Flexural behavior of corroded reinforced concrete beam strengthened with jute fiber reinforced polymer

Abstract

Fiber-reinforced polymer (FRP) is a revolutionary breakthrough in the history of structural engineering innovation due to its unique characteristics to strengthen and repair the deficient reinforced concrete structures. This paper aimed at evaluating the flexural characteristics of jute fiber reinforced polymer (JFRP) bonded reinforced concrete beams. The influence of the test variables comprised of strengthening scheme and corrosion rate for reinforced concrete (RC) beams. The experimental study comprised of casting six RC beams and testing them in flexure loading. To determine the flexural response of RC beams, three beams were fabricated with JFRP laminate having a level of corrosion of 0%, 7.5%, and 15%, whereas three beams were designated as control beams having same corrosion levels with no JFRP. Test results indicated that all JFRP strengthened beams exhibited increased ultimate load, yield load, first cracking load, and lower ductility index compared to control beams. The results also revealed that JFRP strengthening technique improved the flexural strength of the corroded beams efficiently, albeit the ultimate load of the beams diminished with higher corrosion level. Analytical calculations were carried out for quantifying the flexural characteristics and mass loss of beams which provided a good agreement with the test results.

Keywords

corrosion flexural behavior Jute FRP (JFRP)RC beam strengthening

Introduction

Corrosion of steel reinforcing bar has become a severe problem in the existing building, which leads to degradation and long-term deterioration of RC structures. During the service life of RC structures, embedded steel has a contact in corrosion due to poor concrete quality, moisture ingress, proximity to sea-shores, and lack of adequate concrete cover. Steel corrosion occurs mainly employing the decline of cross-sectional areas of reinforcements and expansion of corrosion products which affect the yield, ultimate and ductility performance and result in premature collapse (Al-Saidy et al., 2010). The corrosion is mainly manifested in the mode of cracking and spalling of clear cover of concrete, reducing of bond between concrete and rebar, losing of ultimate strength and reduction of the service life of structures (Azad et al., 2010; Fang et al., 2004; Mancini et al., 2014; Šavija et al., 2013). Therefore, deficient and functionally obsolete corroded RC structures or members require an urgent upgrade in the form of repair and rehabilitation method to overcome the amalgamated cost. As a result, various suitable cost-effective strengthening techniques have been seen in practice which can greatly reduce construction cost, enhance safety and improve the estimated life span of these structures (Valluzzi et al., 2014). One of the viable strengthening techniques is the advancement of using FRP in various forms for structural engineering applications. FRPs have been widely studied as a strengthening scheme due to its advantages including high tensile capacity, light unit weight, excellent durability, resistance to corrosion, versatility and design flexibility (Baggio et al., 2014; Pham et al., 2013; Rahimi and Hutchinson, 2001). Many researchers have reported that FRP has the capability of retrofitting RC structures by restoring the flexural capacity of corroded and control damaged beams (Al-Saidy et al., 2010; Barris et al., 2020; Davalos et al., 2013; Dong et al., 2013; Zhang et al., 2019).

Toutanji et al. (2006) reported that adding CFRP layers on the soffit of the beams increased the flexural strength of the beams, whereas the repaired beam’s ductility reduced significantly than unstrengthened beams. Kim and Shin (2011) tested hybrid FRPs retrofitted beams to study the combined influence of carbon and glass fiber laminates on the flexural performance. It was concluded that hybrid FRPs contributed to the improvement of ultimate capacity and stiffness where the order of the layers played a vital role in controlling ultimate strength and ductility of repairing scheme. Choi et al. (2013) examined the debonding failure of small hybrid FRPs strengthened RC beams. It was reported that higher debonding strength was achieved in stiffer and thinner layer strengthened beams than lesser stiff and thick layer beams, which suggested that the thickness of the material influenced the debonding behavior of repaired beams. The ultimate capacity of RC beams repaired with carbon or glass sheets increased significantly than control beams as reported by Hawileh et al. (2014). Besides, glass and hybrid sheets strengthened beams displayed higher ductility behavior than a single carbon fiber laminate beam. Chen et al. (2018) conducted the flexural response tests of repaired RC beams with basalt FRP sheet in different configurations. It was noted that incorporation of BFRP sheet enhanced the flexural strength where the strengthening performance was more effective in inclined U-jackets than other schemes.

Soudki et al. (2007) studied the influence of CFRP to analyze the flexural behavior of corroded RC beams. It was reported that CFRP influenced the flexural performance of the beams significantly by enhancing the strength and reducing the corrosion effects. Al-Saidy et al. (2010) investigated the structural behavior of CFRP repaired RC beams exposed to accelerated corrosion. The ultimate strength restored in repaired corroded damaged beams with CFRP sheets, however, these beams showed lower ultimate deflection than the corresponding deflection of control beams. Al-Hammoud et al. (2011) used CFRP laminates to strengthen RC beams in the action of corrosion and investigated the fatigue flexural performance. It was noticed that the fatigue capacity in CFRP repaired corroded beams improved than the undamaged control beams. Elghazy et al. (2018) tested nine RC beams in which exposing fabric-reinforced cementitious matrix (FRCM) repaired corroded beams resulted in about 23% reduction in mass loss of steel. The utilization of U-shaped FRCM strips had greater efficacies in declining corrosion percentage and improving the flexural capacity of strengthened beams.

Hence, it can be seen that numerous studies have been performed to strengthen RC beams using CFRP, GFRP, AFRP, and BFRP. However, these fibers are artificial or man-made fibers which require large manufacturing methods and higher production costs. In addition, artificial FRPs are a vital source of CO₂ emissions during their processing and manufacturing stages and causes various environmental issues. From these perspectives, plant-based natural jute fibers can be used as a strengthening scheme for repairing and rehabilitation of existing RC structures considering its environmental and sustainable benefits. Besides, utilization of natural jute fibers can be a value-added cheap solution for the enhancement of performance of RC structures considering its manufacturing methods, reduced energy consumption, eco-friendly, local availability, and low cost. Huang et al. (2016) investigated the effectiveness of natural flax fabric reinforced polymer (FFRP) composite plates to strengthen RC beams varying FFRP thickness and steel ratio. They reported that FFRP plates contributed substantially to increasing the ultimate load, deflection and ductility and behaved as an effective external reinforcement material for strengthening of the beams. Chen et al. (2020) determined the advantages of nature fiber-reinforced polymer (NFRP) over synthetic FRP in terms of strengthening effects and material cost. It was observed that NFRP enhanced the ultimate load of strengthened beams substantially and showed better strengthening performance compared to control and CFRP beams. Sen and Jagannatha Reddy (2013) also investigated the efficiency of JFRP for repairing of RC beams in flexure as compared to CFRP and GFRP.

