Experimental investigation on masonry arches strengthened with PBO-FRCM composite

Composites Part B 100 (2016) 228e239

Contents lists available at ScienceDirect

Composites Part B journal homepage: www.elsevier.com/locate/compositesb

Experimental investigation on masonry arches strengthened with PBO-FRCM composite Valerio Alecci a, Giulia Misseri a, Luisa Rovero a, *, Gianfranco Stipo a, Mario De Stefano a, Luciano Feo b, Raimondo Luciano c  di Firenze, Piazza Brunelleschi 6, 50121 Firenze, Italy Dipartimento di Architettura, Universita  di Salerno, Via Ponte Don Melillo, 84084 Fisciano, SA, Italy Dipartimento di Ingegneria Civile, Universita c  di Cassino e del Lazio Meridionale, Via G. Di Biasio 43, 03043 Cassino, FR, Italy Dipartimento di Ingegneria Civile e Meccanica, Universita a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 April 2016 Received in revised form 16 May 2016 Accepted 30 May 2016 Available online 17 June 2016

In the last two decades, Fiber Reinforced Polymer (FRP) composites were effectively used for increasing the load-carrying capacity of masonry arches and for improving their mechanical response against the most critical arch failure mechanisms. More recently, Fabric Reinforced Cementitious Matrix (FRCM) composites are appearing significantly promising for the strengthening of masonry arches, providing an alternative to the FRP composites, but only few experimental contributions are available to date. In the present paper, the structural performance of masonry arches strengthened both at intrados and at extrados by using a polybenzoxazole (PBO) fabric reinforced cementitious mortar composite was experimentally investigated. Results, in terms of load-carrying capacity, post peak behavior and failure mechanism, were compared with those determined by using a Carbon Fiber Reinforced Polymer (CFRP) composite. The present experimental study contributes to improve the knowledge of the innovative PBOFRCM composites and it provides mechanical data useful for defining masonry strengthening design guidelines. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Fabrics/textile Fiber/matrix bond Mechanical testing Collapse mechanism

1. Introduction Historical masonry buildings, constituting a large part of the world architectural heritage, often include arches and vaults of high architectural value. The maintenance of the safety and usability of these structures is a challenging task for architects and engineers because repairing and/or strengthening interventions need to meet both the seismic safety and the conservation criteria. Nowadays, traditional strengthening techniques of arches and vaults, such as the application of steel profiles at the arches intrados, the insertion of steel rebars, the injection of cementitious mortar and the application of reinforced concrete hoods, are considered unsuitable due to their aesthetic impact, to the irreversibility of the interventions, to the addition of extra weight and stiffness to the structure, etc. The recent wide development of composite materials for constructions allowed to overcome these disadvantages. FRP (Fiber

* Corresponding author. E-mail address: luisa.rovero@unifi.it (L. Rovero). http://dx.doi.org/10.1016/j.compositesb.2016.05.063 1359-8368/© 2016 Elsevier Ltd. All rights reserved.

Reinforced Polymer) strips, applied at intrados or extrados of vaulted structures, demonstrated able to improve their structural performance, both in terms of load carrying capacity and ductility [1e6]. Application of FRP to external or/and to internal surface of arches prevents the typical four hinges mechanism, forcing the arch to collapse by other failure modes: masonry compressive failure, masonry shear failure or FRP debonding from the substrate [7e13]. The key issue of the bond between FRP layer and arch masonry substrate was investigated by experimental tests, numerical models and analytical formulations. By means of experimental tests, it was found that the failure always occurs at the composite/ substrate interface by a debonding phenomenon affecting the superficial layer of the substrate [14,15]. The bond behavior was also studied by a fracture mechanics approach [16e20] and by various numerical models [20e27]. In some cases, different substrates like steel and concrete [28e30] were considered and the bond behavior was investigated successfully applying a mixed mode delamination approach [31,32]. More recently, the scientific community raised the issue of the scarce physical-chemical compatibility of FRP composites with the

V. Alecci et al. / Composites Part B 100 (2016) 228e239

masonry substrate, particularly in the case of historical and artistic constructions for which the compatibility with the original material is often a specific requirement. For this reason, innovative composites made of a fabric embedded in a cement-based mortar (FRCM, Fabric Reinforced Cementitious Matrix) have been proposed as an alternative to FRP composites, especially for strengthening historical and monumental masonry constructions [33e43]. The use of FRCM composites for strengthening masonry members is still to an early stage and few experimental data are available in the literature. Bond performances of FRCM composites when applied to masonry substrate are recently being investigated [44e47]. In particular, in Ref. [44] the bond behavior of a steel reinforced groutestrengthened masonry units was investigated by single lap shear tests; the effect of the surface preparation of bricks as well as the curing time of the grout on bond behavior was also analyzed. In Ref. [45] three FRCM materials made of steel, carbon and basalt textiles embedded in inorganic matrices were investigated by means of single and double lap bond tests. In Ref. [46] double shear tests were carried out on three FRCM composites made of carbon, polybenzoxazole (PBO) and glass textiles embedded in a cementitious matrix. In Ref. [47] bond performance of basalt, carbon, glass and steel textiles embedded in cementitious matrices was investigated in case of application on clay and tuff masonries. Coating treatments of carbon textiles to be embedded in cementitious mortar, in order to increase the bond and tensile strengths, were studied in Ref. [48]. Reduction of bond performances of FRCM composites was investigated [49] in the specific case of application in alkaline and saline environments. The bond performance of FRCM composites was also investigated in case of application on concrete substrate [50,51]. Finally, concerning the structural behavior of masonry arches strengthened through FRCM composites, a carbon-FRCM composite is tested in Refs. [52], a basalt-FRCM composite is studied in Ref. [53] and the use of a composite made of ultra high strength steel filaments twisted to form cords and embedded in a mortar matrix was proposed in Ref. [54]. These results represent an exiguous data sample and, consequently, a much wider quantity of experimental tests are needed in order to define some general conclusions regarding the typical structural behavior, the failure modes, the load-carrying capacity and ductility performances of masonry arches strengthened by FRCM composites. In this perspective, the present paper experimentally investigates the structural behavior of masonry arches strengthened by applying, at the whole intrados or extrados surface, a polybenzoxazole (PBO) fabric reinforced cementitious mortar (FRCM) composite sheet. Results were compared with those determined by testing the same type of masonry arches strengthened by Carbon Fiber Reinforced Polymer (CFRP) sheet at the whole intrados or extrados surface. Ten masonry arch models (1:2 scale) subjected to a vertical force were tested. The experimental study also involved the mechanical characterization of masonry and composites and their components. Furthermore, the bond capacity of the two type of composites was investigated by beam tests. Experimental data provided in the present paper, in terms of load-carrying capacity, post peak behavior and failure mechanism of the tested specimens, allow to improve the knowledge of PBOFRCM composites for strengthening masonry arches and they can be useful to future development of masonry strengthening design guidelines to be used in engineering practice. The rest of the paper is organized into 5 sections. The second section presents the composite materials; the third and the fourth sections cover the experimental investigation and the obtained

