Effects of long-term exposure of NSM CFRP-to-concrete bond to natural and accelerated aging environments

Composite Structures 340 (2024) 118174 Available online 5 May 2024 0263-8223/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Effects of long-term exposure of NSM CFRP-to-concrete bond to natural and accelerated aging environments Aloys Dushimimana a , Jos´ e Sena-Cruz a , * , Luís Correia a , Jo˜ ao Miguel Pereira a , Susana CabralFonseca b , Ricardo Cruz a a University of Minho, ISISE/IB-S, ARISE, Department of Civil Engineering, Guimar˜ aes, Portugal b National Laboratory for Civil Engineering, Materials Department, Lisbon, Portugal ARTICLE INFO Keywords: Durability Natural aging Accelerated aging NSM CFRP-to-concrete bond Epoxy adhesive Concrete Carbon fiber ABSTRACT Carbon fiber reinforced polymer (CFRP) composites can be used to strengthen existing reinforced concrete (RC) structures. The CFRP laminate can be bonded to RC structure using epoxy adhesive via near-surface mounted (NSM) strengthening technique. However, existing literature generally lacks data about durability of NSM CFRPto-concrete bond. In this study, strengthened concrete elements were exposed to laboratory-controlled environments (at approximately 20 ◦C/55 % RH, and water immersion at 20 ◦C) and natural field environments (to promote natural aging induced mainly by carbonation, high temperatures, freeze–thaw attack, and airborne chlorides) for up to four years. Durability tests were conducted yearly for the bond and its constituent materials. The highest bond strength degradations were nearly 12 % and 9 % for the specimens immersed in water and those exposed to freeze–thaw attack, respectively. Besides, environmental conversion factors of 0.88 and 0.93 were derived from a database of existing accelerated, and natural aging data from the present work, respectively. 1. Introduction It is undoubtedly realistic to state that concrete has been proven to be a reliable construction material since its invention. Concrete has very good compressive behavior but generally poor tension properties; however, it can usually be reinforced with e.g., steel bars to become reinforced concrete (RC) with improved tensile properties. This constructive solution has been used for several decades and the growing need to extend and/or maintain the service life of these structures has led to significant investments in the strengthening of RC structures. Fiber Reinforced Polymer (FRP) composites can be used to strengthen RC structures. In particular, carbon FRP (CFRP) can be a good option, owing to its advantageous properties including high strength to weight ratio, high durability, high fatigue resistance [1], and high corrosion resistance [2], among others. CFRP composites in the form of laminate strips and sheets have generally been applied for flexural and shear strengthening of existing RC structures, respectively, where a bonding agent, such as epoxy adhesive, is typically used to bond the CFRP to concrete substrate. Near-surface mounted (NSM) is known as one of the existing strengthening techniques [3], and its durability is mainly addressed in the present work. During application of the NSM technique, the CFRP strip is basically inserted into a groove pre-cut on concrete cover [4], and is bonded to concrete with an adhesive. The NSM technique is more recent and has been repeatedly reported to possess more advantages than other techniques, such as, the externally bonded reinforced (EBR) technique [5–8]. Furthermore, NSM can additionally be prestressed, mainly to benefit from the full usage of the CFRP strain, which leads to efficient use of materials, thereby resulting in reduced crack width and increased cracking and yield response [9,10]. However, the durability of the bond in NSM technique, either non-prestressed or prestressed, still lacks literature under both accelerated aging test (AAT) and natural aging test (NAT) protocols. Besides, the durability of the materials constituting the bond in NSM technique (i.e., concrete, epoxy adhesive, and CFRP) has also not been fully investigated yet, considering NAT protocols (e.g., [11,12]), thus there is a need to conduct further related research. A significant number of studies have been conducted on the AAT of the NSM–to-concrete bond constituent materials. Starting from concrete, carbonation of concrete is an important durability factor. Studies show that high temperatures can make concrete more porous thereby leading to carbonation depth increase, which thereafter can improve concrete properties [13–15], although the carbonation is known to have * Corresponding author. E-mail address: [email protected] (J. Sena-Cruz). Contents lists available at ScienceDirect Composite Structures journal homepage: www.elsevier.com/locate/compstruct https://doi.org/10.1016/j.compstruct.2024.118174 Received 9 January 2024; Received in revised form 21 March 2024; Accepted 3 May 2024 Composite Structures 340 (2024) 118174 2 negative effects on the reinforcing steel bar [16]. In contrast, low temperatures and high relative humidity (RH) [17], as well as continuous cement hydration (leading to micropore closure) [18] can hinder the carbonation depth increase. On the other hand, carbonation depth can be high if concrete is exposed to environments with 50–70 % RH [13] (the optimum increase being at 65 % RH [17]). In a study by [19], both concrete elastic modulus and compressive strength increased after carbonation. Besides, only the compressive strength of concrete increased after exposure to carbonation in [18], and to ultraviolet (UV) radiation at certain humidity and temperature in [20]. Studies have also shown that when concrete is exposed to both carbonation and chlorides, the carbonation can release bound chlorides inward at greater depths [21,22], thereby minimizing the effects of chlorides at the concrete surface and hence protecting the concrete surface region, where FRP is normally bonded during strengthening applications. In studies addressing the durability of epoxy adhesives, it has been found that the properties of the adhesives reduce significantly after exposure to moisture or in the case of full immersion in water [11,23]. Wet-dry cycles were also found to decrease both the tensile strength and elastic modulus of the adhesive [24]. However, exposure to high temperatures improved the adhesive properties through post-curing phenomenon [25], while temperatures close to the glass transition temperature resulted in softening of the polymeric matrix [11]. Besides, high carbonation can accelerate the curing of the epoxy resin [26]. However, exposure of the adhesive to chlorides has no harmful effect on the adhesive properties [27]. Regarding the durability of CFRP composites, studies show that CFRP is generally not affected by degradation agents such as chloride exposure [11], and thermal cycles (TC) in air [28]. However, freeze- –thaw (FT) cycles may reduce both tensile strength and elongation of CFRP [29], UV radiation can affect a few microns from the CFRP surface [30], and some polymeric matrix microcracks may form due to different thermal expansion coefficients between the fiber and matrix after exposure to TC. It is worth noting that existing literature on the durability of the bond constituent materials has generally been conducted using AAT protocols and basically addresses the effect of a single degradation agent; however, in outdoor environment the materials are exposed to more than one degradation agent simultaneously. Hence there is a need to conduct studies that address the combined or synergic effect of different agents to be able to mimic what normally occurs in outdoor environments. Regarding the NSM CFRP-to-concrete bond, there are some studies [7,31–33] that address the durability of the bond under AAT conditioning. In [31], 90 wet-dry cycles lasting for 2160 h were applied on strengthened concrete elements with 60 mm and 90 mm bond lengths, 1.4 mm CFRP laminate thickness,15 mm and 25 mm groove depths, and 4 mm and 8 mm groove widths. The bond strength generally decreased after exposure, except for specimens with 25 mm groove depth and 4 mm groove width. In [33], conditioning specimens to 300 freeze–thaw cycles or immersion in salt water at 20 ◦C decreased the bond strength for concrete elements strengthened with a 1.3 mm CFRP laminate thickness. In [7], immersion in tap water at 22 ◦C (or tap water with 3.5 NaCl) for 5760 h and 11,520 h increased the bond strength of concrete elements strengthened with 1.4 mm CFRP laminate thickness. Furthermore, applying 240 or 480 wet-dry cycles on the specimens in tap water (or in tap water with 3.5 NaCl) also increased the bond strength. Besides, conditioning the specimens with temperature cycles between −15 ◦C and +60 ◦C increased the bond strength; however, temperature cycles between +20 ◦C and +80 ◦C did not affect the bond strength. Finally, conditioning the specimens to 120 freeze–thaw cycles between −18 ◦C to +20 ◦C for 5760 h also did not affect the bond strength, but doubling the number of freeze–thaw cycles between −18 ◦C to +20 ◦C for 4320 h led to a slight decrease in the bond strength. On the other hand, a few studies, e.g., [8,34,35] addressed the durability of NSM with NAT protocols. In these NAT-based studies, it was generally found that the NSM CFRP-to-concrete bond is significantly affected by moisture and freeze–thaw attack. Based on the above mentioned literature on both the NSM CFRP-toconcrete bond and the material constituting the bond, it can