However, a limited number of studies have been conducted to study the feasibility of using JFRP on the investigation of the flexural response of corroded RC beams. Although CFRP or GFRP has superior performance compared to natural FRPs or JFRP, they are non-sustainable materials and generates various environmental-sustainability issues along with the higher cost. In this case, natural jute fibers have the potentiality as a strengthening technique for repairing of existing RC structures considering its environmental and sustainable benefits, moderate properties, local availability, and low cost. In this study, the effects of JFRP on flexural behavior of RC beams were experimentally investigated as a strengthening scheme under normal and corrosion action. The experimental results were also compared with analytical results in terms of ultimate load, deflection, first cracking load, and mass loss of tensile bars.

Experimental program

Specimen configuration and details

Six RC rectangular beams were fabricated and tested in the laboratory. The beam specimens were categorized into two main groups: control beam (CB) and JFRP strengthened beam (JFB). Three beams in the first group were not repaired with JFRP laminates and treated as control beams, while the second group comprised of three beams strengthened with JFRP. Three beams were corroded to different degrees of corrosion that is 0%, 7.5%, and 15% using an impressed electrical current. Table 1 presents the details of the test matrix of the beam specimens. The beam dimensions and reinforcement configuration details with long and cross-section are depicted in Figure 1(a) and (b). It should be noted that the stainless-steel bar of 6 mm diameter was put inside the beam to work as cathode function for the initiation of corrosion.

Table 1.

Description of test matrix.

Specimen designation	Strengthening scheme	Percentage of corrosion (%)	Duration of corrosion (days)
CB-0	No	0	0
JFB-0	Yes	0	0
CB-7.5	No	7.5	4
JFB-7.5	Yes	7.5	4
CB-15	No	15	8
JFB-15	Yes	15	8

CB-0: Control beam at 0% corrosion level; CB-7.5: Control beam at 7.5% corrosion level; CB-15: Control beam at 15% corrosion level; JFB-0: Strengthened beam at 0% corrosion level; JFB-7.5: Strengthened beam at 7.5% corrosion level; JFB-15: Strengthened beam at 15% corrosion level.

Figure 1.

(a) Longitudinal section and (b) cross-section with reinforcement details of the beam specimens.

Material properties

In this study, the same materials were used to perform the casting, reinforcing and strengthening of all beams. The cement utilized in all mixes contained Ordinary Portland cement (OPC) meeting the specifications of ASTM C150 (ASTM, 2001). The coarse aggregate was natural crushed stone chips with a maximum size of 20 mm, unit weight of 1620 kg/m³, FM of 5.65 and specific gravity of 2.78. Locally available Sylhet sand was employed as fine aggregate (FM = 2.93) with specific gravity of 2.64, unit weight of 1565 kg/m³ and water absorption of 2.3%. The basic constituents of the mix ratio were set as 1:1.75:2.75:0.5 by weight of OPC, fine aggregate, coarse aggregate, and water respectively. The mixing was done for all concrete batches according to ASTM C685 (ASTM, 2002). The slump test of fresh concrete was conducted following ASTM C143 (ASTM, 2003). Compressive and splitting tensile strength tests were done on cylindrical molds of 100×200 mm according to ASTM C39 (ASTM, 2010) and ASTM C496 (ASTM, 2011) respectively.

The compressive and splitting tensile strength of concrete was obtained 25 (0.85) MPa and 2.65 (0.05) MPa respectively at 28 days. The value in the bracket presents the standard deviation. The fresh concrete slump was measured 96 mm. The yield strength and modulus of elasticity of 12 mm diameter rebar were found 503 (2.08) MPa and 200 (2.65) GPa respectively, whereas they were 465 (2.52) MPa and 200 (2.31) GPa for 10 mm diameter rebar. Besides, the yield strength was found 387 (2) MPa for 6 mm diameter plain bar. Jute fibers were collected from the local jute factory of Khulna, Bangladesh and examined to find the basic properties. The collected jute fibers were from locally known “Tusha Pat” and golden in color. These jute fibers were roving type entangled with one another which was separated fiber by fiber and used for the preparation of JFRP. The diameter, absorption and density of the jute fibers were obtained about 160–180 µm, 168% (7.64) and 1.41 g/cm³ respectively. In addition, the tensile strength and modulus of elasticity of the fibers were found around 490 (6.81) MPa and 26.5 (1.8) GPa. The elongation percentage of jute fiber at failure was about 1.8% (mm/mm). The epoxy was collected from Mapei Far East Pte. Ltd., Singapore through a local distributor. The properties of epoxy as supplied by the manufacturer had a compressive strength of 80 MPa, tensile strength of 30 MPa, elastic modulus of 4000 MPa and flexural strength of 40 MPa. The mass density of Part A epoxy resin was 1.6 g/cm³ and Part B hardener had a density of 1.8 g/cm³.

