Abstract
Basalt Fiber Reinforced Polymer (BFRP) bars have bright potential application in concrete structures subjected to marine environment due to their superior corrosion resistance. Available literatures mainly focused on the mechanical properties of BFRP concrete structures, while the bond-slip behavior of BFRP bars, which is a key factor influencing the safety and service life of ocean concrete structures, has not been clarified yet. In this paper, effects of BFRP bars size, corrosion age, and concrete strength on the bond-slip behavior of BFRP bars in concrete cured in artificial seawater were investigated, and then an improved Bertero, Popov, and Eligehausen (BPE) model was employed to describe the bond-slip behavior of BFRP bars in concrete. The results indicated that the maximum bond stress and corresponding slip decreased gradually with the increase of corrosion age and size of BFRP bars, and ultimate slip also decreased sharply. The ascending segment of bond-slip curve tends to be more rigid and the descending segment tends to be softer after corrosion. A horizontal end in bond-slip curve indicates that the friction between BFRP bars and concrete decreased sharply.
1. Introduction
Traditional reinforced concrete (RC) structures often encounter durability problems due to cracks caused by corrosion expansion of steel bar [1]. With respect to steel bar, Basalt Fiber Reinforced Polymer (BFRP) bars present superior corrosion resistance as a kind of novel nonmetallic reinforced bars with high chemical stability [2–5]. Recently, BFRP bars are regarded as a potential alternative material to replace traditional steel bar in reinforced concrete structures subjected to extreme corrosive environment, such as ocean concrete structures [6–10].
Numerous researches have been carried out on mechanical properties and bond durability of BFRP bars or BFRP reinforced concrete [11–14]. The tensile strength loss of BFRP bars was as high as 40% after immersion in 55°C alkaline solution for 63 days [11], and the bond strength of BFRP bars in concrete immersed in artificial seawater for 90 days exhibited a 25% reduction [12]. Tighiouart et al. [13] confirmed that it was easy bleeding at the interface between BFRP bars and concrete when the large diameter BFRP bars were used, resulting in a decreased bond strength between BFRP bars and concrete. Shen et al. [14] found that the bond strength of the BRFP bars in concrete remained essentially unchanged after exposure to 40°C artificial seawater for 60 days.
The bond-slip behavior, which affects bar anchoring, lap splice strength, concrete-cover requirement, serviceability, and ultimate states, is a key feature for the successful application of BFRP bars as internal reinforcement in concrete structures [15]. It should be noted that the bond-slip behavior of BFRP bars in concrete was different from that of steel bar [16, 17]. Although some achievements about the bond-slip behavior between BFRP bars and concrete have been made [11, 16–24], most of them were focused on small size BFRP bars (<Ф 10 mm). However, the bond-slip behavior between large size BFRP bars and concrete is different due to larger interfacial transition zone, internal bleeding, and smaller specific perimeter. In addition, long term bond-slip behavior of BFRP bars in concrete exposed to marine environment has not been clarified yet.
Large size BFRP bars (>Ф 20 mm) are expected to be used as reinforced material in ocean concrete structures, especially for south China coastal area. Therefore, effects of BFRP bars size, corrosion age, and concrete strength on the bond-slip behavior of BFRP bars in concrete subjected to artificial seawater were investigated in the present study, and an improved Bertero, Popov, and Eligehausen (BPE) model was employed to describe the bond-slip behavior of BFRP bars. The results will give a better understanding of bond-slip behavior of BFRP bars in concrete subjected to marine environment and finally provide fundamental data for the design and service life prediction of marine concrete structures.
2. Materials and Methods
2.1. Materials
Basalt filaments were pulled over a roller in the pultrusion process to form bars; then, the bars were coated with epoxy resin and twined with nylon wire (Ф 0.5 mm). The epoxy resin was solidified immediately after adhering a layer of quartz sand on the bars. The obtained BFRP bars are regarded as desirable reinforcement material in corrosion environment due to their high strength, low weight, corrosion resistance, and cost performance [25]. BFRP bars with size of Ф 8 mm, Ф 20 mm, and Ф 25 mm were used as shown in Figure 1. Their mechanical properties measured according to ASTM D3039 [26] are listed in Table 1.