be noted that there is a knowledge gap in several aspects, namely: (i) existing durability data are largely based on the AAT protocols with very little knowledge on the behavior of the bond and materials under real outdoor environments; (ii) existing data, apart from being heavily dependent on the AAT protocols, are still insufficient to lead to appropriate predictions of environmental conversion factors (ECF); and, (iii) currently some standards [36,37] recommend ECFs to account for degradation of epoxy/carbon systems due to some generic exposures; however, these ECFs are neither derived from sufficient data, nor from data including both AAT and NAT protocols. Furthermore, there is no information about the factors for specific types of exposures, which is generally the real situation for most of the RC strengthened structures. Regarding the above aspects, the present work intends to contribute to increasing the number of existing data related to the durability of NSM CFRP-to-concrete bond and attempt to address the abovementioned aspects. This is achieved by firstly presenting durability data from AAT (laboratory-controlled environments: specimens conditioned at 20 ◦C/55 %RH or immersed in water at 20 ◦C) and NAT (outdoor environments: specimens kept in outdoor regions with characteristics mainly promoting carbonation, elevated temperatures, freeze–thaw attacks, and airborne chlorides) protocols. Secondly, the ECF from a database of existing AAT data and the ECF from NAT data in the present work, are derived and compared. 2. Experimental program This section first describes the properties of the materials composing the NSM CFRP-to-concrete bond followed by detailing the environments studied, and, finally, the testing methods used. 2.1. Constituent materials Three different materials involved in the bond of NSM CFRP-toconcrete system are analyzed: the concrete, the epoxy adhesive and the CFRP laminate. Details of each are given below. Concrete: The concrete used in the preparation of all specimens was in accordance with the recommendations as per [38]. All concrete properties were as shown in Table 1. Concrete cylinders (Fig. 1a) were produced and used to investigate the variation of concrete elastic modulus (E c ) and compressive strength (f cc ). Concrete prisms (Fig. 1b) were used to assess the variation of concrete pull-off strength (f ct ) and the concrete carbonation depth (C d ). Epoxy adhesive: A two-component commercial cold-curing epoxy resin-based adhesive with properties as shown in Table 1 was used to cast the dog-bone shaped specimens (Fig. 1c) and to investigate the variation of the adhesive elastic modulus (E a ) and tensile strength (f a ). This adhesive was then used as a bonding agent in NSM CFRP-toconcrete bond specimens. The adhesive is a solvent-free, thixotropic, in grey color, developed for bonding carbon fiber laminates to concrete substrates. The properties of this adhesive according to the supplier [39] are shown in Table 1. CFRP laminate: The CFRP laminate (Fig. 1d) produced by S&P [40] was used to strengthen the concrete elements and its properties are as shown in Table 1. This CFRP laminate has a rectangular cross-section of 10 mm ×1.4 mm (width ×thickness). Besides, the variations of both its elastic modulus (E f ) and tensile strength (f f ) were also investigated using CFRP strips. NSM CFRP-to-concrete bond: The properties of the specimens used to characterize the bond between the CFRP laminate and concrete are also shown in Table 1. A. Dushimimana et al. Composite Structures 340 (2024) 118174 3 2.2. Environmental exposure conditions Six different environmental exposure conditions (E1-E6) were used to study the degradation of both the NSM CFRP-to-concrete bond and its constituent materials. Different characterization tests were performed on the specimens before exposure (referred to as T0) and on the specimens collected from the different environments after one year (T1), two years (T2), three years (T3) and four years (T4) of exposure. Description of the exposure conditions in each environment is provided as follows. 2.2.1. Laboratory-controlled environments Two laboratory-controlled environments were considered. In the first environment, denoted as E1, the specimens were kept in an in-house walk-in climatic chamber with a targeted temperature and relative humidity of 20 ◦C and 55 % RH, respectively. The second environment, denoted as E2, consisted of specimens fully and continuously immersed in water at 20 ◦C (see Fig. 2a and Fig. 2b). The specimens in E2 were tested under wet state. More details about these two environments can be found in [8,11]. 