Specimen preparation, jute fabrication, and surface treatment phase

The preparation of beam specimens comprised of formwork preparation, installing the steel reinforcement, concrete mixing and pouring, curing, strengthening with JFRP and artificial corrosion. The formwork preparation for the pouring of concrete is shown in Figure 2(a). For the strengthening scheme of the beams, JFRP laminate of 100 mm width and a total thickness of 6.25 mm was prepared into the requisite length. It should be noted that three sub-layers of JFRP with incorporating epoxy adhesive in each layer formed it the final thickness of 6.25 mm of one layer. Prior to attaching JFRP laminate to the tension face of the beams, the concrete surface was carefully prepared. At first, the bottom surface of the concrete beam was made rough using coarse sandpaper and an air blower was used to remove all accumulated dust and weak concrete. Then the surface was partially coated with putty and cleaned by acetone followed by applying primer to fill the little honeycomb in the soffit of the beam. After preparing the beam surface, the epoxy resin combined with Part-A and Part-B hardener were combined with 3:1 ratio following the manufacturer’s instruction manual. The mixing procedure continued until a completely homogenized grey paste was obtained. The nominal thickness of epoxy adhesive was maintained into JFRP surface of the beam soffit. For this system, three ruler marks were used on both sides of the wooden mold at every 300 mm distance to get a perfect view of the levels and appropriate thickness of JFRP laminate. A rope and bubble leveler were used to check the final thickness of JFRP laminate from one side to another along the length and a trowel was used for the final finishing of the thickness of epoxy. The conventional hand lay-up procedure in layer of epoxy resin was applied to prepare the beam tension face. A layer of epoxy resin mixture was pasted on the wooden mold and jute fiber was pasted into the mold. Then, a layer of epoxy resin was employed over the first layer of jute fiber and finally the last layer of epoxy resin was added. JFRP laminates were put at room temperature to ensure a proper bonding for ultimate setting and then the laminates were separated from the wooden molds. It is worth noting that silicone grease was used as a mold releasing agent on the wooden mold surface before starting to paste the jute sub-layer and epoxy one after another. This was because it would be tough to release JFRP from the mold after exceeding the hardening period. After the setting, the molds were taken to a nearby wooden mill and JFRP was separated very slowly by a portable sawmill cutter blade with the help of experienced technicians. A layer of epoxy resin was created on the beam soffit by the same procedure. Finally, JFRP laminate was attached on the beam tension face and a regular consistent pressure was implemented on the laminate surface to eject the extra epoxy resin for ensuring the bond between epoxy, concrete and laminate. Later, the prepared strengthened beam specimens were stored at room temperature for 72 hours to ensure enough curing of epoxy resin before test day. Figure 2(b) shows the application of epoxy adhesive on JFRP laminate and tension face of the beam.

Figure 2.

(a) Formwork preparation and (b) attachment of epoxy and JFRP laminate on the surface of the specimens.

Artificial corrosion process

The artificial corrosion was stimulated by an impressed electric current through the cathode of 6 mm stainless steel bar, while two tension-steel bars worked as anode as shown in Figure 3. To obtain the density of the current, the sum of impressed current can be divided by the surface vicinity of part of the rebars cage that is immersed into the salted water in concrete (Al-Saidy et al., 2010). A DC power source was created to apply the desired impressed constant current where a direct line of 220 V AC was used. A step-down transformer was used to convert the voltage which was brought down to 12 V. Also, 12 V AC was converted into 12 V DC to incorporate a bridge rectifier function. The DC was controlled through an IC LM 317 to control the amount of current passing through that line. The beams with stainless-steel rebars were attached to the power supply in a parallel circuit to determine the performance under a corrosive environmental condition. A consistent current of 375 mA was induced for a duration of 4 and 8 days to attain the predicted mass loss of 7.5% and 15% respectively in flexural bars of the specimens. To estimate the mass loss, the following equation can be used according to Faraday’s law.

Δ m = \frac{MIt}{zF}

(1)

where $Δ m$ = mass of rebar consumed, M = metal atomic weight, I = current, t = time, z = ionic charge, and F = constant of Faraday.

Figure 3.

Artificial corrosion setup.

Instrumentation and test procedure

Schematic experimental setup of RC beam is presented in Figure 4(a). All beams were tested under a mid-span two-point loading employing two roller position which ultimately made the arrangement four-point as illustrated in Figure 4(b). The monotonic load was applied in a regular interval by a hydraulic jack with load cell through a steel I-section on the top of the beam. Three displacement dial gauges were placed to monitor the vertical deflection at three locations, mid-span and both side quarter-spans from support. The loading values with the corresponding deflections were recorded for each specimen until the failure of the beam. Initially, cracks were detected on the beam surface with a magnifying glass for the corresponding loads. Afterwards, the cracks were observed visually and marked with each increment of the load until the failure of the specimen.

Figure 4.

(a) Schematic view and (b) experimental setup of the tested beam specimens.

Results and discussion

Summary of key test data for load-deflection response

Table 2 summarizes the key performance of tested control and strengthened beams in form of first cracking load, yield load, ultimate load, deflection at yielding and ultimate load and failure pattern. According to ASTM C1018-97 (1997), the first crack is the point on the load-deflection curve at which the form of the curve first becomes nonlinear or little sudden drop from its linear ascending nature at the very beginning of the loading which approximates the onset of cracking in the concrete matrix and its corresponding load is defined first cracking load. In other words, when the applied stress in the tension soffit of the beam exceeded the tensile stress capacity of concrete then the small hairline cracks appeared on the maximum moment zone of the beam. From Figures 6 and 7, the first cracking load can be determined from the corresponding load of that first nonlinear point at very earlier in the load-deflection curve. It was observed from the test results that JFRP strengthened beams with and without the corrosion action exhibited higher first cracking load than the control beams. This enhancement of first cracking load in strengthened beams is attributed to the upgradation of tensile resistance in strong composite action and maximum shear stress developed due to the application of JFRP and epoxy adhesive at the concrete surface.

Table 2.

Test results of beam specimens.

Specimen designation	CB-0	JFB-0	CB-7.5	JFB-7.5	CB-15	JFB-15
$P_{cr}$ (kN)	17.2	27.5	15.2	25.0	10.0	22.1
$P_{y}$ (kN)	50.0	70.3	40.1	55.1	30.0	52.0
$Δ_{y}$ (mm)	2.51	3.05	1.72	2.54	1.67	2.08
$P_{u}$ (kN)	75.0	105.0	55.1	80.3	44.9	70.2
$Δ_{u}$ (mm)	8.85	7.94	6.35	7.89	8.69	5.92
Failure mode	a	c	c	c	a	b

where $P_{cr}$ = First cracking load, $P_{y}$ = yield load, $Δ_{y}$ = deflection at yield load, $P_{u}$ = ultimate load, and $Δ_{u}$ = deflection at ultimate load.