229

results, respectively. The final section deals with the concluding remarks. 2. Composite materials Two strengthening composite materials were proposed in the present experimental study. The first composite, produced by Ruredil S.p.A Company, is made of a polybenzoxazole (PBO) bidirectional balanced textile consisting of 14 mm spaced rovings (Fig. 1a), with a nominal thickness in both fiber directions equal to 0.014 mm. The free space between rovings is roughly 11 mm. Warp

Fig. 1. Composite materials (dimensions in mm) a) PBO balanced textile; b) unidirectional carbon fibers.

230

V. Alecci et al. / Composites Part B 100 (2016) 228e239

and weft rovings are overlapped but not connected at the intersections. The matrix is constituted by a pozzolanic cement-based matrix named Ruregold mx. The second composite, also produced by Ruredil S.p.A Company, was a wet layup consisting of single direction (Fig. 1b), 0.17-mmthickness carbon fibers (RUREDIL X WRAP 310) and a twocomponent epoxy-base matrix used for impregnating fibers and gluing them to the substrate. 3. Experimental investigation The experimental investigation was carried out on unstrengthened and both extrados and intrados strengthened masonry arch models (1:2 scale). A first stage of the experimental campaign was devoted to the mechanical characterization of the masonry and its constituent materials, cement-lime mortar and bricks. Beam tests were performed to investigate the bond performances of the two composites and tension and compression tests were carried out on PBO textile and cement-based matrix respectively.

Fig. 2. Compression test set-up (dimensions in mm).

3.1. Mechanical characterization of masonry Masonry used for the construction of the arch models was mechanically characterized by compression tests and shear tests. Furthermore, the mechanical properties of the masonry constituent materials, cement-lime mortar and bricks, were also determined before their assemblage. In particular, mechanical characterization of cement-lime mortar was performed according to [55]. As required by this Standard, three prismatic specimens of 160
40
40 mm3 size were produced. After 28 days of curing, specimens were subjected to three points bending tests and subsequently the two stumps, produced by the rupture of each prism, were subjected to uniaxial compression tests. Results (in terms of average value) of the three points bending and compression tests are reported in Table 1. Mechanical characterization of bricks used for the arch models assemblage was performed according to UNI EN 772-1 [56]. In particular, compression tests were carried out on 8 cubic specimens with side length of 40 mm obtained by cutting common bricks produced by S. Marco Laterizi Company. Tests were performed by maintaining the bricks flatwise direction. Three points bending tests were conducted on 8 bricks of 250
120
55 mm3 size. In Table 1, the average values of compressive and flexural strength of tested bricks are reported. In a second step, these constituent materials were assembled for creating 6 masonry prisms, 1:2 scaled, subsequently subjected to uniaxial compression tests. The mortar and bricks were assembled in line with the arrangement followed for the construction of the arch models, i.e. placing the bricks flatwise, evoiding the alignment of the vertical joints. In particular, the masonry prism texture was constituted by 6 mortar layers, 5 mm thick, and 14 bricks, 21
46
95 mm3 in size, obtained by cutting common bricks

produced by S. Marco Laterizi Company. Compression tests were performed by using a universal press machine equipped with 10 kN load cell (TCLP-10B, Tokyo Sokki Kenkyujo Co. Ltd). The load was applied using a displacement control device, at the rate of 0.05 mm/ s, in order to describe the whole load path. Four displacement transducers were placed on a metal plate positioned on the top of the prism; one U transducer was positioned on the principal side and one on the back side of the specimen (Fig. 2). In Table 1, the average values of compressive strength, compressive elastic modulus and ultimate strain are reported. In order to obtain reliable data on masonry shear strength, 18 masonry triplets constituted by three bricks and two mortar joints 10 mm thick, were tested. The triplet tests were performed following two different procedures, i.e. with or without lateral compression [57]. A displacement control device made up of a rigid frame with a screw jack, which can be controlled through a flywheel, was used. The test apparatus is schematized in Fig. 3. A metallic plate and cylinder of 15 mm, respectively in thickness and in diameter, were positioned under the lateral bricks of the triplet. The load was applied on the top of the central brick by a spherical hinge placed on a double metallic plate with interposed two metallic cylinders of 120 mm in length (equal to the brick width). The results of the triplet tests, performed both with and without lateral compression (named respectively T-WPR and T-PR) are presented in Table 2, where the shear strength was calculated as Fmax/2A (where Fmax is the maximum load and A is the specimen area parallel to the mortar joint). A linear regression of the experimental results on the (s,t) plane permitted to evaluate the cohesion (0.26 MPa) and the friction coefficient (0.6) of the masonry.

