Characterization and application of FRCM as a strengthening material for shear-critical RC beams

This paper investigates the effectiveness of fabric reinforced cementitious matrix (FRCM) systems in shear-strengthening of reinforced concrete beams. Three types of FRCM systems were considered, namely, polyparaphenylene benzobisoxazole (PBO)-FRCM, Carbon-FRCM, and GlassFRCM. At first, tensile characterization test was performed on 15 FRCM coupons with the aim of identifying the tensile properties of the FRCM systems adopted. After that, seven shear-critical RC beams were tested under three-point loading, with the consideration of two test parameters: (a) FRCM material (glass/carbon/PBO); and (b) strengthening configuration (full/intermittent). The study results revealed the use of FRCM as a strengthening material to achieve a considerable improvement in the structural capacity of shear-critical RC beams. The average gain in the shear capacity of the FRCMstrengthened beams was 57%. The beam specimens strengthened with carbon-FRCM showed the highest improvement as compared to those strengthened with glassand PBO-FRCM systems. As intuitively expected, the shear capacity improvement achieved with the full-length strengthening systems was generally higher than that with the intermittent counterparts.


Introduction
Structural damage is normally encountered in reinforced concrete (RC) structures subject to harsh environment or unexpected extra loads [1]. Consequently, different strengthening techniques of RC members were proposed to enhance their structural capacity such as ferrocement [2] and fibre reinforced polymer (FRP) [3][4][5][6][7][8]. Latterly, fabric reinforced cementitous matrix (FRCM) strengthening systems have emerged as a viable solution for structural rehabilitation and repair. FRCM consists of dry fibres embedded in an inorganic matrix, which is externally applied on the surface for the aim of strengthening RC and masonry structures [9,10].
FRCM systems have been successfully applied in the previous research for strengthening RC members such as slabs [11], columns [12][13][14], and beams critical in flexure [15,16] or in shear [17][18][19][20]. In these studies, a noticeable improvement was generally reported in the loading capacity of the FRCM-strengthened members as compared to that of the non-strengthened benchmarks. Nonetheless, it is well-agreed that a proper understanding of the tensile characteristics of the FRCM composite is required to complete the structural design or numerical modeling of the FRCM strengthening system [21][22][23].
In view of this, the current paper has a two-fold objective: (a) to investigate the effectiveness of FRCM systems in strengthening of shear-critical RC beams; (b) to enrich the literature with contextual data pertinent to the tensile characteristics of commercially available FRCM systems, which can be used for future applications in structural rehabilitation and repair. In this effort, the authors have considered three types of FRCM systems: Glass (G)-FRCM, Carbon (C)-FRCM, and polyparaphenylene benzobisoxazole (PBO)-FRCM. At first, tensile characterization test was performed on 15 FRCM test coupons to obtain the tensile properties of the utilized FRCM systems. After that, a total of seven RC beams critical in shear were tested under three-point loading to assess the efficacy of FRCM strengthening systems, while considering the FRCM type and strengthening configuration as test variables.
In the beginning, the experimental program will be explained in terms of materials description, test specimens, and test setup. After that, results and observations of this experimental study will be presented and discussed, mainly for the tensile characteristics of the FRCM systems as well as the structural performance of the strengthened RC beams. MPa for the yield stress, 0.266% for the yield strain, and 224 GPa for the modulus of elasticity.
Three types of FRCM systems were considered, which are: Glass-FRCM [25], Carbon-FRCM [26], and PBO-FRCM [27]. To prepare the FRCM composite, the textile was embedded in its corresponding mortar in accordance with the manufacturers' recommendations. The amount of mixing water needed per 25 kg of mortar was 5 litres for the Glass-FRCM and 7 litres for the Carbon-and PBO-FRCM systems. The mechanical properties for each textile type as provided by the manufacturers are listed in Table 1, including the textile area per unit width ( ), ultimate strain ( ), elastic modulus ( ), and tensile strength ( ), along with the compressive strength of their associated mortars ( ).

FRCM tensile coupons
In accordance with AC434 [28], five tensile coupons were prepared for each FRCM system. A panel of 410 mm in length, 250 mm in width, and 10 mm in thickness was first cast on a flat wooden mould. After being cured for 28 days, each panel was cut to five number of 410-mm long, 50-mm wide, and 10-mm thick coupons. Each FRCM tensile coupon encompassed a single textile layer, of which the longitudinal direction was parallel to the coupon's length.

