.68	–	2.17
C-B1C1-2	–	44.14	–	0.55	–	1.56	–	2.39
C-B1C1-3	–	47.29	–	0.63	–	1.67	–	2.74
C-2B1-1	33.57	32.37	0.26	0.36	1.19	1.14	1.63	1.57
C-2B1-2	32.51	30.60	0.21	0.33	1.15	1.08	1.31	1.43
C-2B1-3	35.70	33.12	0.26	0.32	1.26	1.17	1.63	1.39
S-B1-1	41.41	30.52	0.21	0.36	1.27	0.94	1.62	1.64
S-B1-2	43.74	39.45	0.21	0.31	1.34	1.21	1.62	1.41
S-B1-3	39.35	28.30	0.20	0.39	1.21	0.87	1.54	1.77
S-B2-1	47.96	45.40	0.28	0.49	1.47	1.39	2.15	2.23
S-B2-2	43.85	43.34	0.29	0.58	1.35	1.31	2.23	2.64
S-B2-3	45.59	39.97	0.23	0.54	1.40	1.23	1.77	2.45
S-B1C1-1	46.40	44.43	0.30	0.40	1.42	1.36	2.31	1.82
S-B1C1-2	39.03	37.63	0.28	0.53	1.20	1.16	2.15	2.45
S-B1C1-3	43.49	42.61	0.33	0.46	1.34	1.31	2.54	2.09
S-2B1-1	36.00	31.47	0.17	0.33	1.11	0.97	1.31	1.50
S-2B1-2	45.25	42.73	0.24	0.31	1.39	1.31	1.85	1.41
S-2B1-3	39.42	30.48	0.16	0.32	1.21	0.94	1.23	1.45

The last digit of ID indicates the numbering of specimens in each group.

Figure 3.

Stress–strain curves of FRP-confined concrete columns.

In general, all the confined columns had an apparent increase in compressive strength and axial strain capacity compared with the unconfined columns. For circular specimens C-B1 and C-B2, the average strength enhancement ratio $\bar{f}'_{cu} / \bar{f}'_{co}$ improved from 1.34 to 1.71 when the number of BFRP layers increased from 1 to 2, while the average strain enhancement ratio ${\bar{ε}}_{ccu} / {\bar{ε}}_{cu}$ increased from 1.91 to 3.96. The data indicated that the enhancement ratios of compressive strength and ultimate strain improved with increasing number of FRP layers. However, for square specimens S-B1 and S-B2, the average strength enhancement ratios of $\bar{f}'_{cc} / \bar{f}'_{co}$ were 1.27 and 1.41, and the average strain enhancement ratios of ${\bar{ε}}_{ccu} / {\bar{ε}}_{cu}$ were 1.59 and 2.45, respectively. It can be found that the confinement effect of circular columns appeared markedly better than that of square columns.

For hybrid BFRP/CFRP confinement, the average compressive strength and ultimate strain improved 64% and 143% for circular column C-B1C1, and 32% and 109% for square column S-B1C1. The strength enhancement of these two specimens was close to that of specimens C-B2 and S-B2, while the strain improvement was significantly lower, which indicated that BFRP confinement had a better influence on the ultimate strain. As shown in Table 2, BFRPs have lower elastic modulus and larger rupture strain than CFRPs, which can deform in accordance with concrete and make the specimens have better deformation capacity.

Moreover, the compressive strengths and axial strain of bidirectional BFRP-wrapped columns C-2B1 and S-2B1 exhibited relatively lower improvement compared with specimens C-B1 and S-B1. With the same areal mass, bidirectional BFRP sheet had only half of the fibers in the resistant direction compared to the unidirectional category, which provided lower lateral confining pressure.

Stress–strain behavior

Figure 4 indicates the typical relationship between concrete axial stress and axial/lateral strain for each specimen. The axial stress values were calculated as the loading force divided by the cross-sectional area, and the strain values were obtained from the strain gauges on the columns. The stress–strain curves exhibited two stages with a small transition zone. In the first stage, the curves showed approximately linear behavior. The stiffness of the confined column was similar to that of unconfined specimens, indicating that wrapped FRP sheets had little confining effect. After axial stress reached the concrete compressive strength, a transition zone to the second part of the curves occurred. The zone signified that volume expansion appeared in the concrete, and FRP sheets started confining effectively. Then, a parabolic branch appeared for the second portion.

Figure 4.

The typical stress–strain curves of specimens: (a) circular column and (b) square column.