Failure mode: (a) Failure initiated in flexure with spalling of compression concrete, (b) failure initiated with spalling of crushing compression concrete and yielding of flexural steel with rupture of JFRP, and (c) failure initiated with crack widening enlargement due to shear failure of concrete.

Crack propagation and failure nature

The first crack for control beam CB-0 formed as flexural crack along the mid-span of the beam at 17.2 kN load. A similar observation of initiation of first crack was stated in the study performed by Chen et al. (2018) and Al-Hammoud et al. (2011). At final failure, flexural cracks formed at high moment region, whereas web shear cracks initiated at high shear zone near the left and right support of the beam. Beam CB-0 failed due to the spalling of compression concrete as shown in Figure 5(a). The first crack for strengthened beam JFB-0 occurred at 27.5 kN load. It formed as flexural crack at the tensile zone of mid-section, which indicated that this beam was deficient at flexural loading in nature. The final cracks as shown in Figure 5(b) formed at high shear region beside the left and right support through the shear failure of concrete. The first crack for beam CB-7.5 formed as flexural crack at 15 kN load. The final cracks as shown in Figure 5(c) mainly spread from the support to high shear region as web shear crack, which propagated from tension to compression zone. For beam JFB-7.5, the first crack occurred as flexural crack at 25 kN load. The failure was caused by shear failure of concrete at the vicinity of high shear zone of the beam as presented in Figure 5(d). This behavior is due to the combination of shear stress and longitudinal flexural stress developed in this strengthened beam. The first crack was seen at 10 kN load as flexural crack for beam CB-15. The final failure of this beam was caused as a result of flexural crushing of compression concrete. The failure mode of the beam is depicted in Figure 5(e) which demonstrated the ductile behavior and exhibited large deflection before final failure. Beam JFB-15 showed the first crack as flexural crack at high moment region, which appeared at 22.1 kN load. The final failure of this beam is shown in Figure 5(f).

Figure 5.

Crack pattern for beams: (a) CB-0, (b) JFB-0, (c) CB-7.5, (d) JFB-7.5, (e) CB-15, and (f) JFB-15.

It can be observed from the load-deflection response and failure modes of JFRP strengthened beams that most of the strengthened beams exhibited the quiet similar response of control beams in the initial linear ascending part of the curves. In the very beginning of the elastic linear part of the curve, when the applied load exceeded the tensile stress of concrete then the small hairline cracks appeared on the beam surface that is termed as first crack and the study revealed that it always appeared in the middle third span of the strengthened beams. Afterwards, when the internal steel reached the yield condition then the additional load was carried by JFRP systems. During this yield stage, the steel reinforcement took a higher load with the combination of JFRP until the ultimate capacity of the strengthened beams reached. This behavior suggested that steel and concrete bonding at the specific tension zone of the beam referred as tension stiffening played a vital role with the combined action of JFRP with the epoxy adhesive that led to stronger bonding which enhanced the ultimate load-carrying capacity of the strengthened beams. It should be noted that at this yield condition, most of the cracks that appeared on the beam surface exceeded the neutral axis of the beam section. At ultimate loading, the little amount of debonding of JFRP laminate took place at both ends of high shear zone induced by web-shear crack for beams JFB-0 and JFB-7.5. This predominant little debonding behavior was attributed to the weak behavior of composite action between the epoxy adhesive and JFRP laminate as well as soft and less stiffened nature of jute fiber. On the other hand, beam JFB-15 failed with the initiation of spalling of compression concrete following the steel yielding then the final rupture of JFRP occurred at the tension soffit of the beam with following the debonding incident. In the ultimate stage of loading, the bonding between the steel, tension concrete, JFRP with epoxy adhesive started to weaken the interfacial bond capacity followed by the steel yielding and then debonding with final rupture of JFRP laminate.

Effects of corrosion on load-mid-span deflection characteristics of control beams

The load-deflection behavior of control beams at mid-span with different level of corrosion contribution is illustrated in Figure 6(a). It can be noted that the ultimate load, yield load and first cracking load of all control beams decreased with the increment of corrosion degree. The decrease of the ultimate load was 27% and 40% in beams CB-7.5 and CB-15 respectively compared to beam CB-0. Al-Saidy et al. (2010) and Shannag and Al-Ateek (2006) observed a similar finding, where the ultimate strength of beams reduced with the increase of corrosion level. Similarly, yielding load decreased by 20% and 40% in beams CB-7.5 and CB-15 respectively compared to beam CB-0, while it was 12% and 41% decrement for first cracking load. The deflection at yield loading also reduced with the increment of corrosion level and beams CB-7.5 and CB-15 showed a reduction of 31% and 33% in yield loading deflection than beam CB-0. On the contrary, deflection at ultimate loading decreased in beam CB-7.5 than beam CB-0; however, it was almost equal in beams CB-0 and CB-15. For beams CB-7.5 and CB-15, the decrease of deflection at ultimate loading was 28% and 2% respectively compared to beam CB-0. Furthermore, a significant difference in first cracking, yield and ultimate load, as well as deflection at yield and ultimate loading, was observed between beams CB-7.5 and CB-15. When the corrosion level increased from 7.5% to 15%; the first cracking, yield and ultimate load reduced, however, the deflection at ultimate loading increased. The reduction in beam CB-15 for first cracking, yielding and ultimate load was 33%, 25%, and 18% respectively than beam CB-7.5, while it was 37% increment for ultimate deflection. The main reason for the decrement in ultimate loading and increment in ultimate deflection was due to the corrosion action in mass loss of tension rebars in corroded control beams. As the tensile bars lost their mass, their centroid changed based on their weights and they were unable to achieve the ultimate capacity that they took before the corrosion.

Figure 6.

Load–deflection curve of: (a) control beams and (b) JFRP strengthened beams at 0%, 7.5%, and 15% corrosion level.