Table 1 Results of uniaxial compression tests and three points bending tests on bricks, cement-lime mortar, and masonry (the standard deviation and the coefficient of variation are reported in parentheses). Specimen

Failure compressive strain

Compressive strength (MPa)

Compressive young modulus (MPa)

Flexural tensile strength (MPa)

Cement-lime Mortar

/

Brick

/

Masonry

0.0076 (0.002; 17.66)

3.22 (0.31; 9.72) 24.08 (2.73; 11.37) 8.53 (1.29; 13.95)

727.7 (69.82; 9.59) 2701,81 (585.05; 21.65) 1753.7 (282.52; 16.11)

1.49 (0.025; 1.68) 5.60 (0.58; 10.44) /

V. Alecci et al. / Composites Part B 100 (2016) 228e239

231

Fig. 3. Triplet test set-up (dimensions in mm).

Table 2 Results of triplet tests. Specimen

Maximum load (N)

Pre-compression (MPa)

Sher strength (MPa)

T-NPR T-WPR-1 T-WPR-2 T-WPR-3 T-WPR-4 T-WPR-5

2978 24016 28387 36908 52987 58553

/ 0.2 0.4 0.6 1.0 1.2

0.26 0.40 0.47 0.61 0.88 0.97

3.2. Mechanical characterization of the FRCM composite The FRCM composite used in the present experimental study, is made of PBO bidirectional balanced textile and an inorganic matrix constituted by a pozzolanic cement-based mortar. Mechanical properties of the PBO-FRCM composite were firstly investigated thorough the mechanical characterization of the constituent materials, cementitious mortar matrix and PBO textile. More precisely, 3 prismatic mortar specimens of 160
40
40 mm3 size were produced e subjected to three points bending tests according to [55]. Subsequently, the two stumps produced by the rupture of each prism, were subjected to uniaxial compression tests. Average values of experimental test results are reported in Table 3. PBO textile was characterized by direct tensile test according to [58]. In particular, 3 textile specimens were obtained from the commercial product and tested under displacement controlled in order to observe the whole loading path (Fig. 4aeb). Load was

Fig. 4. Direct tensile test on PBO fiber textile (a) and stress-strain curve (b).

applied through a 600 kN hydraulic actuator; global measures were acquired by linear variable differential transformer integrated in the testing machine with 1 mm of resolution while local measures were obtained by U extensometers, having 50 mm gage length, positioned in the middle of the textile specimen. The average values

Table 3 Results of uniaxial compression tests and three points bending tests on the cementitious matrix (the standard deviation and the coefficient of variation are reported in parentheses). Component material

Compressive strength (MPa)

Compressive young modulus (MPa)

Flexural tensile strength (MPa)

Ruregold MX

20.22 (2.12; 10.51)

2874.72 (458.38; 15.94)

6.15 (0.35; 5.7)

232

V. Alecci et al. / Composites Part B 100 (2016) 228e239

Table 4 Mechanical properties of textiles. Component material

Tensile strength (MPa)

Tensile strength per unit width (N/mm)

Tensile young modulus (MPa)

Ultimate strain

PBO Carbon

3328a 4800b

46.59a 816.0b

223382a 240000b

0.0149a 0.02b

a b

Data experimentally determined in this study. Data provided by the manufacturer - Ruredil S.p.A., 2013.

of mechanical properties of PBO textile are summarized in Table 4 and compared with those referred to the carbon fibers as supplied by the manufacturer. 3.3. Bond capacity of the PBO-FRCM and CFRP composites The characterization of the composite materials used to strengthen the masonry arches was also performed in terms of bond capacity. The investigation of the bond capacity and of the mechanisms at the fiberematrix interface play a crucial role for understanding the mechanical behavior of FRCM systems. In particular, beam tests were carried out on 24 specimens on which the FRCM composite and the FRP composite were applied in order to investigate the bond capacity (Fdb), i.e. the maximum force that can be transferred to the masonry substrate. It is determined as the tensile force in the composite layer corresponding to the maximum load supported by the specimen. Each of the twenty-four tested specimens was made of two bricks connected by a metallic hinge on the compression side and by the FRCM or FRP composite sheet on the tension side. The composite was applied along the whole length of the specimen comprised between the two steel cylinder supports. A FRCM composite sheet 95 mm width (the same width of the sheet applied to strengthen the masonry arches) and a FRP composite sheet 20 mm width were applied to 12 specimens respectively. Reduced width (20 mm) of the FRP composite sheet was used in order to produce the specimen failure due to composite debonding (avoiding the premature specimen failure due to the masonry crushing) and capturing the corresponding Fdb value. As illustrated in Fig. 5, the load was applied by a spherical hinge placed on a metallic plate and transferred to the specimen by two metallic cylinders. Each specimen was equipped with three displacement transducers (type CE cantilever): one of them was positioned on the metal hinge to measure the vertical displacement of the mid-span cross-section and two at the specimen ends to measure the displacement of the supports (Fig. 6). The test was performed under

Fig. 6. Specimen 2BT subjected to beam test.

displacement controlled at the rate of 0.05 mm/s. In Table 5, test results in terms of maximum applied load Fmax are reported). The failure of specimens strengthened by FRCM composite was characterized by a textile-matrix slip occurred at the mid-span cross-section while cementitious matrix remained perfectly attached to the substrate. The specimens strengthened by FRP composite failed due to the FRP debonding at the compositeesubstrate interface affecting the superficial layer of the substrate. The bond capacity Fdb of the composite material was obtained through the equilibrium between internal moment and the mid-span bending moment relative to the maximum applied force Fmax. The tests provided an average bond capacity Fdb,m for the FRCM composite (sheet 95 mm width) and the FRP composite (sheet 20 mm width) equal to 5096 N and 8198 N respectively. 3.4. Arch testing Experimental campaign addressed ten 1:2 scaled masonry arches (Fig. 7). Two specimens were tested in the unstrengthened configuration (specimens 1-US and 2-US); four specimens were

Fig. 5. Beam test set-up (dimensions in mm).

