PSI - Issue 44

Annalisa Napoli et al. / Procedia Structural Integrity 44 (2023) 2182–2189 Annalisa Napoli, Roberto Realfonzo / Structural Integrity Procedia 00 (2022) 000–000

2183

2

and mechanical properties of the composite. The most commonly used fibers in FRCM composites are basalt (B), carbon (C), alkali-resistant (AR), glass (G) and poliparafenilenbenzobisoxazole (PBO); lately, unidirectional steel (S) textiles composed of properly spaced ultra-high tensile strength steel cords are also being used for these applications. The experimental studies published on FRCM systems are less numerous than those performed on fiber reinforced polymers (FRPs), also because of the more recent market entry. By focusing on the FRCM confinement of masonry, the existing knowledge is rather poor, fragmented, and sometimes controversial; therefore, developing reliable models for the estimate of the compressive strength and ultimate strain of the FRCM confined masonry is a challenging task. On the poor availability of information, just think that about confinement applications the first experimental studies found by the authors date from 2015 only (Carloni et al. 2015); on the fragmentation of information, it is highlighted that, when a large number of tests could not be carried out, researchers often chose to investigate many parameters at the same time, thus, losing sight of the real focus of the performed study. With the purpose to organize all the available information in a systematic framework, in a recent paper the authors published a wide experimental database based on which analytical proposals for estimating the compressive strength of masonry members confined by FRCM, suitable for both any fiber (B, C, G, PBO, and S) and masonry type (natural and artificial) were proposed (Napoli and Realfonzo 2022). In order to advance the knowledge gained so far, a further upgrade of the collected database is presented in this paper, based on which new predictive models for the compressive strength of the FRCM confined masonry were derived by applying best-fit techniques to the experimental results and compared with those previously found by the authors (Napoli and Realfonzo 2022). The accuracy of the analytical proposals is also examined by considering the typology of masonry, artificial and natural, the latter poorly investigated in the literature. Even though further validation is needed based on the availability of an even more increasing amount of experimental data, the present study highlights the improved accuracy of the proposed models with respect to the companion formulas reported in some international guidelines, such as CNR-DT 215 (2018) and ACI 549.6R (2020). 2. The new database The new database includes 261 uniaxial compression tests - against 222 indicated in the previous collection (Napoli and Realfonzo 2022) - performed on FRCM confined masonry with circular (28 tests) or square/rectangular cross sections (233 tests). The addition of new data has basically expanded the data sample previously available for PBO FRCM (from 14 to 20 members) and S-FRCM systems (from 53 to 69 members) applied to square/rectangular specimens. The new members, all with square cross-section, are made of clay brick (CB) masonry and their mass density g m is supposed to be in the typical range 1500-2000 kg/m 3 , not being provided by the source’s authors; the arrangement of the masonry units follows the “ scheme 1a ” among the layouts shown in Napoli and Realfonzo (2022). Table 1 reports the main details of these additional tests. In particular, the following information is reported: • the first column includes the source from which each test has been found; • Dataset ID denotes the label of the single test; • B and H are, respectively, the width and the depth of the square/rectangular cross-section ( B = H in this case); • L and r c are, respectively, the height and the corner radius of the column; • f mat,c , f mat,b , E mat and t mat are, respectively, the compressive strength, the flexural strength, the elastic modulus in compression and the overall thickness of the FRCM matrix; •  f , f f,u , E f ,  f,u , and t f are, respectively, the density, the tensile strength, the elastic modulus and the ultimate strain of the dry strengthening sheet, and the equivalent thickness of the single layer; • n f and L b are, respectively, the number and the overlapping length of the FRCM layers; • f m0 is the compressive strength of the unconfined masonry column; • k h and k v are the coefficients of horizontal and vertical efficiency, the former calculated according to the guidelines DT 215 (2018) and the latter estimated as suggested by DT 200 (2013). To this purpose, it is highlighted that the guidelines for FRCM applications (DT 215 2018) do not account, to date, for the possibility of using a discontinuous confinement (probably due to the lack of a sufficient number of experimental data); • ̅  =     ⁄ is the normalized compressive strength of the FRCM confined masonry; • FM is the dominant failure mode exhibited by test specimens, i.e.: jacket failure (JF), debonding of the reinforcement at the overlap region (DB) and fiber-matrix slippage (S).