Effects of corrosion and strengthening scheme on load-mid-span deflection characteristics of strengthened beams

Figure 6(b) presents the load-deflection response at mid-span of JFRP strengthened beams in the action of different corrosion levels. When the cross-section of bars reduced due to the corrosion, the first cracking, yield and ultimate load also reduced for all strengthened beams. The ultimate load reduced by 24% and 33% in beams JFB-7.5 and JFB-15 respectively compared to beam JFB-0, while it was 13% and 20% for first cracking load. For yield load, the decrement was 22% and 17% in beams JFB-7.5 and JFB-15 respectively than beam JFB-0. The similar findings were observed in a study reported by Elghazy et al. (2018), where the reduction of ultimate and yield strength was 10% and 15% for a mass loss of 22.5% due to the corrosion effect. It was concluded that this behavior was due to the losing of lugs of corroded reinforcing bars and increasing of mass loss in the bars (Elghazy et al., 2018). The deflection at yield and ultimate loading also declined, when the cross-sectional area of steels reduced and the degree of corrosion increased more. Beams JFB-7.5 and JFB-15 exhibited a decrement of 1% and 25% deflection at ultimate loading than beam JFB-0, whereas it was 17% and 32% for deflection at yield loading. On the contrary, the decrease of deflection at ultimate loading was 25% in beam JFB-15 compared to beam JFB-7.5. Notable variation was observed for first cracking, yield and ultimate loading, when the degree of corrosion increased. The decline in ultimate loading was 25% when the corrosion level increased from 7.5% to 15%. The reduction for first cracking load in beam JFB-15 was around 12% than beam JFB-7.5, while it was 6% decrement for yield loading. The yield and ultimate load in strengthened beams decreased gradually with the increment of corrosion level due to the reduction of mass in tensile bars. However, JFRP laminates used its tensile strength for the enhancement of ultimate strength in all strengthened beams.

Comparison of load-deflection characteristics of control and strengthened beam

Control and strengthened beam (0% corrosion level)

The load-mid span deflection response of beam CB-0 and JFB-0 at 0% corrosion is presented in Figure 7(a). It can be seen that the stiffness of beam JFB-0 was higher than beam CB-0 by a small-scale amount. For control beam, it is well known that tension steel and compression concrete carried most of the load due to the flexural loading in RC beams. However, numerous analytical and finite element analysis revealed that not only compression concrete but also the surrounding tension-steel zone form a bond and contribute to upgrading the load-carrying capacity of the beam. For strengthened beam, the attached JFRP laminate with epoxy adhesive acted as a stronger stiffed area on the tension zone of the beam together with the tension stiffening zone concrete and reinforcement. The composite action of concrete in tension zone, reinforcement and JFRP laminate with epoxy adhesive resulted in much stronger bond in the new tension stiffening zone and contributed to the increase of stiffness in the strengthened beam than the control beam. The first cracking, yield and ultimate load increased by 58%, 40%, and 40% respectively in beam JFB-0 than beam CB-0. In another study, the increment of first cracking, yield and ultimate load of repaired beam was obtained 58%, 32%, and 52% respectively than that of control beam (Maaddawy and Soudki, 2005). The ultimate load increased substantially in beam JFB-0 than beam CB-0, however, deflection at ultimate loading decreased by 10% in this case. This behavior of strength increment in repaired beam compared to control beam is also the same as that in the study by Sobuz et al. (2012). The main reason for the significant improvement of ultimate load in strengthened beam was due to the addition of JFRP laminates on the tension face of the beam. Following the internal yielding of steel, the supplementary tensile strength was gained by JFRP, which contributed to the increase of ultimate loading capacity. Other researchers (Maaddawy and Soudki, 2005) also noted the identical behavior, in which CFRP laminate in strengthened uncorroded specimens, significantly contributed to the strength gaining compared with that of control beams.

Figure 7.

Load–deflection response of control and strengthened beam at: (a) 0%, (b) 7.5%, and (c) 15% corrosion level.

Control and strengthened beam (7.5% corrosion level)

Figure 7(b) shows the load-mid span deflection response of 7.5% corroded control beam CB-7.5 and strengthened beam JFB-7.5. The ultimate load, yield load, first cracking load and ultimate deflection in beam JFB-7.5 increased by about 45%, 38%, 66%, and 24% respectively compared to beam CB-7.5. Other researchers (Al-Saidy et al., 2010) also found the similar response, in which it was mentioned that the percentage increment of ultimate load and yield load was 34.2% and 23.9% respectively for the beam having 5% corrosion than control beam. In the present study, the stiffness of these two beams showed similar pattern up to 40 kN load, however, the first cracking load and yield load was higher in beam JFB-7.5 than beam CB-7.5. It is noticed that beam JFB-7.5 gained a significant amount of ultimate load and deflection at ultimate loading compared to beam CB-7.5. This is attributed to the bonding of JFRP sheet with the tension side of the beam, which was mainly responsible for the substantial increase of flexural capacity with significant deformation of the beam in the action of 7.5% corrosion.

Control and strengthened beam (15% corrosion level)

The flexural response of 15% corroded beam CB-15 and JFB-15 is shown in Figure 7(c). The increase of ultimate, yield and first cracking load in strengthened beam JFB-15 was 55%, 73% and 120% respectively than control beam CB-15. Similar findings were reported by several researchers (Al-Saidy et al., 2010; Elghazy et al., 2018; Sobuz et al., 2012). The stiffness of both curves was nearly equal up to 10 kN load and then it was higher for beam JFB-15 due to JFRP addition on the soffit of the beam. On the other side, deflection at ultimate loading in beam JFB-15 decreased by a significant amount that was 32% than beam CB-15, which followed the similar observation made by other researchers (Al-Saidy et al., 2010; Sobuz et al., 2012). The significant change of ultimate load was obtained in beam JFB-15 than beam CB-15. As a matter of fact, the effect of JFRP laminate with the combination of epoxy adhesive was more significant for beam JFB-15 considering the strong composite substrate. This is due to the fact that beam JFB-15 failed due to rupture of JFRP which formed after the yielding of flexural steel and spalling of crushing compression concrete. On the other hand, beams JFB-0 and JFB-7.5 failed due to shear failure of concrete with the widening of cracks. This finding is in line with Al-Saidy et al.’s (2010) study which investigated the structural performance of corroded RC beams strengthened with CFRP sheets. They reported that ultimate strength of strengthened beams at 5% and 10% corrosion were 23.9% and 9.3% higher than control beam, while it was 2% lower for strengthened beam at 15% corrosion level. This was mainly due to premature debonding failure in strengthened beam specimen at 15% corrosion level. Another reason is that epoxy adhesive itself exhibits outstanding corrosion resistance due to its excellent chemical properties and adhesion properties.

