Monotonic and cyclic flexural behavior of plain concrete beams strengthened by fabric-cement based composites M. Gencoglu , Istanbul Technical University, Faculty of Civil Engineering, Department of Reinforced Concrete Structures, Ayazaga Campus, 34469 Maslak-ISTANBUL, TURKEY

B. Mobasher , Arizona State University, Department of Civil and Environmental Engineering, Main Campus, Tempe, Arizona 85287, USA

Keywords: Pultrusion, fabric, cement paste, cyclic, flexural, ductility, strengthening, concrete beam ABSTRACT: The limited strain capacity of cementitious materials makes them tension-weak, brittle, and considerably notch sensitive. Textile fabrics have been recently developed as a new class of cement based materials with considerable improvement in strength and ductility. In this study use of pultruded cement fabrics in increasing the flexural capacity of concrete subjected to flexural loads are studied as a potential strengthening method. Plain concrete beams with size of (100 × 100 × 460 mm) were produced. Samples were strengthened by using alkali resistant (AR) glass and polyethylene (PE) fabrics impregnated with cement paste. Two levels of strengthening consisting of four and eight layers fabric were used on five replicate specimens in each category. Three point bending tests were performed on a closed-loop controlled servo-hydraulic material test system. The strengthened concrete beams were subjected to monotonic and cyclic flexural loads to determine the effects of strengthening process, type, and number of fabric layers on the flexural behavior of concrete beams. The recorded measurements on the test specimens were evaluated from the viewpoint of flexural strength, ultimate deflection, and rigidity degradations of specimens. The fabric reinforced concrete composites (FRCC) significantly contributed on the flexural load carrying capacity, ductility, and rigidity of concrete beams. The contributions of these composites varied according to fabric type and amount of fabric layer. The increase in the flexural capacity was also associated with the apparent change in the mode of failure from a flexural mode to shear mode.

1 INTRODUCTION The reinforced concrete structures are subjected to cyclic loads during dynamic loads such as earthquake shocks, traffic loads on the bridges, etc. It is well known that plain concrete is brittle and weak under flexural loads. To eliminate the disadvantages of plain concrete is added fibers into concrete mix. However, the existing concrete elements are strengthened by steel plates, carbon fibers or fabric-cement based composites so that the strength and ductility of these elements subjected to dynamic loads are improved. Mobasher and Lee (1996) expressed that alumina and carbon fibers were stabilized the microcracks and increase the strength of fiber reinforced concrete (FRC) composites and also polypropylene fibers significantly increase the toughness of hybrid composites. Lok and Pei (1998) developed the analysis model concerned with the flexural behavior of steel E-mail addresses: [email protected] (M. Gencoglu), [email protected] (B. Mobasher). 1 Fax: 90 212 285 6587. 2 Fax: 1 480 965 0557. © Millpress, The Netherlands, ISBN 9789059660540

reinforced concrete using the relationship between tensile behavior and flexural response. Another research done by Bindiganavile and Bantia (2001a and b) illustrated that the flexural strength of fiber reinforced concrete was higher under impact loading than under quasi-static loading and polymeric fibers (with suitable length, geometry and deformations) reinforced concrete may absorb fracture energies very close to that of steel fiber reinforced concrete under impact loading. Daniel and Loukili (2002) investigated the behavior of high-strength fiber-reinforced concrete (HSFRC) beams under cyclic loads. They illustrated that HSFRC with tensile reinforcement ratio of 0.55% exhibited behavior similar to that of a HSC beam with a tensile reinforcement ratio of 0.97% and the fibers have no influence on strength deterioration during loading cycles at a given displacement. Peled and Mobasher (2005) found that tensile strength and toughness of the pultruded fabric-cement composites was extremely higher than ones of the conventional GFRC (with short fibers). The objective of this study was to develop new and alternative strengthening methods for concrete 1961

structure elements and to evaluate the performance of the strengthened plain concrete beams using fabriccement based composites under monotonic and cyclic flexural loads. In addition to these aims, the influences of the layer numbers and fabric type on the flexural behavior of the strengthened concrete beams were also studied as experimental parameters. 2 RESEARCH SIGNIFICANCE The primary objective of this research is to improve the flexural strength and ductility of plain concrete beams using FRCC and is to explore alternative and economical strengthening methods and materials for concrete structures. The effects of layer numbers of FRCC on the flexural behavior of the strengthened concrete beams under monotonic and cyclic loads were studied from view point of strength, ductility and initial stiffness. Because of that, the plain concrete beams were strengthened by four and eight layered FRCCs. Furthermore, two different fabric types (AR Glass and PE) were used in this work to determine the effect of fabric type on the flexural behavior of the strengthened concrete beams