RC beam specimens
Seven 2100-mm long, 150-mm wide, and 330-mm deep reinforced concrete beams were tested. Bars of 16 mm in diameter were used as flexural reinforcement, while 8mm diameter stirrups spaced at 100 mm were used as shear reinforcement in the non-critical regime. The beam specimens had no stirrups within the 550-mm long critical span as illustrated in Figure 1. One of the beam specimens was kept non-strengthened as a benchmark, while the rest were FRCM-strengthened within the critical shear span with different types and configurations ( Figure 1). A 34mm clear concrete cover was typically considered, leading to an effective depth of 280 mm in RC beam specimens. Table 2 presents the test matrix of the RC beam specimens, in which two parameters were investigated: (i) FRCM type: glass-, PBO-, and carbon-FRCM.
(ii) Strengthening configuration: the beam specimen was strengthened along the critical shear span either fully by a continuous FRCM plate or intermittently by a set of 120mm wide FRCM strips spaced at 95 mm.
The FRCM system was applied on each side of the strengthened beam and composed of (i) an internal mortar layer of 5 mm in thickness, placed after roughening the concrete surface; (ii) two layers of textile separated by a 3-mm thick intermediate mortar layer; and (iii) an external layer of 5-mm thick mortar.

Test setup for FRCM tensile characterization
The tensile characterization test setup for FRCM coupons is illustrated in Figure 2. Tensile load was uniaxially applied at a rate of 0.25 mm/min. Metal plates were attached to the specimen's ends using epoxy with a 150mm bond length. The specimen was gripped by a double clevis-type connection from one end and a single clevistype connection from the other. A clip-type displacement transducer was placed to continuously measure the extension of the test coupon. Hence, the FRCM strain can be calculated as the measured coupon's extension divided by the 200-mm extensometer length. For each load step, the axial tensile stress can be determined as the recorded load (P) divided by the FRCM equivalent area. This FRCM equivalent area is actually the average textile area per unit width (Af), provided in Table 1 for each textile type, multiplied by the specimen's width (50 mm).

Fig. 2.
Tensile characterization test setup. RC beams were tested under three-point loading ( Figure  3) using the Instron 1500HDX Static Hydraulic Universal Testing machine located in Qatar University Structures Laboratory. Loading was applied on the tested beams at a rate of 1 mm/min until failure. Strain gauges (with 5% maximum limit and 5-mm gauge length) were fixed to the longitudinal bars. The deflection was measured below the loading point at each step using a linear variable displacement transducer. Data acquisition of the measurements was done at a 1-Hz frequency range. Figure 4 shows a representative stress-strain diagram for each FRCM system. In general, the stress-strain diagram for a FRCM system is composed of two characteristic phases: (i) the prior-cracking stiff section phase represented by a steep curve; followed by (ii) a reduced-slope curve indicating the cracked section phase until failure. The initial stiffness of G-FRCM was observed to be lower than that of PBO-FRCM and C-FRCM. This can be attributed to the lower elastic modulus of glass textile as compared to that of PBO or carbon (Table 1).

Fig. 4.
Stress-strain diagrams for the FRCM systems adopted. [28], a segment was considered between two data points at stress levels of 0.6 , and 0.9 , within the cracked-section curve. The slope of the line connecting these two points represents the cracked modulus of elasticity. The mean values of for G-FRCM, PBO-FRCM, and C-FRCM are 60 GPa, 112 GPa, and 135 GPa, respectively.  The G-FRCM system exhibited a more brittle failure behavior than that observed for the C-FRCM and PBO-FRCM counterparts. This can be evidenced by the harsh drop in the stress-strain diagram of G-FRCM at failure (Figure 4). Such observation can be explained by the relatively lower tensile strength of the glass textile compared to that of PBO or carbon (Table 1), which in turn results in the premature rupture of the textile inside the G-FRCM coupons. On the other hand, the Carbon-FRCM and PBO-FRCM systems generally showed a fabric-slippage type of failure. This explains the softening in the stress-strain curve after exceeding , . All cracks were developed within the non-bonded area of the tensile coupons with no sign of slippage of the epoxy-bonded metal plates. The carbon-FRCM and PBO-FRCM systems have generally shown a higher number of cracks than that of the glass-FRCM counterpart. This can be probably attributed to the lower loading capacity of the glass-FRCM coupon than that of the carbon-FRCM and PBO-FRCM counterparts.  Table 4 presents a summary of the test results for RC beam specimens. For each specimen, the second and the third columns of Table 4 list the ultimate load carrying capacity ( ) and the gain in as compared to that of the non-strengthened benchmark, respectively. As for the deformational characteristics, the deflection value ( ), the , ⁄ ratio, and the flexural strain ( ) measured at failure are listed in the fourth, fifth, and sixth columns of Table 4, respectively.