V. Alecci et al. / Composites Part B 100 (2016) 228e239

233

Table 5 Average values of beam-test results. (the standard deviation and the coefficient of variation are reported in parentheses). Specimen code

Composite material

Composite sheet width (mm)

Maximum force Fmax (N)

1-12F

FRCM

95

1-12BT

FRP

20

2378 (112.37; 4.72) 3518 (112.19; 3.19)

loading history, up to the point of a conventional test-end corresponding to a residual strength equal to 80% of the peak load. The load was applied at a quarter of the span (375 mm from the left abutment) using a squared steel plate 20 mm thick and it was measured through a load cell with a capacity of 10 kN (TCLP-10B tension/compression load cell). Vertical displacements were measured by means of two displacement transducers (type CE cantilever) positioned on the metallic plate (Fig. 7). The transducers and the load cell were connected to a computer through an electronic control unit that directly displayed the load-displacement curve. During the tests, observation of the specimens and analysis of the contemporary displayed load-displacement curve allowed to record cracks, stiffness variation, hinges formation, failure modes. 4. Test results

Fig. 7. Geometry (mm) of the arch and testing set-up.

strengthened with PBO-FRCM composite, two at the extrados and two at the intrados (specimens 1e2/PeS and 1e2/PiS respectively); four specimens were strengthened with CFRP composite, two at the extrados and two at the intrados (specimens 1e2/CeS and 1e2/CiS respectively). Bricks were 95
46
21 mm3 sized with average thickness of the mortar layers of 5 mm, according to the scale of the arch models. The arches had a 1500 mm span, with 866 mm intrados radius, 961 mm extrados radius, 432.5 mm rise, and 95
95 mm2 cross section (width by thickness). With the aim of comparison, the PBO-FRCM and the CFRP composite sheets had the same width (95 mm) and were applied on the whole arch surface, in case of the extrados as well as of the intrados application. Consequently, in virtue of the different nominal thickness of the PBO and carbon textile, the fiber transversal section in the applied composite sheet is constituted by different areas: 16.15 mm2 for CFRP and 1.33 mm2 for PBO-FRCM. The cement matrix for PBO-FRCM composite was prepared according to the manufacturer’s data sheet. The bricks were accurately cleaned before mortar application; then, a first layer of cementitious matrix of about 3 mm thick was applied. Then, the PBO textile was positioned on this matrix layer and, subsequently, a second cementitious matrix layer was applied. The composite layer reached a final thickness of about 6 mm. For the CFRP composite application, a two-component epoxybase matrix was used for impregnating fibers and gluing them to the substrate, reaching a final thickness between 1 and 1.5 mm. All of the strengthened arches were tested after 28 days of curing in a controlled temperature environment (20  C and 50% relative humidity). The test apparatus and the arch geometry are illustrated in Fig. 7. Each unstrengthened and strengthened arch was subjected to a vertical force, applied by using a displacement control device made of a screw jack controlled through a flywheel. The use of the displacement control device permitted to observe the whole

Test results are illustrated thorough load-displacement diagrams (Figs. 8e10) and reported in Table 6 for unstrengthened (US) arches and PBO-FRCM strengthened arches at intrados (PiS) and extrados (PeS). Furthermore, with aim of comparison, the results concerning the CFRP strengthened arches at intrados (CiS) and extrados (CeS) are also reported. The following structural parameters are identified: maximum load, stiffness, kinematic ductility and available kinematic ductility. Stiffness was determined in the linear range of the loadedisplacement curve, in order to avoid the influence of the non-linear start of the load path produced by the contact between load plate and specimen surface. Kinematic ductility (i.e. the capacity of the specimen to show large displacements after the linear elastic phase up to the maximum load) was calculated as the ratio of the displacement measured at the maximum load, to the displacement corresponding to the linear elastic limit of the load path; available kinematic ductility (i.e. the capacity of the specimen to show large displacements after the maximum load up to the ultimate load) was taken as the ratio of the ultimate displacement (displacement corresponding to the ultimate load, conventionally assumed to be equal to 80% of the maximum load) to the displacement at the maximum load [59,60]. The fifth column of the Table 6 shows the observed failure mode for each tested specimen. 4.1. Unstrengthened and PBO-FRCM strengthened arches Fig. 8 shows the load-displacement curve of the loaded cross section of the unstrengthened and both intrados and extrados PBOFRCM strengthened arches. It can be noted the considerable contribution of PBO-FRCM material in increasing the maximum load and ductility (kinematic ductility as well as available kinematic ductility). The initial stiffness was similar in the unstrengthened and intrados strengthened arches while the extrados strengthened arches showed an higher stiffness. An important difference was observed comparing the effect of the strengthening at the intrados and extrados surfaces. In fact, the arches strengthened at the extrados exhibited a higher kinematic ductility than those strengthened at the intrados. On another hand, the maximum load didn’t show considerable variations.

234

V. Alecci et al. / Composites Part B 100 (2016) 228e239

Table 6 Results of tests on unstrengthened and strengthened arches (Us ¼ unstrengthened; CiS ¼ CFRP intrados strengthened; CeS ¼ CFRP extrados strengthened; PiS ¼ PBO-FRCM intrados strengthened; PeS ¼ PBO-FRCM extrados strengthened). Specimen

Maximum load (N)

Tangent stiffness (N/mm)

Kinematic ductility

Available Kinematic ductility

Failure mode

1-US 2-US 1-PiS 2-PiS 1-PeS 2-PeS 1-CiS 2-CiS 1-CeS 2-CeS

910 1066 5280 5672 4968 4813 15198 14940 11345 11955

7179 6197 5891 5510 16221 9946 5329 5364 10106 8271

1.85 1.11 2.77 3.64 20.32 19.46 1.37 1.20 10.93 7.88

1.10 1.18 1.42 2.12 1.11 1.51 1.15 1.32 / /

Hinges mechanism Hinges mechanism Debondingþmasonry crushing Debondingþmasonry crushing Shear sliding Shear sliding Masonry crushing Masonry crushing Shear sliding Shear sliding

Fig. 8. Load-displacement curves of the unstrengthened and PBO-FRCM strengthened arches (Us ¼ unstrengthened; iS ¼ intrados strengthened; eS ¼ extrados strengthened).