Obvious differences in the second part of the stress–strain curve were found according to the cross-sectional shape and FRP types. The strain-hardening response is observed from the stress–strain curves of all cylinders except for specimen C-2B1 with bidirectional BFRP confinement, as shown in Figure 4(a), while the curves of square columns present an obvious descending branch except for the hybrid-confined specimen S-B1C1, as shown in Figure 4(b). However, one layer of the bidirectional BFRP jacket provided deficient confinement, leading to the exception of cylinder C-2B1. On the other hand, only the stress–strain curves of square column S-B1C1 confined with BFRP/CFRP showed a flat-topped stage in the second portion. According to the study by Ribeiro et al. (2018), the hybridization of BFRP and CFRP promotes synergies to increase the failure strain of CFRPs. Thus, the hybrid effect finally maximized the effectiveness of FRP jacket and then resulted in a pseudo-ductile response of the specimen.

Numerical analysis

Finite element model

Three-dimensional numerical models of FRP-confined circular and square concrete columns are established as in Figure 5. Two rigid plates were added to the column ends. The column bottom was constrained in all translational degrees of freedom, while the top was applied with a concentric load. The C3D8R element was used for concrete, and the four-node shell element S4R was selected for FRP. According to the test results, the FRPs were continuously wrapped around the concrete with sufficient lap length, which made them deform accordantly before the FRP rupture. Thus, the bond–slip behavior between the FRP and concrete was disregarded, and perfect bonding between two materials was assumed in the numerical simulation.

Figure 5.

The finite element models of FRP-confined columns: (a) circular columns and (b) square columns.

The elastic modulus in fiber direction E₁ of the FRP sheets is shown in Table 2. But for the bidirectional BFRP sheets, E₁ is only considered for the warp fiber that lies toward the tensile loading direction, and E₂ is the in-plane elastic modulus that lies perpendicular to the warp direction, which is calculated as 7.8 GPa by the Halpin–Tsai equations (Jones, 1999), as shown in Figure 6(b). The hybrid BFRP/CFRP jacket in columns is defined by composite layup as in Figure 6(a). The bidirectional BFRP jacket of the columns is also approximated as two layers with half thickness in the warp and weft directions, respectively, as depicted in Figure 6(b). Since bidirectional BFRP had the same fibers in the two directions, the mechanical properties in two directions were considered to be the same.

Figure 6.

Ply stack plot of FRP elements: (a) hybrid BFRP/CFRP and (b) bidirectional BFRP.

The Poisson’s ratio $ν_{c}$ of concrete was set to 0.2, and the elastic modulus E_c was 28.2 GPa, which was estimated as $E_{c} = 4730 \sqrt{{f'}_{co}}$ by ACI 318-11/318R-11 (2011). The compressive stress–strain curve was recommended by Chen (1982). The concrete damaged plasticity (CDP) model was selected. The damage parameters were chosen from Chinese Design Code for Concrete Structure (GB 50010-2010, 2010), and the compressive damage parameter $d_{c}$ can be derived from equations (1) and (2).

d_{c} = {\begin{matrix} 1 - \frac{{f'}_{co} n}{E_{c} ε_{co} [n - 1 + {(\frac{ε}{ε_{co}})}^{n}]}, \frac{ε}{ε_{co}} ⩽ 1 \\ 1 - \frac{{f'}_{co}}{E_{c} ε_{co} [α_{c} {(\frac{ε}{ε_{co}} - 1)}^{2} + \frac{ε}{ε_{co}}]}, \frac{ε}{ε_{co}} > 1 \end{matrix}

(1)

n = \frac{E_{c} ε_{co}}{E_{c} ε_{co} - {f'}_{co}}

(2)

where $α_{c}$ is the parameter of the descending branch in compressive stress–strain curve, and $f'_{co}$ and $ε_{co}$ can be obtained from Table 3.

The load was defined as axial displacement, which increased gradually until FRP ruptures. For insufficient confined columns, the calculation was terminated when the load decreased to 85% of the bearing capacity according to ACI 440 2R-08 (2002).

Comparison between numerical and tested results

The simulated stress–strain curves are shown in good agreement with the test results in Figure 7. The initial loading parts of the simulated curves usually provided a good approximation. But the simulated peak load was generally slightly higher than the test values. However, differences clearly appeared in the late plastic stage, which were ascribed to concrete strain fluctuations caused by rapid cracking in the experiment.

Figure 7.

Comparison of axial stress–strain curves between simulated and tested results.