load carrying capacity
In general, the ultimate load carrying capacity for the FRCM-strengthened beams was higher than that of the benchmark specimen, demonstrating the successful application of the FRCM strengthening systems to enhance the structural capacity of shear-critical RC beams. The average improvement of the load carrying capacity after shear-strengthening was 86.2, 39.0, 46.6% for carbon, PBO, and glass FRCM systems, respectively. Figures 5-(a), (b), and (c) show the load-deflection diagrams for the beam specimens strengthened with carbon-, PBO-, and glass-FRCM systems, respectively. The carbon-FRCM strengthening system generally showed better improvement in than that of the PBOand Glass-FRCM systems. The RC beams with full-length strengthening configuration exhibited significantly higher enhancement in as compared to those with the intermittent strengthening configuration. This indicates the importance of the strengthening material's amount in the critical region. Furthermore, it was observed that using carbon-FRCM strengthening system was more effective in increasing the load carrying capacity of RC beams as compared to those with glass-and PBO-FRCM strengthening systems.

Deformational characteristics
In general, the deflection measured at failure ( ) for the strengthened beams was higher than that of the benchmark specimen ( , = 3.25 ). The beams strengthened with carbon-FRCM systems exhibited higher values of than those measured for Glass-and PBO-FRCM counterparts, which can be an indication to a superior structural performance for the beams strengthened with carbon-FRCM systems. As an example, the C-Full specimen has a value of 7.75 mm, which is ) and G-Full ( = 6.10 ) specimens. In addition, the specimens with full-length strengthening configuration showed slightly higher values of compared to those with the intermittent strengthening. For instance, the P-Intermittent specimen has a value of 4.6 mm, which is lower than that measured for P-Full specimen ( = 5.35 ). With reference to the flexural strain, all beam specimens have failed prior to the yield of the main steel ( , < = 0.266%), with the exception of C-Full that has failed just after achieving the yield tensile strain ( , = 0.286%). The , value for the FRCMstrengthened beams was commonly higher than that of the benchmark specimen, indicating the successful application of the FRCM system in postponing the brittle shear failure and, to some extent, allowing the main steel to be a little closer to the yield region.

Conclusions
This paper has investigated the effectiveness of FRCM systems in shear-strengthening of reinforced concrete beams. Three FRCM systems were considered, namely, carbon-FRCM, glass-FRCM, and PBO-FRCM. In the first part of this paper, tensile characterization test was performed on 15 FRCM coupons to measure the uniaxial tensile properties of the adopted FRCM systems. In the second part of this effort, a total of 7 RC beams were tested under three-point loading. Two test variables were considered, namely, the FRCM material and the strengthening configuration. Based on the study results, the following conclusions have been drawn: -A typical shape was observed for the stress-strain diagrams of the FRCM tensile coupons with two distinct phases, of which the first is a steep curve indicating the non-cracked section, and the second is a gentle curve representing the cracked section. -The cracked modulus of elasticity was calculated as 60 GPa, 112 GPa, and 135 GPa for glass-FRCM, PBO-FRCM, and carbon-FRCM, respectively. -FRCM has been successfully used to improve the load carrying capacity of shear-critical RC beams. The average gain in the load carrying capacity achieved with FRCM-strengthening was 57%. -A better strength improvement was reported for the fulllength strengthening configuration than that of the intermittent counterpart. -Carbon-FRCM systems were more effecient than PBOand glass-FRCM counterparts in improving the structural performance of RC beams.
This paper was made possible by NPRP grant # NPRP 7-1720-2-641 from the Qatar National Research Fund (a member of Qatar Foundation). The findings achieved herein are solely the responsibility of the authors.