Results reported in Table 6 confirm the effectiveness of PBOFRCM composite in increasing the maximum load (by approximately 400% for extrados strengthening and 480% for the intrados one) and the ductility. In particular, the kinematic ductility resulted considerably increased in case of extrados application (approximately 13 times the value of the unstrengthened arches) with respect to the case of intrados application (2 times the value of the unstrengthened arches).

4.2. Failure modes of unstrengthened and PBO-FRCM strengthened arches As was expected, the unstrengthened arches exhibited a collapse mechanism with four alternate (intrados/extrados) hinges (Fig. 9). The first hinge occurred at the arch extrados at the loaded cross section and the second hinge at the intrados, in a symmetric position with respect to the previous one. The third and fourth hinge occurred approximately on the left and right abutment respectively. In Fig. 9, the position and order of formation of the hinges on the unstrengthened arches are represented. The application of PBO-FRCM strengthening composite at the extrados or intrados of the arches, modified the collapse mechanism. Concerning the extrados application, it was observed that the onset of the hinge formation on the loaded cross-section, revealed by thin cracks appeared on the mortar joint, coincided with the

Fig. 9. Sequence of hinges formation on the unstrengthened arches.

linear elastic phase limit corresponding to a load of 3780 N for 1-US arch and 3825 for 2-US arch. A further increase of the load produced the opening of the first hinge (Fig. 10a). At the same time, a slight debonding at the composite-to-masonry interface was observed, also revealed by a sudden load reduction at constant

V. Alecci et al. / Composites Part B 100 (2016) 228e239

235

displacement rate (Fig. 13). The presence of PBO-FRCM composite at the extrados partially prevented the opening of the relative hinges so, in this case, only superficial cracks of the cementitious matrix were observed approximately where second and third hinges occurred in the unstrengthened arches (Fig. 10b). In this phase the load slightly increased reaching a peak load of 4968 N and 4813 N for 1-PeS and 2-PeS arches respectively. After, additional cracks occurred on the matrix allowing a large displacements up to the arch collapse, due e after many warnings e to shear sliding at the right abutment (Fig. 10c). The intrados PBO-FRCM strengthened arches exhibited a different failure mode involving the debonding at the PBO textilemortar interface and the crisis of the masonry near the loaded cross section. More precisely, both in case of 1-PiS and 2-PiS specimens, the linear elastic limit coincided with the formation of a first crack (corresponding load was approximately 2900 N in arch 1-PiS and 3650 in arch 2-PiS) where second hinge occurred in the unstrengthened arches (Fig. 11a). In the case of intrados strengthening, in fact, the presence of the PBO-FRCM material prevented the opening of hinge on the loaded cross section. With the increase of the load, the crack opened and a first hinge formed at a corresponding load of 4980 N in arch 1-PiS and 4100 in 2-PiS arch. After this phase, 1-PiS arch quickly reached its maximum load capacity (5280 N). Some cracks were observed on the composite matrix and on the mortar joints near the loaded cross section. This did not occur in the 2-PiS arch that showed a second hinge at the left abutment (5190 N). Subsequently, superficial cracks of the cementitious matrix and a crack involving mortar joints and bricks at the loaded cross section were observed (Fig. 11b). The corresponding maximum load was 5872 N. The post-peak phase, both in case of 1-PiS and 2-PiS arches, was characterized by an increasing of the masonry crushing and of the debonding phenomenon at the textile-mortar interface. Many warnings were provided before failure. At the end of the test, the PBO-FRCM composite remained fixed to the substrate but an evident debonding at the PBO textilemortar interface resulted (Fig. 11c). 4.3. Comparison with the CFRP strengthened arches The comparison between the load-displacement curves of PBOFRCM and CFRP strengthened arches are plotted in Figs. 12 and 13 for intrados and extrados strengthening respectively. Obviously, the contribution of the CFRP composite in terms of maximum load is much higher than that provided by the PBO-FRCM composite, similarly in the case of intrados (Fig. 12) and extrados strengthening (Fig. 13), also due to the different fiber amount of the two composite sheets. Conversely, PBO-FRCM strengthened arches showed a more ductile behavior. In fact, CFRP strengthened arches always had a brittle failure and a consequent low available kinematic ductility both in case of intrados (Fig. 12) and extrados (Fig. 13) strengthening. Only in the case of the extrados strengthening, the brittle collapse was preceded by some warning corresponding to the ascending branch of the load path (between the linear elastic limit and the maximum load) characterized by a reduced slope and large displacements (Fig. 13). In order to make a more reliable comparison between the two composites about their effectiveness for improving the arch models load-carrying capacity, the exploitation of the bond capacity of the strengthening sheet was evaluated as the ratio between the average maximum load, determined by testing the arches with

Fig. 10. Failure details of PBO-FRCM extrados strengthened arches: a) hinge formation on the loaded cross section; b) superficial cracking of the cementitious matrix; c) shear sliding at the right abutment.