2.2.2. Natural outdoor environments Four different outdoor environments were selected in different regions of Portugal with different meteorological conditions. These environments were selected with the aim of promoting the degradation of the NSM CFRP-to-concrete bond and its constituent materials. The first outdoor environment, denoted as E3, was chosen to mainly promote concrete carbonation, by exposing the specimens to air pollution (all along the year) from highway with heavy traffic and the international airport of Lisbon. The second environment (E4) was chosen to mainly promote the freeze–thaw attacks, and therefore, the specimens were placed on the highest mountain of Portugal (at the altitude of 1600 m) with some seasonal snowfall. The third environment (E5) was chosen to mainly promote the degradation due to high temperatures, so the specimens were placed in a region with yearly elevated temperatures (Fig. 2c and Fig. 2d). The fourth environment (E6) was selected just close to the Atlantic Ocean to promote airborne chloride attack, and the effects of high seasonal humidity. Although these outdoor environments were chosen to primarily promote the mentioned degradation agents, the specimens kept outdoors are likely to be affected by more than one degradation agent. For example, in E3, in addition to carbonation, high temperatures can also play an important role as the region generally experiences hot weather throughout the year. Typical variations in temperature and relative humidity in the laboratory and outdoor environments are shown in Fig. 3. Particularly, in E1, the relative humidity varies in the range of 45–77 % RH, which promotes concrete carbonation [13,17]. The RH variations are mainly due to inevitable opening/closing of the door of the climatic chamber for maintenance and data collection and fall within the moderate humidity range (exposure class XC3), according to Table 1 Properties of the NSM CFRP-to-concrete bond specimens and its constituent materials. Concrete Epoxy adhesive Concrete class C30/37 (cylinder/ cube in [MPa]) Type of adhesive Cold-curing S&P Resin 220 Max aggregate size [mm] 12.5 Density, at 23 ◦C [g/ cm 3 ] 1.7–1.8 Cement type CEM II/A–L 42.5R Flexural elastic modulus [GPa] >7.1 Slump [mm] 160–210 (slump class S4) Tensile strength [MPa] 19.9 (after 7d curing at 20 ◦C) Water-to-cement ratio 0.4 Glass transition temperature [T g ] in ◦C 46.2 (after 7d curing at 23 ◦C) Exposure class XC4(P) Compressive strength [MPa] >70 E c [GPa] 29.1 (at 28 days) Shear strength [MPa] >26 f cc [MPa] 41.5 (at 28 days) BS by pull-off, on concrete [MPa] 3 (after 3d curing at 20 ◦C) CFRP laminate CFRP-to-concrete bond Type and trademark S&P clever (CFK 150/2000) CFRP cross-section [mm 2 ] 10 ×1.4 Prefabricated by Pultrusion Concrete cube dimensions [mm 2 ] 200 ×200 Fiber orientation Unidirectional Bond length [mm] 60 Fiber content [%] 68 Groove depth [mm] 15 Fiber matrix Vinyl ester resin (with T g ≈85◦C) Groove width [mm] 5 External surface Black, smooth Epoxy adhesive thickness [mm] 5 / 3.6 at the CFRP level Elastic modulus [GPa] >170 Epoxy adhesive depth [mm] 15 Tensile strength [MPa] >2000 Type of test Single-lap shear test Notes: f cc : average compressive strength; E c : average elastic modulus; BS: bond strength; T g : glass transition temperature; 3d: three days; 7d: seven days. Fig. 1. NSM CFRP-to-concrete bond constituent materials: (a) concrete cylinder, (b) concrete prism, (c) epoxy adhesive dog bone shaped specimens and (d) CFRP laminate strips. A. Dushimimana et al. Composite Structures 340 (2024) 118174 4 the Eurocode 1992–1-1:2004 (E) [41]. These actions may have promoted the flow of CO 2 from the laboratory environment into the climate chamber. Furthermore, a significant difference in terms of temperature and relative humidity can also be observed in outdoor environments (E4 and E5). More details can be found in [42], where a general view is that E4 and E6 tend to experience low temperatures and high relative humidity; while E3 and E5 experience high temperatures and low relative humidity. The variation of annual temperatures and relative humidity is tabulated in Table 2, where it can be noted that the maximum temperatures from E3 and E5 were the highest, as expected due to their high daily temperatures as compared to other environments. 2.3. Characterization test methods This section provides information about the tests adopted for the characterization of both the NSM CFRP-to-concrete bond and its constituent materials. In all the cases, specimens were tested before ageing (T0) and after being exposed to the environments during one (T1), two (T2), three (T3) and fours (T4) years. In the case of environment E2, all the specimens were tested in wet state. 