(a) Specimen photograph
(b) Specimen geometry
ASTM type II 42.5 Portland cement, river sand (fineness modules of 2.7), crushed limestone (particle size of 5.0 mm–31.5 mm), tap water, and polycarboxylate superplasticizer (water reducing ratio of 25%) were used to prepare concrete according to mixture proportions shown in Table 2, and 28-day compressive strength of concrete tested as specified in ASTM C39 [27] is also listed.
2.2. Specimen Preparation
The pullout specimen consisted of 600 mm long BFRP bars embedded centrally in concrete column (Ф 150 mm × 150 mm) as shown in Figure 2(a). The embedded length was kept constant at 5 times as the diameter of the BFRP bars according to FILEM/FIP recommendation RC6 [28], and PVC tube was used at the loading end of the bar to minimize the stress concentration near the loading plate. Steel pipes were used as anchors at the loading end and were cast with epoxy resin before casting. Fresh concrete was cast into the mould and then vibrated for 30 seconds (Figure 2(b)); demoulding was performed after curing at 20 ± 1°C and 90% relative humidity chamber for 24 h. The exposed free end of the bars was sealed by epoxy resin (Ф 10 mm); the specimens were cured in lime-saturated water at 20 ± 1°C for 27 days and then immersed artificial seawater (NaCl 24.53 g/L, MgCl2 20 g/L, Na2SO4 4.09 g/L, and CaCl2 1.16 g/L according to ASTM D 1141-98 [29]) under 40°C; schematic of conditioning container and details are depicted in Figure 3.
(a) Specimen geometry
(b) Specimen photograph
2.3. Pullout Test Procedures
Figure 4 depicts the typical details and schematic for pullout test according to ASTM D7913 [30]. The tests were carried out with a MTS testing machine in displacement control mode at a rate of 2 mm/min. The displacements of the free end and loading end of BFRP bar were measured with linear variable displacement transducers (LVDT), and the applied load and displacements were recorded automatically throughout a data-acquisition system. The slip of the BFRP bars in concrete can be obtained as follows: where is the slip of the BFRP bars (mm); is the displacement of the loading end of the BFRP bars (mm); is the displacement of the free end of the BFRP bars (mm).
The bond stress can be calculated assuming a uniform bond stress distribution along the embedded length of the bar in concrete using where is the bond stress (MPa); is the pullout force which was measured by pressure sensor (N); is the diameter of BFRP bars (mm); (mm) is bond length of the BFRP bars in concrete.
3. Results and Analysis
3.1. Effect of Corrosion Age on the Bond-Slip Behavior of BFRP Bars in Concrete
The bond-slip behavior of the BFRP bars in concrete shown in Figure 5 can be described by four parameters: and are the maximum bond stress and corresponding slip, while and are ultimate slip and corresponding bond stress. Both and decreased gradually with the increase of corrosion age. For instance, and of BRFP bars (Ф 25 mm) in concrete before corrosion were 21.5 MPa and 3.0 mm and reduced to 17.8 MPa and 1.7 mm after 90-day corrosion in artificial seawater, respectively, indicating that the maximum bond stress decreased with the prolong of corrosion age. of BRFP bars (Ф 25 mm) increased from 1.1 MPa to 8.1 MPa after 90 days of corrosion, while was decreased sharply from 10.0 mm to 5.2 mm. It should be noted that a horizontal end in bond-slip curve was observed after corrosion, which means significant reduction in ultimate slip after corrosion. Similar tendency was found for BRFP bars with size of Ф 8 mm, and both , and , of BRFP bars with size of Ф 8 mm had higher value than those of BRFP bars with size of Ф 25 mm.