FRP stress distributions

The FRP stress distributions of representative specimens C-B2 and S-B2 are depicted in Figure 8. The hoop stress gradually increases from the column end to the middle section along the height in Figure 8(a) and (b). In square columns, it decreases from the middle part of sides to the sectional corners as in Figure 8(b). The major BFRP stress in circular columns was higher than that in square columns, which verified the more effective confinement in the former.

Figure 8.

Hoop stress distributions in BFRP jacket (Pa): (a) C-B2 and (b) S-B2.

For the hybrid BFRP/CFRP-wrapped column S-B1C1, the stress distributions in two respective plies are generally similar, as depicted in Figure 9(a) and (b). The hoop stress of the outer CFRP ply was approximately 1.5 times that of the inner BFRP ply, which was equal to the ratio of elastic modulus of CFRP to BFRP. The stress distributions of two respective plies in bidirectional BFRP-confined column S-2B1 are shown in Figure 10(a) and (b). The hoop stress of warp fiber ply was much larger than that of the weft fiber ply, which explained the BFRP fiber in the warp ply clearly provided a favorable confinement effect in concrete.

Figure 9.

Hoop stress distributions in hybrid FRP of S-B1C1 (Pa): (a) BFRP ply and (b) CFRP ply.

Figure 10.

Hoop stress distributions in bidirectional FRP of S-2B1 (Pa): (a) warp fiber ply and (b) weft fiber ply.

Sectional concrete stress distributions

The axial concrete stress distributions in the middle height sections of the square columns are shown in Figure 11, which is the dangerous position for failure showing the largest stress values. The stressed distribution at four column section areas all appeared as arching action. The highest stress was obtained at the corners of each column section, and they gradually dropped to approximately 52% of the maximum at the central part of the section and remained nearly consistent in the core area. Moreover, the minimum stress values were seen as triangular contour areas in the middle of the four sides, and these values were approximately 31% of the maximum.

Figure 11.

Concrete stress distributions in the middle cross section (Pa): (a) S-B1, (b) S-B2, (c) S-B1C1, and (d) S-2B1.

By comparing Figure 11(a) with 11(b), the stress in specimen S-B2 is in the range of 13%–25% higher than that in specimen S-B1, which shows that the confinement obviously improves with increasing number of BFRP layers. Figure 11(b) and (c) demonstrates that the stress distributions in BFRP/CFRP-confined column S-B1C1 are generally similar to those in column S-B2, which indicates a similar confining efficiency. However, bidirectional BFRP-confined specimen S-2B1 shows relatively weaker confinement effect as shown in Figure 11(d), in which the axial concrete stress is circa 26% lower than that in specimen S-B1. Generally, these results were all in accordance with the experimental results.

Theoretical model analysis

Strength and strain models for circular columns

For FRP-confined concrete circular columns, the strain-hardening curves were observed in the vast majority of existing test results. The strength and ultimate strain models for FRP-confined cylinders with ascending behavior can be expressed in the following equations (Ozbakkaloglu et al., 2013; Teng et al., 2002)

\frac{{f'}_{cu}}{{f'}_{co}} = 1 + k_{1} \frac{f_{l}}{{f'}_{co}}

(3)

\frac{ε_{ccu}}{ε_{co}} = 1 + k_{2} \frac{f_{l}}{{f'}_{co}}

(4)

f_{l} = \frac{2 f_{frp} t}{d}

(5)

where $f_{l}$ is the confining pressure provided by BFRP fabric, and $k_{1}$ and $k_{2}$ are the strength and strain confinement coefficient, respectively.

For regression analysis, test data containing a total of 29 BFRP-confined plain concrete cylinders with strain-hardening curves were considered from existing literature works (Ma et al., 2018; Sadeghian and Fillmore, 2018; Suon et al., 2019; Wei et al., 2018). All wrapped columns failed by FRP rupture. The test data covered circular specimens with the unconfined concrete compressive strength ranging from 15.8 to 40.3 MPa, the diameter from 112.8 to 150.0 mm, and the length-to-diameter ratio L/d from 2.0 to 4.6. Figure 12 demonstrates the fitting results, and equations (3) and (4) can be rewritten as follows by substituting new parameters

\frac{{f'}_{cu}}{{f'}_{co}} = 1 + 3.41 \frac{f_{l}}{{f'}_{co}}

(6)

\frac{ε_{ccu}}{ε_{co}} = 1 + 21.63 \frac{f_{l}}{{f'}_{co}}

(7)

Figure 12.