2.3.1. Concrete A universal testing machine (UTM) with a maximum load capacity of 2000 kN was used to perform compression tests for the concrete elastic modulus (E c ) and compressive strength (f cc ). Under non-destructive test (Fig. 4a), three LVDTs spaced at 120◦were used to measure the data required to estimate the E c according to EN 12390–13:2013 [43]. Furthermore, a destructive test (Fig. 4b) was performed on the same specimen to determine the f cc according to NP EN 12390–3:2011 [44]. A total of 75 specimens were tested to determine both the E c and the f cc . Additionally, the pull-off test was performed using the DYNA Z5 testing machine according to EN 1542:1999 [45]. After allowing the epoxy adhesive to cure for 7 days at room temperature, metal dollies bonded to the concrete cores (using the epoxy adhesive) were manually loaded (Fig. 4c) to pull-off the formed cylinder, until the cylinder’ split occurred (Fig. 4d). A total of 100 specimens were tested (series of 4 tests). Further details of this type of test can be found in [8]. In addition to these tests, the carbonation ingress of concrete was also evaluated by spraying phenolphthalein indicator on the cylindrical cores drilled from concrete prisms (series of 4 specimens). After spraying, the clear areas of the specimens indicated the concrete carbonated areas, while the purple-red areas indicated the concrete non-carbonated areas. These tests were performed according to [8], to measure the Fig. 2. Typical examples of studied environments: (a) exposure of the constituent materials specimens to water immersion (concrete, epoxy adhesive and CFRP laminate); (b) exposure of the specimens of NSM CFRP-to-concrete bond to water immersion (E2); exposure of (c) constituent materials and (d) NSM CFRP-toconcrete bond specimens to natural outdoor environments (E5). Fig. 3. Typical meteorological records from laboratory (E1) and outdoor (E5, E6) environments. A. Dushimimana et al. Composite Structures 340 (2024) 118174 5 carbonation depth. 2.3.2. Epoxy adhesive and CFRP laminate Tensile tests of both epoxy adhesives and CFRP laminates were performed using MTS UTS machine (Fig. 5a). A total of 125 epoxy adhesive specimens (series of at least 5 specimens for each environment) were tested according to EN ISO 527–2:2012 [46] as shown in Fig. 5b and the elastic modulus was determined according to EN ISO 527–2:2012 [46] by calculating the slope of the secant line on the stress–strain curve between 0.05 % and 0.25 % of the strains. On the other hand, a total of 150 CFRP laminate specimens were also tested in the same configuration (Fig. 5c) as that of the adhesives and according to EN ISO 527–5:2009 [47]. 2.3.3. NSM CFRP-to-concrete bond A total of 50 concrete cubes, each with 200 mm, were used to prepare the specimens strengthened according to NSM technique. The concrete cubes were strengthened with CFRP strips, each being strengthened by 2 CFRP laminates with a bond length of 60 mm (Fig. 6a). For that purpose, two opposed faces parallel to the casting direction were used. The bond length adopted aimed at avoiding the CFRP laminate failure and also to be large enough to represent the system and minimize some inevitable effects such as geometric irregularities [8]. Two of these concrete blocks (4 strengthening systems) were tested at the beginning to serve as reference specimens, whereas the remaining ones were exposed to different environments and tested after exposure. A total of 100 pull-out tests were performed. The test setup is shown in Fig. 6b, from which the main parts can be highlighted: (i) LVDT support fixation (Fig. 6c); (ii) horizontal and vertical movements restraint (Fig. 6d and Fig. 6e); and (iii) LVDT installation (Fig. 6e). These tests were conducted on four specimens each year (i.e., from T1 to T4). A servo-controlled equipment was used, with the applied force measured by a load cell with a maximum capacity of 200 kN. The tests were performed under displacement control at the loaded end with a rate of 2 μ m/s, using a displacement measured through the LVDT1 (Fig. 6a) placed at the loaded end section as a control variable. 3. Results and discussion This section presents the results from the different tests performed (as described in Section 2), followed by a comprehensive analysis and discussion. Furthermore, comparative studies are also performed between the results from this work and those from the existing literature to derive and propose the environmental conversion factors. 