236

V. Alecci et al. / Composites Part B 100 (2016) 228e239

strengthening both at intrados (Fmax,in) and extrados (Fmax,ex), and the average bond capacity (Fdb), determined through beam test (Table 7). The value of Fdb for a FRP sheet of 95 mm width was calculated linearly increasing the value obtained through beam test by applying a strip 20 mm width [61e15]. The Fdb value for a FRP sheet of 95 mm is approximately 8 times the value of Fdb for a PBOFRCM sheet of 95 mm. The values reported in Table 7 highlight that, in terms of loadcarrying capacity, PBO-FRCM strengthened arches exploit better the bond capacity of the strengthening sheet with respect to CFRP strengthened arches. In order to more deeply study the comparison between the two composites, the limiting area fraction of the strengthening sheet corresponding to the composite sheet failure was also investigated. Force equilibrium and strain compatibility give the following equation [62]:

6lim ¼

εM;u 0:8 εM;u ND $  $ εF;u 1 þ εεF;u εF;u NM;u

(1)

M;u

where ulim is the normalized strengthening sheet area fraction, εF,u is the ultimate strain of the strengthening sheet calculated through beam test, εM,u is the ultimate strain of the masonry calculated by compression test on masonry prisms, Nd is the design normal force, and NM,u is the resistant normal force of the masonry cross-section. In Fig. 14, diagram representing the normalized strengthening sheet area fraction ulim as a function of the ratio Nd/NM,u is reported for the PBO-FRCM and CFRP composites. The value of u referred to the CFRP composite sheet used to strengthen the arches was found to be equal to 0.38 while the maximum ulim is equal to 0.26, as reported in Fig. 14. In agreement with these data, the failure mechanisms of the CFRP strengthened arches highlighted that the bond capacity of the CFRP sheet was not fully exploited. In fact, brittle failure mechanisms, such as the masonry crushing (in case of intrados strengthening) and the sliding at the arch abutment (in case of extrados strengthening), preceded the debonding of the CFRP composite sheet. More precisely, 1-CiS and 2CiS arches failed due to the crushing of the masonry at the loaded cross section when a maximum load equal to 15198 N and 14940 N respectively was reached. In the case of 2-CiS arch, the masonry crushing involved a large debonding of the CFRP sheet from the substrate. This debonding was a consequence of the masonry crushing: cracks, arisen close to the point of application of the load, rapidly increased and developed up to the arch intrados triggering the debonding process. The complete debonding occurred in a very short time and without providing additional warnings.

5. Conclusions

Fig. 11. Failure details of PBO-FRCM intrados strengthened arches: a) first hinge; b) masonry crushing; c)debonding at the textile-mortar interface.

The present experimental investigation covers the structural performance of masonry arches strengthened both at intrados and at extrados by using a polybenzoxazole (PBO) fabric reinforced cementitious mortar (FRCM). Results were also compared with those determined by using a Carbon Fiber Reinforced Polymer (CFRP). More precisely, the experimental work involved the test of ten masonry arch models (1:2 scale) subjected to a vertical force. The experimental study also concerned the mechanical characterization of masonry and composites and the investigation e through beam tests e on the bond capacity of the two types of composite. Concerning the arch testing, the principal mechanical parameters (load-carrying capacity, stiffness and ductility) were determined and failure mechanisms were accurately described and interpreted. The following conclusions can be deduced from the experimental results:

V. Alecci et al. / Composites Part B 100 (2016) 228e239

237

Fig. 12. Load-displacement curves of the PBO-FRCM and CFRP intrados strengthened arches. (CiS ¼ CFRP intrados strengthened; PiS ¼ PBO-FRCM intrados strengthened).

Fig. 13. Load-displacement curves of the PBO-FRCM and CFRP extrados strengthened arches. (CeS ¼ CFRP extrados strengthened; PeS ¼ PBO-FRCM extrados strengthened).

Table 7 Exploitation of the composite bond capacity (average values). Fdb ¼ bond capacity; Fmax,in ¼ maximum load of intrados strengthened arches; Fmax,ex ¼ maximum load of extrados strengthened arches. Composite material

Bond capacity Fdb (N)

Maximum load Fmax,in (N)

Fmax,in/Fdb

Maximum load Fmax,ex (N)

Fmax,ex/Fdb

FRCM FRP

5096 38940

5476 15069

1.07 0.39

4890 11650

0.96 0.30

- PBO-FRCM composite demonstrated effective in increasing the maximum load (more than 400% of the values of the unstrengthened arch) both in the case of extrados and intrados application. - In terms of ductility, a difference between intrados and extrados application of PBO-FRCM composite was observed. In particular, kinematic ductility resulted approximately 13 times the value of the unstrengthened arches in the case of extrados application

while 2 times the value of the unstrengthened in the case of intrados application. - The comparison between results obtained from PBO-FRCM and CFRP strengthened arches (with the same composite sheet width) highlighted that the contribution of CFRP composite in terms of maximum load is much higher than that provided by PBO-FRCM composite, similarly in the case of intrados and extrados strengthening.

238

V. Alecci et al. / Composites Part B 100 (2016) 228e239

Fig. 14. Normalized area fraction ulim versus Nd/NM,u ratio for the PBO-FRCM and CFRP composite sheet.

Conversely, in terms of ductility, CFRP strengthened arches always showed a brittle failure and a consequent low available kinematic ductility both in the case of intrados and extrados strengthening. - The comparison between the maximum load determined by testing strengthened arches (both at intrados and extrados) and the bond capacity of the strengthening composite sheet, determined through beam test, highlighted that PBO-FRCM strengthened arch models better exploited the bond capacity of the strengthening sheet. The laboratory data demonstrated that PBO-FRCM composite is an effective solution for the strengthening of masonry members, being able to improve their load-carrying capacity as well as their ductility capacity (specific requirement of the current seismic codes). Furthermore, the high physical-chemical compatibility with the masonry substrate makes this innovative composite particularly useful in the case of strengthening interventions on historical and artistic buildings for which the compatibility with the original substrate is often a specific requirement. References [1] Ianniruberto U, Rinaldi Z. Systems of arches and columns strengthened with FRP at the extrados. In: Modena, Lourenco, Roca, editors. Structural analysis of historical constructions. London: Taylor and Francis Group; 2005. ISBN 0415363799. [2] Rovero L, Focacci F, Stipo G. Structural behavior of arch models strengthened using FRP strips of different lengths. J Compos Constr 2013;17(2):249e58. [3] De Lorenzis L, Dimitri R, La Tegola A. Reduction of the lateral thrust of masonry arches and vaults with FRP composites. Constr Build Mater 2007;21(7): 1415e30. [4] Briccoli Bati S, Rovero L, Tonietti U. Strengthening of masonry arches with composite materials. J Compos Mater 2007;11(1):33e42. [5] Borri A, Castori G, Corradi M. Intrados strengthening of brick masonry arches with composites materials. Compos Part B Eng 2011;42:1164e72. [6] Castori G, Borri A, Corradi M. Behavior of thin masonry arches repaired using composite materials. Compos Part B 2015. http://dx.doi.org/10.1016/ j.compositesb.2015.09.008. [7] Cosenza E, Manfredi G, Occhiuzzi A, Pecce M. Toward the investigation of the interface behaviour between tuff masonry and FRP fabrics. In: Proceeding of the third Italian Conference of Mechanics of masonry structures strengthened with composite materials (MuRiCo/3), Venice, Italy; April, 2009.