Ductility characteristics

Ductility is used to describe the ability of any material to sustain inelastic deformation before fracture. Ductility in RC member can be generally measured by ductility ratio or index. A structural member having higher ductility index is capable of undergoing inelastic displacement before failure (Teo et al., 2006). The deformation-based approach is frequently utilized and followed by the deformation difference between ultimate and service condition (Alengaram et al., 2008; Teo et al., 2006; Wang and Belarbi, 2011). The displacement ductility index can be determined according to Equation (2).

μ_{d δ} = \frac{δ_{u}}{δ_{y}}

(2)

where $μ_{d δ}$ = ductility ratio, $δ_{u}$ = mid-span ultimate deflection, and $δ_{y}$ = mid-span deflection at time of yielding of tensile reinforcement.

The displacement ductility indexes for all tested beams were determined using Table 2. These values were obtained 3.53, 3.69, and 5.2 for control beams at 0%, 7.5%, and 15% corrosion level respectively, while they were 2.6, 3.11, and 2.85 for JFRP strengthened beams. It can be noticed that control beams achieved more ductility index than strengthened beams in all corrosion levels. This finding is similar to that shown by another study (Al-Saidy et al., 2010), where the ductility of all repaired beams was lower concerning control beams. Strengthened beams at 0%, 7.5%, and 15% corrosion level showed a decrease in ductility index by 26%, 16%, and 45% respectively than the corresponding control beams. This behavior is due to rupture of JFRP and shear failure as well as the addition of JFRP layer on the soffit of the beam.

Analytical investigation

Flexural response of RC beam specimens

The analytical approach was followed according to code BS 8110-1 (1997) for evaluation of flexural response due to the addition of JFRP laminate to the concrete surface. The stress and strain profile of RC beam section repaired with JFRP at soffit level of the beam is presented in Figure 8. The analytical cracking moment $M_{cr}$ can be measured based on the following equation:

M_{cr} = \frac{f_{r} I_{g}}{y_{t}}

(3)

Figure 8.

Stress-strain diagram of RC beam section.

where $f_{r}$ = rupture strength of concrete, I = second moment of inertia, and $y_{t}$ = distance account from the neutral axis to outer face of the beam.

The following equations from the equilibrium of section of stress-strain diagram can be applied for the calculation of ultimate flexural strength of RC control beam.

F_{st} = F_{cc} + F_{sc}

\Rightarrow 0.95 A_{s} f_{y} = 0.45 {f_{c}}^{'} bs + 0.95 A_{s} f_{y}

(4)

The yield condition of tension steel can be checked by the subsequent equation:

\frac{0.0035}{x} = \frac{ε_{st}}{d - x}

Then, the strain of steel can be obtained as:

ε_{st} = (\frac{d - x}{x}) * 0.0035 (For the tension steel yield condition)

(5)

ε_{sc} = (\frac{x - d^{'}}{x}) * 0.0035 (For compression steel yield condition)

(6)

The corresponding stress and force in compression steel $(f_{sc})$ can be measured by using the following equation:

f_{sc} = ε_{sc} * E_{sc}

\Rightarrow f_{sc} = (\frac{x - d^{'}}{x}) * 0.0035 * E_{sc}

(7)

Then F_{sc} = (\frac{x - d^{'}}{x}) * 0.0035 * E_{sc} * A_{sc}

(8)

Taking moment at top of the section,

M_{ult} = F_{st} * d - F_{sc} * d^{'} - F_{cc} * 0.45 * x

(9)

The ultimate flexural capacity of JFRP strengthened beam can be determined by using the following equations from the equilibrium of section of stress-strain diagram.

F_{st} + F_{f} = F_{cc} + F_{sc}

\Rightarrow 0.95 A_{s} f_{y} + 0.95 A_{f} f_{y} = 0.45 {f_{c}}^{'} bs + 0.95 A_{s} f_{y}

(10)

Then, the strain of steel can be calculated as the same way from Equation (5) and (6). The following equation can be used to check JFRP reinforcement yield condition:

\frac{0.0035}{x} = \frac{ε_{f}}{d_{f} - x}

\Rightarrow ε_{f} = (\frac{d_{f} - x}{x}) * 0.0035

(11)

If JFRP reinforcement does not yield, then we can consider

F_{st} = 0.95 A_{s} f_{y}

(12)

F_{cc} = 0.45 {f_{c}}^{'} bs

(13)

The corresponding stress and force in compression steel $(f_{sc})$ can be obtained from equation (14) and (15).

f_{sc} = ε_{sc} * E_{sc}

\Rightarrow f_{sc} = (\frac{x - d^{'}}{x}) * 0.0035 * E_{sc}

(14)

Then F_{sc} = (\frac{x - d^{'}}{x}) * 0.0035 * E_{sc} * A_{sc}

(15)

From the proportions of strain distribution diagram, JFRP laminate strain is as follows:

\frac{0.0035}{x} = \frac{ε_{f}}{d_{f} - x}

\Rightarrow ε_{f} = (\frac{d_{f} - x}{x}) * 0.0035

(16)

The corresponding stress and force in fiber laminate $f_{f}$ is as follows:

f_{f} = ε_{f} * E_{f}

\Rightarrow f_{f} = (\frac{d_{f} - x}{x}) * 0.0035 * E_{f}

(17)

Then F_{f} = (\frac{d_{f} - x}{x}) * 0.0035 * E_{f} * A_{f}

(18)

Then, we can apply Equation (10) of equilibrium of section from stress-strain diagram and take moment at top of the section,

M_{ult} = F_{st} * d + F_{f} * d_{f} - F_{sc} * d^{'} - F_{cc} * 0.45 * x

(19)