3.1. Test results from the bond constituent materials 3.1.1. Concrete Concrete compressive strength (f cc ), elastic modulus (E c ), pull-off strength (f ct ), and carbonation depth (C d ) for all testing time series (T0-T4) and all studied environments (E1-E6) are plotted in Fig. 7a while average values are shown in Table 3. The results show that f ct Table 2 Variation of annual temperatures and relative humidity in the studied environments. Variable / Environment E1 E2 E3 E4 E5 E6 Temp [◦C] Year 1 (T1) Max 20.2 (22.5) 20.2 (20.3) 23.1 (46.2) 17.8 (32.4) 24.0 (44.6) 21.6 (36.0) Min 19.8 (13.5) 20.1 (19.0) 13.4 (3.3) 11.1 (−4.7) 11.1 (−1.9) 11.8 (1.5) Avg 19.9 20.0 17.4 14.3 17.4 16.15 RH [%] Year 1 (T1) Max 62.8 (79.5) 100.0 87.9 (100.0) 83.3 (100.0) 84.9 (100.0) 88.7 (100.0) Min 57.8 (31.0) 100.0 48.8 (11.0) 50.9 (4.0) 40.3 (9.0) 59.8 (18.0) Avg 60.0 100.0 70.7 67.2 63.5 76.1 Temp [◦C] Year 2 (T2) Max 20.4 (22.0) 20.2 (20.4) 22.7 (39.7) 14.0 (29.6) 23.6 (39.9) 22.5 (39.5) Min 20.1 (19.0) 20.5 (19.5) 13.9 (4.3) 7.4 (−4.6) 8.9 (0.3) 13.0 (2.0) Avg 20.2 20.6 17.5 10.6 17.3 17.2 RH [%] Year 2 (T2) Max 63.1 (77.5) 100.0 90.1 (100.0) 87.6 (100.0) 87.9 (100.0) 90.5 (100.0) Min 60.1 (41.5) 100.0 52.9 (16.0) 54.7 (4.0) 42.0 (11.0) 63.5 (28.5) Avg 61.8 100.0 74.4 72.8 67.1 78.6 Temp [◦C] Year 3 (T3) Max 20.1 (22.0) 21.0 (22.0) 22.8 (40.7) 13.2 (29.5) 22.8 (41.4) 16.9 (31.1) Min 19.8 (17.0) 20.2 (19.6) 13.5 (1.4) 4.0 (−6.4) 11.4 (−4.1) 9.1 (−1.1) Avg 19.9 20.3 17.3 9.7 17.1 13.0 RH [%] Year 3 (T3) Max 63.9 (77.5) 100.0 87.9 (100.0) 86.0 (100.0) 86.2 (100.0) 93.0 (99.0) Min 61.7 (43.0) 100.0 48.8 (16.0) 50.5 (6.0) 45.6 (9.0) 61.5 (23.0) Avg 62.9 100.0 70.7 71.3 66.5 79.4 Temp [◦C] Year 4 (T4) Max 20.5 (23.0) 20.2 (20.3) −20.7 (31.1) 25.5 (46.0) 19.9 (34.0) Min 20.1 (17.0) 20.3 (19.5) −12.7 (2.0) 13.9 (2.0) 11.1 (2.0) Avg 20.3 20.4 −16.5 19.3 15.1 RH [%] Year 4 (T4) Max 62.8 (77.5) 100.0 −83.3 (100.0) 84.9 (88.5) 88.7 (100.0) Min 57.8 (47.0) 100.0 −50.9 (4.0) 40.3 (10.0) 59.8 (22.0) Avg 60.0 100.0 −67.2 63.4 76.1 Notes: RH: Relative humidity; Temp: Temperature; Max/Min/Avg: yearly maximum/minimum/average value; the value in parentheses stands for the annual peak value. Fig. 4. Tests for concrete characterization: (a) Elastic modulus setup; (b) compression test setup; (c) Sandblasting and creating the cylindrical cores; (d) concrete pulloff test machine. A. Dushimimana et al. Composite Structures 340 (2024) 118174 6 decreased in all studied environments. On the other hand, both f cc and E c generally improved with time in all outdoor environments. The highest improvements in f cc and E c were found in E4 and E6 at T2, with increases of 21.2 % and 12.0 % in E4, and 20.9 % and 12.0 % in E6, respectively. The observed increase of f cc and E c in different environments can be attributed to different factors such as carbonation, temperature variations and continuation of cement hydration. In fact, the C d is observed to increase linearly with time in E1 (Fig. 7a) due to the relative humidity range [45–77 % RH], see Fig. 3, which is already known to favor the ingress of CO 2 [48]. Fig. 5. Tensile tests using a MTS UTS machine: (a) overview of MTS with video extensometer; (b) epoxy adhesive test; (c) CFRP laminate test. Fig. 6. NSM specimen and single-lap shear test setup: (a) specimen’s geometry and test configuration (NSM); (b) photograph of the test; (c) mounting LVDTs supports; (d) installing the specimen; (e) positioning of LVDTs. Note: all units in [mm]. A. Dushimimana et al. Composite Structures 340 (2024) 118174 7 36 42 47 52 24 27 30 33 36 2 3 4 5 6 8 11 14 T0 T1 T2 T3 T4 T0 T1 T2 T3 T4 Compressive strength [MPa] Carbonation [mm] Elastic modulus [GPa] T0 T1 T2 T3 T4 Tensile strength [MPa] T0 T1 T2 T3 T4 E1 E2 E3 E4 E5 E6 Environmental exposure Concrete properties 2030 2320 2610 2900 156 182 208 234 8 16 24 32 3 6 8 11 CFRP Tensile strength [MPa] T0 T1 T2 T3 T4 T0 T1 T2 T3 T4 CFRP Elastic modulus [GPa] T0 T1 T2 T3 T4 Adhesive Tensile strength [MPa] T0 T1 T2 T3 T4 Adhesive Elastic modulus [GPa] seitreporpevisehdayxopEdnaPRFC E1 E2 E3 E4 E5 E6 Environmental exposure Fig. 7. Variation of material properties under different environments from 0 to 4 years of exposure: (a) concrete, and (b) epoxy adhesive and CFRP laminate. Table 3 Average values of concrete compressive strength, elastic modulus, pull-off strength, and carbonation depth after 0 (T0), 1 (T1), 2 (T2), 3 (T3), and 4 (T4) years of different environmental exposures (E1 to E6). Environment T0 T1 T2 T3 T4 T0 T1 T2 T3 T4 Compressive strength f cc [MPa] Elastic modulus E c [GPa] REF 41.5 (4.4) −- −- −- −- −- −- −- −- −- E1 −- 42.8 (2.4) 43.3 (1.4) 42.8 (3.8) 40.3 (1.4) 29.1 (5.0) 28.0 (0.7) 28.7 (1.7) 28.5 (1.6) 27.7 (0.4) E2 −- 40.7 (0.7) 38.7 (2.9) 41.9 (1.5) 41.1 (1.8) −- 28.2 (2.8) 27.7 (2.0) 28.1 (7.2) 30.8 (4.5) E3 −- 46.3 (0.9) 46.0 (3.4) 48.4 (2.5) 46.6 (0.6) −- 29.4 (0.8) 27.0 (8.0) 31.7 (3.4) 29.7 (2.9) E4 −- 46.5 (3.4) 50.3 (0.8) 49.2 (4.8) 48.6 (0.7) −- 28.6 (3.4) 29.7 (12.1) 32.6 (2.7) 30.9 (4.0) E5 −- 44.9 (1.0) 48.2 (1.5) 44.8 (0.7) 45.1 (0.9) −- 28.6 (2.6) 29.1 (1.6) 30.0 (0.3) 29.6 (4.2) E6 −- 47.1 (2.4) 50.2 (1.3) 47.6 (1.4) 47.5 (0.5) −- 30.2 (3.7) 32.6 (0.5) 30.8 (1.5) 32.1 (2.6) Pull-off strength f ct [MPa] Carbonation depth C d [mm] REF 3.4 (13.3) −- −- −- −- 0.0 −- −- −- −- E1 −- 2.9 (10.4) 3.2 (1.3) 2.9 (18.4) 2.9 (10.5) −- 7.4 (19.9) 7.7 (14.8) 8.4 (9.2) 8.9 (12.8) E2 −- 2.5 (5.3) 2.3 (11.8) 2.5 (13.1) 2.6 (7.4) −- 7.2 (15.2) 5.5 (17.7) 7.6 (16.4) 6.4 (15.3) E3 −- 3.2 (3.6) 2.8 (9.7) 2.6 (20.1) 3.0 (4.7) −- 10.1 (5.5) 9.3 (18.2) 9.6 (23.4) 9.0 (18.8) E4 −- 3.2 (13.5) 3.1 (5.3) 2.9 (12.2) 2.9 (0.1) −- 7.8 (7.5) 8.4 (14.1) 6.5 (8.4) 4.9 (24.1) E5 −- 3.1 (6.3) 3.1 (16.4) 2.7 (5.7) 2.6 (7.7) −- 7.8 (10.9) 8.0 (12.5) 10.3 (12.4) 8.0 (12.5) E6 −- 2.7 (4.4) 2.8 (8.8) 2.9 (8.1) 2.7 (10.5) −- 8.0 (14.1) 7.3 (13.9) 4.9 (22.9) 8.0 (12.8) Notes: all values in parentheses express coefficient of variation; REF: Reference values from the specimens tested before exposure (i.e., at T0). A. Dushimimana et al. Composite Structures 340 (2024) 118174 8 However, trends with lower C d values can be observed from E2 and E4 probably due to increased humidity and moisture in these two environments. This agrees well with the findings by [49] who reported that the increased humidity, rain, and snow slowed down the ingress of CO 2 after exposing concrete to outdoor environments for up to 4 years. Most importantly, the specimens in E2 and E4 are generally found to be partially carbonated (Fig. 8a). Partial carbonation has negligible effects on the concrete properties [50,51] for high relative humidity environments, hence the effects of carbonation in E2 and E4 on the f cc and E c are negligible. The highest C d is observed in E3 as expected although there is a surprising decrease in the last year T4 that may be due to climate change issues. Besides, even in other outdoor environments (E5, E6) the C d is observed to be comparable to that in E3, which may reflect the effects of climate change that causes the abundance of CO 2 in the atmosphere. On the other hand, the observed increase in f cc and E c in E6 may be further attributed to the synergistic effect between carbonation and chlorides as previously reported in [21]. Studies also show that high temperatures can make concrete more porous thereby increasing the CO 2 penetration [13,14], which may be the reason for the high C d observed in E5. Hence, the presence of carbonation in E3, E5, E6 can be considered as one of the possible factors that increased the concrete f cc and E c . This is in line with other previous studies where carbonation was found to increase the f cc [18] and both the f cc and E c [19]. On the other hand, as previously described for outdoor environments, high temperatures with low RH were recorded in E3 and E5, so this may also have played a major role in increasing the f cc and E c . This is in agreement with a study by [52], where high temperature variations increased the f cc . Hence, it can be noted that there is a synergistic effect between high temperatures and carbonation, but high temperatures may have played a greater role than the presence of carbonation due to the inconsistent trend of the latter, as shown in Fig. 7a. Contrary, the effects of temperatures on the increase in the f cc and E c may be thought negligible in E4 and E6 as these environments experienced lower temperature values, hence other factors may have contributed to the observed increased concrete properties. For example, as a result of the high RH in E4 and E6, continued hydration of cement may occur over time, with the increase of the number of C-S-H silicates, which then occupy the empty voids and push out the diffused moisture in the voids, thereby reducing the porosity and improving concrete properties. Increase of the C-S-H silicates is already known to improve the f cc [53,54]. Increased curing moisture has also been linked to increase in the f cc and E c [19]. To summarize, the above results show that the increase in concrete properties can result from a combination of more than one factor. In fact, these can be the leading factors in each environment: (i) carbonation in E1; (ii) the competing mechanisms between the increased porosity versus cement hydration in E2; (iii) the synergistic effect between carbonation and high temperatures in E3, (iv) the synergistic effect between the continuation of cement hydration and increased humidity in E4; (v) the synergistic effect between high temperatures and carbonation in E5; and, (vi) the synergistic effect of carbonation and chlorides and continuation of cement hydration in E6. Furthermore, comparing the degradation agents, the presence of cement hydration resulting from high relative humidity regions leads to the highest improvement in the concrete f cc and E c . The typical failure modes from concrete compression and tensile tests are shown in Fig. 8b, Fig. 8c and Fig. 8d, respectively. In