[8] Valluzzi MR, Valdemarca M, Modena C. Behavior of brick masonry vaults strengthened by FRP laminates. J Compos Constr 2001;5(3):163e9. [9] Foraboschi P. Strengthening of masonry arches with fiber-reinforced polymer strips. J Compos Constr 2004;8(3):191e202. [10] Briccoli Bati S, Rovero L. Towards a methodology for estimating strength and collapse mechanism in masonry arches strengthened with fibre reinforced polymer applied on external surfaces. Mater Struct 2008;41:1291e306. [11] Oliveira DV, Basilio I, Lourenco PB. Experimental behavior of FRP strengthened masonry arches. J Compos Constr 2010;14(3):312e22. [12] Tao Y, Stratford TJ, Chen JF. Behaviour of a masonry arch bridge repaired using fibre-reinforced polymer composites. Eng Struct 2010;33:1594e606. [13] Kesteloot S, Djelal C, Benslimane I, Khoufache N. Strengthening of masonry arches using carbon plates. Constr Build Mater 2015;81(15):216e23. [14] Valluzzi MR, Oliveira Daniel V, Caratelli Angelo, Zuccarino Gina. Round robin test for composite-to-brick shear bond characterization. Mater Struct 2012;45(12):1761e91. [15] Rotunno T, Rovero L, Tonietti U, Briccoli Bati S. Experimental study of bond behavior of CFRP-to-brick joints. J Compos Constr 2014. http://dx.doi.org/ 10.1061/(ASCE)CC.1943-5614.0000528. 04014063. [16] De Lorenzis L, Zavarise G. Interfacial stress analysis and prediction of debonding for a thin plate bonded to a curved substrate. Int J Nonlinear Mech 2009;44(4):358e70. [17] De Lorenzis L, Zavarise G. Debonding analysis of thin plates from curved substrates. Eng Fract Mech 2010;77(16):3310e28. [18] Milani G, Milani E, Tralli A. Upper Bound limit analysis model for FRP reinforced masonry curved structures. Part I: unreinforced masonry failure surfaces. Comput Struct 2009a;87(23e24):1516e33. [19] Milani G, Milani E, Tralli A. Upper Bound limit analysis model for FRPreinforced masonry curved structures. Part II: structural analyses. Comput Struct 2009b;87(23e24):1534e58. [20] D’Antino T, Carloni C, Sneed LH, Pellegrino C. Matrix-fiber bond behavior in PBO FRCM composites: a fracture mechanics approach. Eng Fract Mech 2014;117(02):94e111. [21] Caporale A, Luciano R, Rosati L. Limit analysis of masonry arches with externally bonded FRP reinforcements. Comput Methods Appl Mech Eng 2006;196: 247e60. [22] Caporale A, Luciano R. Limit analysis of masonry arches with finite compressive strength and externally bonded reinforcement. Compos Part B Eng 2012a;43:3131e45. [23] Caporale A, Feo L, Luciano R. Limit analysis of FRP strengthened masonry arches via nonlinear and linear programming. Compos Part B Eng 2012b;43: 439e46. [24] Caporale A, Feo L, Luciano R, Penna R. Numerical collapse load of multi-span masonry arch structures with FRP reinforcement. Compos Part B Eng 2013;54: 71e84. [25] Caporale A, Feo L, Luciano R. Debonding of FRP in multi-span masonry arch structures via limit analysis. Compos Struct 2014;108:856e65. [26] Cancelliere I, Imbimbo M, Sacco E. Experimental tests and numerical modeling of reinforced masonry arches. Eng Struct 2010;32(3):776e92. [27] Carozzi FG, Milani G, Poggi P. Mechanical properties and numerical modeling of fabric reinforced cementitious matrix (FRCM) systems for strengthening of masonry structures. Compos Struct 2014;107:711e25.