superplasticizer (SP) were used. The percentages of cement, silica fume and super-plasticizer were respectively 42, 5 and 0.1 by volume fraction, while in all mixtures the water/binder ratio by weight was 0.37. After the strengthening process, the compression force was applied to the beam surfaces strengthened by fabrics with cement paste so that the bond effects between plain concrete and fabric cement based composite were enhanced. This compression force was held on the strengthened specimens throughout 24 hours and then the plain concrete beams strengthened by fabric cement based composites were held in water (22 ◦ C) for cure of 28 days after hardening of the cement based composites on the plain concrete beams. 4 EXPERIMENTAL SET UP AND PROGRAM Three point bending tests were performed on a closedloop controlled servohydraulic material test system for monotonic and cyclic loading. A test span of 406.4 mm (16 in) was used. The capacity of loading frame 89 kN (20 kips). A three point bending loading fixture was used to eliminate extraneous deformations such as support settle-ments and specimen rotations.

3 TEST SPECIMEN PREPARATION AND MATERIALS The twenty five plain concrete beams with size of 100 × 100 × 460 mm were produced and then were held in water (22 ◦ C) for cure of 28 days. The compressive strength of plain concrete was 27.5 MPa next 28 days after casting concrete. The twenty of these beams were strengthened using AR Glass and PE fabrics with cement paste. The fabrics were immersed in a slurry infiltration chamber, and then pulled through a set of rollers to squeeze the paste in the openings of the fabric, and remove excessive paste. All of the strengthened concrete beams were notched at the mid-span to simulate flexural crack before strengthening process and the penetration of cement paste into notch was prevented during strengthening process. Then the cement paste in thickness of 2∼3 mm was firstly laid and the fabrics with cement paste were mounted to the bottom face of beam with respect to the flexural load layer by layer (see Figure 1). In order to develop a mixture with optimal penetrability within the fabric openings tertiary blends of cement–silica fume in addition to

Figure 1. The strengthened concrete beam during strengthening process. 1962

Figure 2. The flexural set up; a) schematic drawing, b) the view of test set up. Structural Engineering, Mechanics and Computation 3, A. Zingoni (ed.)

A digital controller was programmed to carry out the closed-loop tests. Figure 2a and b represent the schematic drawings and views of the flexural test set up. In the three point bending test, the deflection of the beam was measured using spring-loaded LVDT with a range of 5.08 mm (0.20 in.). To obtain the cyclic flexural behavior of the strengthened concrete beams using fabric-cement based composites, loading/unloading tests were performed. The ascending cycles were controlled with the target deflection recorded by means of LVDT for each cycle while the unloading was under load control at a rate of 44.5 N/s (10 lb/s). For each fabric type and each category of beams strengthened by four and eight layers fabric, three specimens were tested under monotonic flexural loads and two specimens were subjected to cyclic loads. 5 EVALUATIONS OF TEST RESULTS The test results of the strengthened plain concrete beams tested under the monotonic and cyclic flexural loads were evaluated from view point of flexural load-carrying capacities, ultimate deflections and initial stiffness. In addition to these, the envelopes of cyclic loading/unloading responses of test specimens were also obtained and used to compare with each other. The area under the load-deflection response of each specimen subjected to monotonic flexural loads was calculated using Simpson’s numerical integration algorithm to determine the toughness variations with respect to the fabric type and the layer numbers of FRCCs. 5.1 Flexural tests under monotonic loads The flexural load-deflection curves of the strengthened concrete beam are presented in Figure 3. This figure indicate that the non-notched plain concrete beams can carry a little more flexural loads but extremely smaller ultimate deflection than the beams strengthened by PE

fabrics with cement paste. However, it can be seen that the plain concrete beams under flexural loads are quite brittle as well known. When the plain concrete beams are strengthened by four layers of AR Glass fabric with cement paste, Figure 3 shows that the flexural loadcarrying capacity of the strengthened beams can be enhanced about 1.5 folds over the load-carrying capacity of plain concrete beams but the ductility of plain concrete beams can not be improved. Furthermore, it is seen that the both flexural load carrying capacity and the ultimate deflection of the strengthened concrete beam can be enhanced by using the eight layers of AR Glass fabric with cement paste. In addition to these results, Figure 3 indicates that the PE fabrics with cement paste can not increase the flexural load carrying capacity of the strengthened concrete beam as well as AR Glass fabrics with cement paste but the ultimate deflections of the strengthened concrete beams using PE fabrics with cement paste are so much higher than ones of AR Glass fabrics with cement paste. 5.2 Flexural tests under cyclic loads The flexural load-carrying capacities of the strengthened concrete beams under cyclic loads are almost same as load-carrying capacities under monotonic loads (see Figure 4a-d). On the other hand, the ultimate deflections of the strengthened concrete beams under cyclic loads are smaller than ones in the case of monotonic loads. The concrete beams strengthened by PE fabrics with cement paste have more cycles than eight while the strengthened beams using AR Glass fabrics with cement paste have only a few cycles. After the strengthened beams reach the maximum flexural loads, the flexural strength of the beams strengthened by AR Glass fabrics continuously decrease. Whereas, the strengthened concrete beams using PE fabrics continue to carry the flexural loads increasingly after the first drop of flexural load level and the failure of the plain concrete part of strengthened beams. These results can be easily seen from the envelopes of cyclic flexural loads of the strengthened beams (see Figure 4a-d). The initial stiffness of strengthened concrete beams for each cycle were calculated as the slope of cyclic loads-deflection curves and the variations of initial stiffness-max deflection at each loop are presented in Figure 5a and b. These figure indicate that the initial stiffness of concrete beams strengthened by AR Glass fabric with cement paste are nearly ten times higher than that of the strengthened concrete beams using PE fabrics with cement paste up to the first maximum load of specimens. 5.3 Failure shapes of specimens