Fitting results of circular columns: (a) the strength model and (b) the strain model.

According to the existing literature, the strength confinement coefficient $k_{1}$ in equation (3) normally changed from 2.00 to 4.51 for FRP-confined concrete, in which $k_{1}$ was normally in the range of 2.00–4.14 for CFRP confinement and was approximately 3.40 for AFRP confinement (Ozbakkaloglu et al., 2013). For the ultimate strain expression, the constant value in equation (4) varied from 1 to 2, and the strain confinement coefficient $k_{2}$ was in the range of 6.30–16.00. For instance, $k_{2}$ can be found to range from 7.30 to 15.20 for CFRP-confined concrete and 9.50 for AFRP-confined concrete (Ozbakkaloglu et al., 2013). Generally, it can be concluded that BFRP confinement provided better improvement of the ultimate strain and an intermediate level of strength enhancement among FRP strengthening.

Strength and strain models for square columns

For FRP-confined concrete square and rectangular columns, the strength and strain models changed in format by different shape factors (ACI 440 2R-08, 2002; Al-Salloum, 2007; Lam and Teng, 2003; Wu et al., 2007). Considering the strain-softening behavior of test results, the modified strength and ultimate strain model were based on the model by Al-Salloum (2007) and Wu et al. (2007) as follows

\frac{{f'}_{cc}}{{f'}_{co}} = 1 + k_{1} \frac{b}{D} \frac{f_{l}}{{f'}_{co}}

(8)

f_{l} = \frac{2 f_{frp} t}{D} k_{e}

(9)

k_{e} = 1 - \frac{2 {(1 - \frac{2 r}{b})}^{2}}{3 [1 - (4 - π) {(\frac{r}{b})}^{2}]}

(10)

D = \sqrt{2} b - 2 r (\sqrt{2} - 1)

(11)

where b is the length of the square section, D is the diagonal length of the square section, r is the corner radius, and $k_{e}$ is the shape factor

\frac{ε_{ccu}}{k_{r} ε_{cu}} = 1 + k_{2} \frac{f_{l}}{{f'}_{co}}

(12)

k_{r} = (2 - 1.6 α) \frac{r}{h} + 0.8 α

(13)

f_{l} = f_{frp} t \frac{b + h}{bh}

(14)

α = \frac{30}{{f'}_{co}}

(15)

where $k_{r}$ is the reducing factor for ultimate strain, $α$ is the influencing efficiency considering concrete strength, and b and h are the short and long sides of the rectangular section, respectively.

For BFRP-confined concrete square specimens, a total of 24 tested columns results were available for the assessment of the strength model (Liu et al., 2015; Suon et al., 2019). However, for the strain model, only 20 tested column findings remained, and the strain data from Zhou (2012) were added. These results covered square specimens with the section length ranging from 100 to 150 mm, the corner radius from 0 to 30 mm, and the unconfined compressive concrete strength from 15.6 to 32.9 MPa. The fitting results are shown in Figure 13, and the modified formulas for the confined columns with strain-softening curves can be expressed as follows

\frac{{f'}_{cc}}{{f'}_{co}} = 1 + 3.68 \frac{b}{D} \frac{f_{l}}{{f'}_{co}}

(16)

\frac{ε_{ccu}}{k_{r} ε_{cu}} = 1 + 10.22 \frac{f_{l}}{{f'}_{co}}

(17)

Figure 13.

Fitting results of square columns: (a) the strength model and (b) the strain model.

Based on the experimental data of AFRP-, GFRP-, and CFRP-confined concrete columns, the strength confinement coefficient $k_{1}$ was calculated as 3.14 in equation (8), and the strain confinement coefficient $k_{2}$ was calculated as 6.30 in equation (12) (Al-Salloum, 2007; Wu et al., 2007). Compared to the results of conventional FRPs, the BFRP confinement coefficients presented larger values, indicating better enhancement effects on the strength and ultimate strain.

Evaluation of the predicted models

The statistical performance of the predicted models is listed in Table 5, in which statistical indicators including average value (AVE), standard deviation (SD), and coefficient of variation (COV) are given in equations (18)–(20), respectively

Table 5.

Statistical performance of the predicted models.

Predicted model	Ratio of test and predicted values
	AVE	SD	COV (%)
The strength model of circular columns	1.00	0.08	8.08
The strain model of circular columns	1.02	0.16	15.84
The strength model of square columns	1.03	0.08	7.26
The strain model of square columns	1.05	0.19	18.33

AVE: average value; SD: standard deviation; COV: coefficient of variation.