(a) Ф 8 mm
(b) Ф 25 mm
3.2. Effect of Bars Size on Bond-Slip Behavior of BFRP Bars in Concrete
As shown in Figure 6, the maximum bond stress and corresponding slip decreased with the diameter increase of BFRP bars, and ultimate slip also decreased slightly. After 90 days of corrosion, both the maximum bond stress and corresponding slip of the BFRP bars decreased about 8.0%. An obvious horizontal end was observed for all specimens after corrosion, indicating that the friction between BFRP bars and concrete decreased sharply.
(a) Before corrosion
(b) After 90 days of corrosion
3.3. Effect of Concrete Strength on Bond-Slip Behavior of BFRP Bars in Concrete
With the increase of concrete strength, the maximum bond stress increased gradually, while slip corresponds to maximum bond stress and ultimate slip decreased as shown in Figure 7. After 90 days of corrosion, both the maximum bond stress and corresponding slip of the BFRP bars increased with the increase of concrete strength. BFRP bars in high strength concrete had lager ultimate slip. It can be inferred that high strength concrete is benefit to the bond performance between BFRP bars and concrete.
(a) Before corrosion
(b) After 90 days of corrosion
4. Discussion
BPE model [31] is generally used to describe the bond-slip of steel bar in concrete (Figure 8(a)). The ascending segment of bond-slip curve can be expressed as follows:where is the maximum bond stress; is the corresponding slip; is a curve-fitting parameter .
(a) Original BPE model
(b) Improved BPE model
For a second segment with a constant bond up to a slip , a linearly descending segment developed from to ; and a horizontal segment developed for , with a value of τ due to the development of friction .
Cosenza et al. proposed an improved BPE model to describe the bond-slip curve of FRP bars in concrete by ignoring the second segment (Figure 8(b)) [32]. The ascending segment is coincident with the original BPE model but the slope of descending segment is , which is given by where is parameter accounting for the softening, is the ultimate slip, and is friction slip due to bond resistance.
Obviously, the improved BPE model is more suitable to describe the bond-slip behavior of BRFP bars in concrete. The improved BPE model can be further described as , , and , and (area underneath the ascending segment as shown in Figure 9) can be written as follows:Then, and in the improved BEP model can be calculated from by
It can be inferred that a low value of means bond stress develops without significant slip, and the ascending segment can be well reproduced using a rigid law. For the descending segment, a low value of and indicates larger slips, and the descending segment can be well fitted using a softening law. Tables 3–6 and Figure 10 showed that the ascending segment tends to be more rigid and the descending segment tends to be softer after corrosion, and the corrosion has much more significant effect on the bond-slip behavior of large size BFRP bars in concrete.
(a) Ф 8 mm
(b) Ф 25 mm
5. Conclusions
Main conclusions that can be drawn from the present study were as follows:(1)The maximum bond stress and corresponding slip decreased gradually with the increase of corrosion age, and ultimate slip also decreased sharply from 10.0 mm to 5.2 mm. A horizontal section in bond-slip curve was observed after corrosion, indicating the friction between BFRP bars and concrete decreased sharply.(2)The maximum bond stress and corresponding slip decreased with the diameter increase of BFRP bars, and ultimate slip also decreased slightly. Both the maximum bond stress and corresponding slip of the BFRP bars increased with the increase of concrete strength. BFRP bars in high strength concrete possessed larger ultimate slip.(3)The ascending segment of bond-slip curve tends to be more rigid and the descending segment tends to be softer after corrosion, and the corrosion has much more significant effect on the bond-slip behavior of large size BFRP bars in concrete.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was funded by National Key Research and Development Program (no. 2016YFB0303502), Water Conservancy Science and Technology of Guangdong Province of China (2015-10 and 2016-23) and the China Postdoctoral Science Foundation (no. 2016M590776). Their financial supports are gratefully acknowledged.