Comparison between experimental and analytical investigation

Table 3 summarizes the experimental and analytical results of ultimate load, first cracking load and ultimate deflection of control and JFRP strengthened beam at 0% corrosion level. It can be noticed that there was a slight variation between analytical and experimental results. The analytical first cracking load predicted lower value than the experimental one for control beam; besides, the ultimate capacity decreased in this case. For control beam, the percentage increment of first cracking load was approximately 9%, whereas ultimate capacity decreased by about 15% between experimental and analytical value. The theoretical first cracking load of strengthened beam exhibited a lower value than the experimental observation. The analytical ultimate capacity of strengthened beam was 93.35 kN with a decrease of 11% than the experimental value of 105 kN. Additionally, ultimate deflection of control and repaired beam from theoretical prediction showed lower value than the tested beams. Furthermore, the ultimate capacity of strengthened beam showed higher value than the corresponding theoretical capacity, which leads to the full composite action and development of high shear capacity-achieving its highest shear stress before any notable yielding of JFRP. It can be noted that first cracking load provided a very reliable agreement from its theoretical value, whereas the ultimate capacity and deflection showed lower prediction with experimental tested results.

Table 3.

Experimental and theoretical results of beam specimens.

Specimen designation	Experimental results			Analytical results			$\frac{P_{u (\exp)}}{P_{u (a n l)}}$
Specimen designation	$P_{cr}$ (kN)	$P_{u}$ (kN)	$Δ_{u}$ (mm)	$P_{cr}$ (kN)	$P_{u}$ (kN)	$Δ_{u}$ (mm)	$\frac{P_{u (\exp)}}{P_{u (a n l)}}$
CB-0	17.2	75.0	8.85	18.75	63.82	6.21	1.18
JFB-0	27.5	105.0	7.94	23.26	93.35	6.30	1.12

where $P_{cr}$ = first cracking load, $P_{u}$ = ultimate load, and $Δ_{u}$ = deflection at ultimate load.

Comparison between theoretical and actual mass loss

Table 4 lists the theoretical and actual mass loss of beam specimens during the corrosion stage. The calculation of mass loss from Faraday’s law as shown in Table 4 demonstrated the occurrence of uniform corrosion. It can be seen that there was a good agreement achieved between theoretical and experimental mass loss. The theoretical value evaluated higher percentages compared to the actual mass loss for the specified beam specimens. This change in theoretical and actual mass loss attributed to the change of environmental conditions where some factors including the reactivity of metal, presence of salt, presence of oxygen affected the corrosion stage. The percentage deviation of mass loss for beam CB-7.5 was 11% between theoretical and actual result, whereas it was 9% for beam JFB-7.5. On the other hand, the percentage deviation of mass loss for beam CB-15 and JFB-15 was 10% and 12% respectively between theoretical and actual value. According to Elghazy et al. (2018), the calculations of mass loss are greatly varied due to the irregular manner of corrosion.

Table 4.

Comparison between theoretical and actual mass loss.

Specimen	CB-7.5	JFB-7.5	CB-15	JFB-15
Theoretical	7.5%	7.5%	15%	15%
Actual	6.7%	6.8%	13.5%	13.2%

Conclusion

This study presented an investigation on flexural performance of control and JFRP strengthened RC beams in normal and corrosion stage along with the analytical calculations. Based on the findings of this study, the following conclusions can be drawn:

JFRP strengthened RC beams exhibited higher ultimate load, yield load and first cracking load compared to control beams with and without the effect of corrosion. It suggests that the strength loss in corroded beams restored due to the addition of JFRP laminate on the tension face of the beams. This behavior is attributed to the contribution of composite action of epoxy adhesive, concrete and JFRP laminate, which leads to achieving the full flexural strength of the beams.

Regarding the effects of strengthening and varying concentration of corrosion level, the improvement of ultimate loading in strengthened beams JFB-0, JFB-7.5, and JFB-15 was achieved about 40%, 45.7%, and 56.3% compared to control beams CB-0, CB-7.5, and CB-15 respectively. The attachment of JFRP laminate on the soffit of the beam exhibited better upgrading in flexural strength at higher corrosion level. This is due to the fact that higher shear stress and composite action developed substantially for each sub-layer of JFRP at the tension face for beam JFB-15 which resisted the corrosion phenomenon.

The use of JFRP laminate in strengthened beams exhibited lower ultimate deflection compared to control beams. The strengthening technique of JFRP increased the strength and stiffness of the beams but resulted in a loss of ductility under flexural loading than control beams. This behavior ensured that JFRP laminate and concrete composite action with premature debonding failure took place on the soffit of the beams during the plastic deformation phase.

The theoretical results showed close agreement with the experimental values where theoretical calculations provided a reasonable prediction of first cracking load, ultimate load, ultimate deflection, and mass loss compared to the testing results.

In general, this study investigated the flexural behavior of natural JFRP strengthened RC beams exposed to the normal and corrosive environment. However, this study needs to be further studied and extended with large scale experimentation considering the number of layers, different end anchorage system around the beam to prevent the debonding phenomenon, angle strip throughout the whole length to prevent the shear failure and more varying concentration of corrosion rate. Moreover, further work is required in an aggressive environment incorporating the durability aspects and time-dependent response of beams under sustained loads with jute fiber.

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Abu Sayed Mohammad Akid

Md. Habibur Rahman Sobuz

References

Alengaram

Jumaat

Mahmud

(2008) Ductility behaviour of reinforced palm kernel shell concrete beams. European Journal of Scientific Research 23: 406–420.

Al-Hammoud

Soudki

Topper

(2011) Fatigue flexural behavior of corroded reinforced concrete beams repaired with CFRP sheets. Journal of Composites for Construction 15: 42–51.

Al-Saidy

Al-Harthy

Al-Jabri

, et al. (2010) Structural performance of corroded RC beams repaired with CFRP sheets. Composite Structures 92: 1931–1938.

ASTM (1997) ASTM C1018-97: Standard Test Method for Flexural Toughness and First-Crack Strength of Fiber-Reinforced Concrete (Using Beam With Third-Point Loading). West Conshohocken, PA: ASTM International.