AVE = \frac{1}{N} \sum_{i = 1}^{N} {(\frac{Test results}{Predicted results})}_{i}

(18)

SD = \sqrt{\frac{1}{N} {\sum_{i = 1}^{N} ({(\frac{Test results}{Predicted results})}_{i} - AVE)}^{2}}

(19)

COV = \frac{SD}{AVE} \times 100 %

(20)

where N is the total number of data in the test results.

For BFRP-confined circular and square columns, the predicted models show good agreement with the test results, as shown in Table 5. These results indicated that the predicted models of circular columns fit better with the test data than the models of square columns, and the predicted strength models were also in closer agreement than the strain models, with significantly smaller SD and COV values. However, two data with error of more than 30% were abandoned as unreasonable points for the strain model of circular columns, which was due to larger confining pressure and strain fluctuations in the experiment.

Conclusion

In this research work, the compressive behaviors of BFRP-wrapped concrete columns were investigated focusing on the influences of cross-sectional shape, BFRP type, and number of layers. According to the experimental, numerical, and theoretical analyses, the following conclusions can be drawn:

By experimental results, it was found that BFRP confinement remarkably improved the compressive strength and ultimate strain of concrete with increasing layers, especially in the circular columns. The two-layer unidirectional BFRP-confined columns showed the best compressive behavior, in which strength and ultimate strain were improved 71% and 296% for circular section, and 41% and 145% for square section.

The hybrid BFRP/CFRP obtained the intermediate level of confinement effectiveness, but it fulfilled the effectiveness of two kinds of FRP. The numerical model indicated that the hoop stress in the CFRP ply was approximately 50% higher than that in the BFRP ply, and the experimental curves of square columns exhibited obvious descending behavior except for the hybrid confined columns. The bidirectional BFRP showed relatively lower effectiveness of confinement with the same number of FRP layers, because only the warp fiber layer provided the main confinement effect as reflected by numerical model, and the strain-hardening response of the stress–strain curve was observed for circular columns except for the bidirectional BFRP-confined specimens in the tests.

By regression analysis of existed experimental data, modified strength and strain models for BFRP-confined circular and square concrete columns were proposed, which showed that the strain confinement coefficient was significantly larger than that of conventional FRPs. For instance, it was calculated as 20.76 for BFRP-confined concrete cylinders and is normally in the range of 7.30–15.20 for CFRP confinement. This indicated that BFRP has a better confined effect on the ultimate strain of concrete.

It should be pointed out that more experiments should be carried out to enlarge the database for BFRP-wrapped concrete, and more parameters need to be investigated.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article was sponsored by the National Natural Science Foundation of China (Grant No. 51208166) and the Natural Science Foundation of Anhui Province (Grant No. 1408085MKL14).

ORCID iD

Peng Gao

References

ACI 318-11/318R-11 (2011) Building code requirements for reinforced concrete and commentary.

ACI 440 2R-08 (2002) Guide for the design and construction of externally bonded FRP systems for strengthening concrete structures.

Al-Salloum

(2007) Influence of edge sharpness on the strength of square concrete columns confined with FRP composite laminates. Composites Part B: Engineering 38(5): 640–650.

Aslani

Kohnehpooshi

(2018) Structural behavior of FRP strengthened reinforced concrete shear walls with openings using finite element method. Advances in Structural Engineering 21(7): 1072–1087.

ASTM D3039/D3039M-08 (2008) Standard test method for tensile properties of polymer matrix composite materials.

Baky

Ebead

Neale

(2010) Statistical analyses and parametric study for reinforced concrete beams strengthened in flexure with FRPs. Advances in Structural Engineering 13(5): 805–822.

Campione

La Mendola

Monaco

, et al. (2015) Behavior in compression of concrete cylinders externally wrapped with basalt fibers. Composites Part B: Engineering 69: 576–586.

Chen

(1982) Plasticity in Reinforced Concrete. New York: McGraw-Hill.

Deng

Raphael-Daud

(2012) Axial compression behaviors of concrete confined by basalt fiber sheets. Applied Mechanics and Materials 238: 13–16.

10.

Dhand

Mittal

Rhee

, et al. (2015) A short review on basalt fiber reinforced polymer composites. Composites Part B: Engineering 73: 166–180.

11.

GB 50010-2010 (2010) Design code for concrete structure.

12.