V. Alecci et al. / Composites Part B 100 (2016) 228e239 [28] Bruno D, Greco F, Lo Feudo S, Nevone Blasi P. Multi-layer modeling of edge debonding in strengthened beams using interface stresses and fracture energies. Eng Struct 2016;109:26e42. [29] Bruno D, Carpino R, Greco F. Modelling of mixed mode debonding in externally FRP reinforced beams. Compos Sci Technol 2007;67:1459e74. [30] Greco F, Lonetti P, Nevone Blasi P. An analytical investigation of debonding problems in beams strengthened using composite plates. Eng Fract Mech 2007;74(3):346e72. [31] Bruno D, Carpino R, Greco F, Lonetti P. Energy release rate and mode partition for interlaminar crack in circular laminated beams. Int J Solids Struct 2006;43: 1201e23. [32] Bruno D, Greco F, Lonetti P. A coupled Interface-multilayer approach for mixed mode delamination and contact analysis in laminated composites. Int J Solids Struct 2003;40:7245e68. [33] Briccoli Bati S, Marilli F, Rovero L, Tonietti U. Mechanical behavior of masonry specimens reinforced with CFRCM (carbon fiber reinforced cement matrix) sheet. In: Proceedings of 4th international conference on restoration, recycling and rejuvenation techonology for engineering and architecture application (RRRTEA ’04), Cesena, Italy; June, 2004. p. 86e91. [34] Prota A, Marcari G, Fabbrocino G, Manfredi G, Aldea C. Experimental in-plane behavior of tuff masonry strengthened with cementitious matrixegrid composites. J Compos Constr 2006;10(3):223e33. [35] Carbone I, De Felice G. Debonding of C-FRCM composite on masonry support. In: Proceedings of 9th international conference on protection of historical buildings (PROHITECH 09), Rome; June, 2009. p. 577e83. [36] Ascione L, De Felice G, De Santis S. A qualification method for externally bonded Fibre Reinforced Cementitious Matrix (FRCM) strengthening systems. Compos Part B Eng 2015;78:497e506. [37] Garmendia L, Marcos I, Garbin E, Valluzzi MR. Strengthening of masonry arches with Textile-Reinforced Mortar: experimental behavior and analytical approaches. Mater Struct 2014;47(12):2067e80.  JT. Strengthening masonry [38] Garmendia L, Larrinaga P, San-Mateos R, San-Jose vaults with organic and inorganic composites: an experimental approach. Mater Des 2015;85:102e14. [39] D’Ambrisi A, Focacci F, Luciano R, Alecci V, De Stefano M. Carbon-FRCM materials for structural upgrade of masonry arch road bridges. Compos Part B Eng 2015. http://dx.doi.org/10.1016/j.compositesb.2015.01.024. [40] Babaeidarabad S, De Caso F, Nanni A. URM walls strengthened with fabricreinforced cementitious matrix composite subjected to diagonal compression. J Compos Constr 2014a;(2):18. http://dx.doi.org/10.1061/(ASCE) CC.1943-5614.0000441. 04013045. [41] Babaeidarabad S, De Caso F, Nanni A. Out-of-plane behavior of URM walls strengthened with fabric-reinforced cementitious matrix composite. J Compos Constr 2014b;(4):18. http://dx.doi.org/10.1061/(ASCE)CC.19435614.0000457. 04013057. [42] D’Ambrisi A, Feo L, Focacci F. Experimental and analytical investigation on bond between Carbon-FRCM materials and masonry. Compos Part B Eng 2013;46:15e20. [43] D’Ambrisi A, Mezzi M, Feo L, Berardi VP. Analysis of masonry structures strengthened with polymeric met reinforced cementitious matrix materials. Compos Struct 2014;113:264e71. [44] Razavizadeh A, Ghiassi B, Oliveira DV. Bond behavior of SRG-strengthened

[45]

[46]

[47]

[48] [49]

[50]

[51] [52] [53]

[54]

[55]

[56] [57] [58]

[59] [60]

[61]

[62]

239

masonry units: testing and numerical modeling. Costr Build Mater 2014;64: 387e97. De Felice G, De Santis S, Garmendia L, Ghiassi B, Larrinaga P, Lourenco PB, et al. Mortar-based systems for externally bonded strengthening of masonry. Mater Struct. 2014;47(12):2021e37. Alecci V, De Stefano M, Luciano R, Rovero L, Stipo G. Experimental investigation on bond behavior of cement-matrixebased composites for strengthening of masonry structures. J Compos Constr 2015. http://dx.doi.org/ 10.1061/(ASCE)CC.1943-5614.0000598. Ascione L, De Felice G, De Santis S. A qualification method for externally bonded fibre reinforced cementitious matrix (FRCM) strengthening systems. Compos Part B Eng 2015;78:497e506. Donnini J, Corinaldesi V, Nanni A. Mechanical properties of FRCM using carbon fabrics with different coating treatments. Compos Part B Eng 2016;88:220e8. Nobili A. Durability assessment of impregnated glass fabric reinforced cementitious matrix (GFRCM) composites in the alkaline and saline environments. Constr Build Mater 2016;105:465e71. D’Ambrisi A, Feo L, Focacci F. Experimental analysis on bond between PBOFRCM strengthening materials and concrete. Compos Part B Eng 2013;44: 524e32. D’Ambrisi A, Feo L, Focacci F. Bond-slip relations for PBO-FRCM externally bonded to concrete. Compos Part B Eng 2012;43:2938e49. Bednarz L, Gorski A, Jasienko J, Rusinski E. Simulations and analyses of arched brick structures. Autom Constr 2011;20(12):741e54.  JT, Garcìa D, Larrinaga P. Rehabilitation of masonry Garmendia L, San-Jose arches with compatible advanced composite material. Constr Build Mater 2011;25(12):4374e85. Borri A, Casadei P, Castori G, Hammond J. Strengthening of brick masonry arches with externally bonded steel reinforced composites. J Compos Constr 2009;13(6):468e75. European Standard UNI EN 1015-11. Methods of test for mortar for masonry determination of flexural and compressive strength of hardened mortar. 2007. European Standard UNI EN 772-1. Methods of test for masonry units determination of compressive strength. 2011. European standard UNI EN EN 1052-3. Methods of test for masonry e determination of initial shear strength. 2007. ASTM D5034. American society for testing and materials. Standard test method for breaking strength and elongation of textile fabrics (grab test). 2013. Rovero L, Focacci F, Stipo G. Structural behavior of arch models strengthened using FRP strips of different lengths. J Compos Constr 2013;17(2):249e58. Alecci V, Briccoli Bati S, Ranocchiai G. Concrete columns confined with CFRP wraps. Mater Struct 2014;47:397e410. http://dx.doi.org/10.1617/s11527013-0068-7. Consiglio Nazionale Delle Ricerche “Istruzioni per la Progettazione, l’Esecuzione ed il Controllo di Interventi di Consolidamento Statico mediante l’utilizzo di Compositi Fibrorinforzati. Materiali, strutture di c.a. e di c.a.p., strutture murarie.” CNR-DT 200 R1/2013, Rome, Italy (in Italian). Triantafillou TC. Strengthening of masonry structures using epoxy-bonded FRP laminates. J Compos Constr 1998;2(2):96e104.