ASTM (2001) ASTM C150: Standard Specification for Portland Cement. Philadelphia, PA: ASTM International.

ASTM (2002) ASTM C685: Standard Specification for Concrete Made by Volumetric Batching and Continuous Mixing. Annual Book of American Society for Testing Materials, vol. 4.02. West Conshohocken, PA: ASTM International.

ASTM (2003) ASTM C143-03: Standard Test Method for Slump of Hydraulic-Cement Concrete. West Conshohocken, PA: ASTM International.

ASTM (2010) ASTM C39: Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens. West Conshohocken, PA: ASTM International.

ASTM (2011) ASTM C496: Standard Test Method for Splitting Tensile Strength of Cylindrical Concrete Specimens. West Conshohocken, PA: ASTM International.

10.

Azad

Ahmad

Al-Gohi

BHA

(2010) Flexural strength of corroded reinforced concrete beams. Magazine of Concrete Research 62: 405–414.

11.

Baggio

Soudki

Noël

(2014) Strengthening of shear critical RC beams with various FRP systems. Construction and Building Materials 66: 634–644.

12.

Barris

Sala

Gómez

, et al. (2020) Flexural behaviour of FRP reinforced concrete beams strengthened with NSM CFRP strips. Composite Structures 241: 112059.

13.

BS 8110-1 (1997) Structural Use of Concrete – Part 1: Code of Practice for Design and Construction. London: British Standards Institute.

14.

Chen

Yang

, et al. (2020) Eco-friendly and mechanically reliable alternative to synthetic FRP in externally bonded strengthening of RC beams: Natural FRP. Composite Structures 241: 112081.

15.

Chen

Pham

Sichembe

, et al. (2018) Experimental study of flexural behaviour of RC beams strengthened by longitudinal and U-shaped basalt FRP sheet. Composites Part B: Engineering 134: 114–126.

16.

Choi

Utui

Kim

(2013) Experimental and analytical investigations on debonding of hybrid FRPs for flexural strengthening of RC beams. Composites Part B: Engineering 45: 248–256.

17.

Davalos

Parish

Chen

, et al. (2013) Effect of anchoring schemes for beams aged by accelerated corrosion and strengthened with carbon fibre-reinforced polymer. Structure and Infrastructure Engineering 9: 229–241.

18.

Dong

Wang

Guan

(2013) Structural behaviour of RC beams with external flexural and flexural–shear strengthening by FRP sheets. Composites Part B: Engineering 44: 604–612.

19.

Elghazy

El Refai

Ebead

, et al. (2018) Post-repair flexural performance of corrosion-damaged beams rehabilitated with fabric-reinforced cementitious matrix (FRCM). Construction and Building Materials 166: 732–744.

20.

Fang

Lundgren

Chen

, et al. (2004) Corrosion influence on bond in reinforced concrete. Cement and Concrete Research 34: 2159–2167.

21.

Hawileh

Rasheed

Abdalla

, et al. (2014) Behavior of reinforced concrete beams strengthened with externally bonded hybrid fiber reinforced polymer systems. Materials & Design 53: 972–982.

22.

Huang

Yan

, et al. (2016) Reinforced concrete beams strengthened with externally bonded natural flax FRP plates. Composites Part B: Engineering 91: 569–578.

23.

Kim

Shin

(2011) Flexural behavior of reinforced concrete (RC) beams retrofitted with hybrid fiber reinforced polymers (FRPs) under sustaining loads. Composite Structures 93: 802–811.

24.

Maaddawy

Soudki

(2005) Carbon-fiber-reinforced polymer repair to extend service life of corroded reinforced concrete beams. Journal of Composites for Construction 9: 187–194.

25.

Mancini

Tondolo

Iuliano

, et al. (2014) Local reinforcing bar damage in rc members due to accelerated corrosion and loading. Construction and Building Materials 69: 116–123.

26.

Pham

Doan

Hadi

MNS

(2013) Strengthening square reinforced concrete columns by circularisation and FRP confinement. Construction and Building Materials 49: 490–499.

27.

Rahimi

Hutchinson

(2001) Concrete beams strengthened with externally bonded FRP plates. Journal of Composites for Construction 5: 44–56.

28.

Šavija

Luković

Pacheco

, et al. (2013) Cracking of the concrete cover due to reinforcement corrosion: A two-dimensional lattice model study. Construction and Building Materials 44: 626–638.

29.

Sen

Jagannatha Reddy

(2013) Strengthening of RC beams in flexure using natural jute fibre textile reinforced composite system and its comparative study with CFRP and GFRP strengthening systems. International Journal of Sustainable Built Environment 2: 41–55.

30.

Shannag

Al-Ateek

(2006) Flexural behavior of strengthened concrete beams with corroding reinforcement. Construction and Building Materials 20: 834–840.

31.

Sobuz

Ahmed

Sutan

, et al. (2012) Bending and time-dependent responses of RC beams strengthened with bonded carbon fiber composite laminates. Construction and Building Materials 29: 597–611.

32.

Soudki

El-Salakawy

Craig

(2007) Behavior of CFRP strengthened reinforced concrete beams in corrosive environment. Journal of Composites for Construction 11: 291–298.

33.

Teo

Mannan

Kurian

(2006) Flexural behaviour of reinforced lightweight concrete beams made with oil palm shell (OPS). Journal of Advanced Concrete Technology 4: 459–468.

34.

Toutanji

Zhao

Zhang

(2006) Flexural behavior of reinforced concrete beams externally strengthened with CFRP sheets bonded with an inorganic matrix. Engineering Structures 28: 557–566.

35.

Valluzzi

Modena

de Felice

(2014) Current practice and open issues in strengthening historical buildings with composites. Materials and Structures 47: 1971–1985.

36.

Wang

Belarbi

(2011) Ductility characteristics of fiber-reinforced-concrete beams reinforced with FRP rebars. Construction and Building Materials 25: 2391–2401.

37.

Zhang

Jin

, et al. (2019) Effect of corroded tension reinforcements on flexural performance of reinforced recycled aggregate concrete beams strengthened with CFRP. Composites Part B: Engineering 162: 589–599.