Jing

Cao

(2005) A model for calculating the axial s-strain curve of square section concrete column confined by FRP. China Civil Engineering Journal 38(12): 32–37 (in Chinese).

13.

Jones

(1999) Mechanics of Composite Materials. 2nd ed. London: Taylor & Francis.

14.

Lam

Teng

(2003) Design-oriented stress-strain model for FRP-confined concrete in rectangular columns. Journal of Reinforced Plastics and Composites 22(13): 1149–1186.

15.

Lam

Teng

(2004) Ultimate condition of fiber reinforced polymer-confined concrete. Journal of Composites for Construction 8(6): 539–548.

16.

Zeng

, et al. (2017) Experimental study on axial compressive behavior of hybrid FRP confined concrete columns. Computers and Concrete 19(4): 395–404.

17.

Liu

Wang

, et al. (2015) Effect of cross-sectional aspect ratio and basalt fiber-reinforced polymer-confined number on axial compression behavior of short columns. Journal of Reinforced Plastics and Composites 34(10): 782–794.

18.

Ludovico

Prota

Manfredi

(2010) Structural upgrade using basalt fibers for concrete confinement. Journal of Composites for Construction 14(5): 541–552.

19.

Yan

, et al. (2018) Testing and analysis of basalt FRP-confined damaged concrete cylinders under axial compression loading. Construction and Building Materials 169: 762–774.

20.

Mander

Priestley

Park

(1988) Theoretical stress-strain model for confined concrete. Journal of Structural Engineering 114: 1804–1826.

21.

Monaldo

Nerilli

Vairo

(2019) Basalt-based fiber-reinforced materials and structural applications in civil engineering. Composite Structures 214: 74–94.

22.

Ozbakkaloglu

Lim

Vincent

(2013) FRP-confined concrete in circular sections: review and assessment of stress–strain models. Engineering Structure 49: 1068–1088.

23.

Rashid

RSM

Aboutaha

(2014) Analytical model for CFRP strengthened circular RC column under elevated temperature. Computers and Concrete 13(4): 517–529.

24.

Ribeiro

Sena-Cruz

Branco

, et al. (2018) Hybrid FRP jacketing for enhanced confinement of circular concrete columns in compression. Construction and Building Materials 184: 681–704.

25.

Sadeghian

Fillmore

(2018) Strain distribution of basalt FRP-wrapped concrete cylinders. Case Studies in Construction Materials 9: 1–17.

26.

Suon

Saleem

Pimanmas

(2019) Compressive behavior of basalt FRP-confined circular and non-circular concrete specimens. Construction and Building Materials 195: 85–103.

27.

Teng

Chen

Smith

, et al. (2002) FRP: Strengthened RC Structures. Chichester: John Wiley & Sons.

28.

Teng

Jiang

Lam

, et al. (2009) Refinement of a design-oriented stress–strain model for FRP-confined concrete. Journal of Composites for Construction 13(4): 269–278.

29.

Wei

Zhang

, et al. (2018) Behaviour of concrete confined by both steel spirals and fiber-reinforced polymer under axial load. Composite Structures 192: 577–591.

30.

Lü

(2007) Design-oriented stress–strain model for concrete prisms confined with FRP composites. Construction and Building Materials 21(5): 1107–1121.

31.

Wei

(2015) General stress-strain model for steel- and FRP-confined concrete. Journal of Composite for Construction 19(4): 04014069.

32.

Xia

Xian

Kafodya

, et al. (2014) Compression behavior of concrete cylinders externally confined by flax fiber reinforced polymer sheets. Advances in Structural Engineering 17(12): 1825–1834.

33.

Zhang

Wei

Bai

, et al. (2019) Stress-strain model of an FRP-confined concrete filled steel tube under axial compression. Thin-Walled Structures 142: 149–159.

34.

Zhou

(2012) Research on the behavior of rectangular concrete short columns confined with BFRP sheets. Master’s Thesis, Liaoning University of Technology, Jinzhou, China (in Chinese).

Compressive behavior of circular and square concrete column externally confined by different types of basalt fiber–reinforced polymer

Abstract

Keywords

Introduction

Experimental program

Specimen design

Material properties

Test and instrumentation

Experimental results and discussion

Failure modes

Compressive strength and ultimate strain

Stress–strain behavior

Numerical analysis

Finite element model

Comparison between numerical and tested results

FRP stress distributions

Sectional concrete stress distributions

Theoretical model analysis

Strength and strain models for circular columns

Strength and strain models for square columns

Evaluation of the predicted models

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

References