Figure 3. The variations of flexural load versus deflection of specimens under monotonic loads. © Millpress, The Netherlands, ISBN 9789059660540

The failure shapes of the strengthened beams under the cyclic loads are same as that of specimens subjected to 1963

Figure 4ad. The variation of flexural load versus deflection of specimens under cyclic loads.

1964

Structural Engineering, Mechanics and Computation 3, A. Zingoni (ed.)

Figure 5ab. The variations of initial stiffness versus max. deflection at each loop.

monotonic loads. At the concrete beams strengthened by four layers of AR Glass fabrics, the notch at the midspan of beam rapidly developed after a few cycles and these specimens suddenly failed and the fabrics were broken from the notch place. The strengthened beams using eight layers of AR Glass fabric were consecutively broken from two different sections. After the plain concrete part of the strengthened beams using eight layers AR Glass fabric was broken from the notch, the specimens continued to carry flexural loads for a few cycles and then AR Glass fabrics were also broken from a point near to the supports. Although the plain concrete part of the strengthened beams using PE fabrics (4 and 8 layers) was broken from the notch, the concrete beams strengthened by PE fabrics did not completely fail and continue to carry the flexural loads. But the bond loss between plain concrete and PE cement based composite occurred and developed step by step at each stage of cyclic loading until the bond loss of 280 mm (∼11 in). The damage views of the strengthened beams at the end of tests are presented in Figure 6a–c. © Millpress, The Netherlands, ISBN 9789059660540

Figure 6ac. At the end of tests, the damage views of the strengthened plain concrete beams using, a- AR Glass fabric with 4 layers, b- AR Glass fabric with 8 layers, c-PE fabric with 8 layers.

6 CONCLUSIONS Based on the research presented, the following conclusions can be made: 1. An alternative and economical strengthening method for concrete structural elements was successfully developed by using AR Glass and PE fabric with cement paste in this study. 2. The influences of fabric types on the flexural behavior of the strengthened concrete beams using fabric-cement based composites were clearly demonstrated in this study. Superior increments on flexural behavior of strengthened beams using AR Glass fabrics from view point of the both strength and initial stiffness were observed composites as compared with the test results of concrete beams strengthened by PE fabrics. 1965

3. The considerable improvement in ultimate deflections of the plain concrete beams strengthened by PE fabrics with cement paste was observed as compared with plain concrete beams and the strengthened concrete beams using AR Glass fabrics and plain concrete beams. 4. The precautions which improve the bond effects between concrete and fabric-cement based composites can be studied as a subject another research. 5. After placing of fabric cement based composites on the hardened plain concrete beams, it can be expressed that the intensity of the applied static pressure improves the bond forces between concrete and fabric-cement based composites. But, it was observed that this static pressure is not enough to prevent the bond loss for especially the strengthened concrete beams using PE fabrics with cement paste.

1966

REFERENCES Mobasher, B., and Lee, C. L., (1996). “Mechanical Properties of Hybrid Cement-Based Composites”, ACI Materials Journal, 93(3), 284–292. Lok, T. and Pei, J. (1998). “Flexural Behavior of Steel Fiber Reinforced Concrte”, Journal of Materials in Civil Engineering, V. 10(2), 86–97. Bindiganavile, V. and Banthia, N., (2001a). “Polymer and Steel Fiber-Reinforced Cementitious Composites under Impact Loading, Part 1: Bond-Slip Response”, ACI Materials Journal, 98(1), 10–16. Bindiganavile, V. & Banthia, N.,(2001b). “Polymer and Steel Fiber-Reinforced Cementitious Composites under Impact Loading, Part 2: Flexural Toughness”, ACI Materials Journal, 98(1), 17–24. Daniel, L. and Loukili A., (2002). “Behavior of High-Strength Fiber-Reinforced Concrete Beams under Cyclic Loading”, ACI Materials Journal, 99(3), 248–256. Peled, A. and Mobasher, B. (2005). “Pultruded Fabric-Cement Composites”, ACI Materials Journal, V. 102, No. 1, January–February 2005, 15–23.

Structural Engineering, Mechanics and Computation 3, A. Zingoni (ed.)