particular, it can be seen that the failure mode from tensile tests occurs deep inside the concrete, as mentioned earlier, which can lead to misleading values since the inner region is ideally not exposed. Finally, the duration and Fig. 8. (a) Carbonation of concrete; typical failure modes from: (b) concrete under compression; (c,d) concrete under tension (pull-off tests); (e) epoxy adhesive under tension; and, (f to h) CFRP laminate under tension. A. Dushimimana et al. Composite Structures 340 (2024) 118174 9 severity of exposure may have a greater impact on the carbonation depth and elastic modulus of the concrete compared to other properties, due to significant fluctuations rather than a gradual increase or decrease over time. It is possible that the exposure duration was shorter or longer in year i than in year i-1 (for i =0,…,4), which could have caused these fluctuations. Similarly, fluctuations may also result from differences in exposure severity (i.e., more severe exposures in year i can be thought to lead to higher changes in the properties than in year i-1). 3.1.2. Epoxy adhesive The average values of the tensile strength (f a ), average elastic modulus (E a ), and ultimate tensile strain (%) of the epoxy adhesive for all testing times and all environments are shown in Fig. 7b, their values in Table 4, and their failure modes in Fig. 8e. The increase in E a from T0 to T1 in E1, E3, E4, and E5, is mainly due to the post-curing phenomenon. However, there was a significant decrease in the E a in E2 which can be attributed to the plasticization effect, and a slight decrease in E6 due to moisture, as the effects of chlorides on the adhesive properties are negligible [27]. Furthermore, a general decrease in the E a can be observed during the following consecutive years, except for E2 that showed marginal variations after T1. The adhesive f a shows a similar trend as that of the E a ; however, the effects of the post-curing phenomenon are more pronounced for the latter. The highest decreases in E a and f a are found from the specimens immersed in water (E2), with reductions of 75.4 % (at T2) and 66.3 % (at T2), respectively, compared to the value recorded at the initial stage (T0). The reasons for the observed decrease in both the E a and f a may be as follows: (i) marginal variations in E1 are due to the effects of increased carbonation, which tends to cause the epoxy resin to cure faster as previously observed in [26], (ii) negative effects of water ingress in E2 and moisture effects in both E4 and E6, which were also found by [11,23], also, in a study by [24], where both the adhesive E a and f a decreased due to exposure to wet-dry cycles (case of E4), (iii) in E3, temperatures higher than Tg and UV (ultra-violet) radiation can be considered as the main factors that deteriorated the properties of the epoxy adhesive. The failure mode (Fig. 8e) of the adhesive was always an abrupt break in the testing zone of constant cross-section. Table 4 Average values of CFRP and epoxy adhesives’ tensile strength, elastic modulus, tensile strain after 0 (T0), 1 (T1), 2 (T2), 3 (T3), and 4 (T4) years of different environmental exposures (E1 to E6). Environment T0 T1 T2 T3 T4 T0 T1 T2 T3 T4 Adhesive: Tensile strength f a [MPa] CFRP laminate: Tensile strength f f [MPa] REF 19.9 (3.0) −- −- −- −- 2405 (3.8) −- −- −- −- E1 −- 19.5 (1.8) 18.2 (2.8) 19.8 (4.9) 16.3 (14.6) −- 2674 (2.72) 2528 (4.4) 2469 (6.4) 2484 (3.0) E2 −- 7.2 (3.1) 6.7 (2.7) 7.4 (7.1) 8.4 (3.6) −- 2688 (3.4) 2460 (7.1) 2713 (4.5) 2522 (7.1) E3 −- 19.9 (3.1) 17.4 (5.3) 16.7 (5.9) −- −- 2792 (3.7) 2590 (5.4) 2546 (5.1) 2427 (6.8) E4 −- 20.1 (3.4) 17.2 (4.3) 16.5 (9.8) 15.7 (20.1) −- 2757 (2.9) 2617 (4.5) 2492 (5.0) 2516 (3.9) E5 −- 21.9 (5.2) 18.0 (3.6) 17.7 (6.5) 17.0 (5.8) −- 2611 (5.0) 2619 (5.3) 2427 (4.1) 2575 (3.9) E6 −- 17.7 (6.4) 15.8 (4.3) 18.0 (4.2) 15.3 (2.6) −- 2667 (3.0) 2640 (2.9) 2561 (2.8) 2605 (5.1) Adhesive: Elastic modulus E a [GPa] CFRP laminate: Elastic modulus E f [GPa] REF 6.5 (3.0) 164 (1.2) E1 −- 6.6 (1.3) 6.1 (1.4) 6.5 (6.0) 5.6 (14.6) −- 179 (1.6) 165 (2.7) 170 (2.7) 172 (9.1) E2 −- 1.9 (5.1) 1.6 (4.0) 1.9 (14.6) 2.0 (12.4) −- 174 (0.7) 168 (0.6) 175 (1.8) 173 (3.2) E3 −- 6.7 (4.4) 6.0 (5.4) 5.3 (8.9) −- −- 177 (1.8) 172 (1.1) 171(3.3) 180(5.3) E4 −- 7.2 (1.4) 5.4 (6.9) 5.8 (4.0) 5.4 (21.1) −- 175 (1.8) 174 (4.5) 180 (5.7) 162 (5.3) E5 −- 7.5 (5.7) 6.1 (5.0) 5.8 (6.5) 5.7 (7.0) −- 173 (2.0) 176 (1.5) 181 (4.5) 173 (4.8) E6 −- 6.2 (5.4) 5.0 (10.0) 6.2 (6.2) 4.8 (10.0) −- 171 (1.4) 165 (2.5) 163 (2.8) 167 (5.0) Adhesive: Ultimate tensile strain [%] CFRP laminate: Ultimate tensile strain [%] REF 0.4 (6.2) −- −- −- −- 14.6 (3.8) −- −- −- −- E1 −- 0.4 (13.0) 0.3 (11.7) 0.3 (8.8) 0.4 (11.3) −- 14.9 (3.2) 15.3 (6.1) 14.5 (7.3) 14.8 (5.3) E2 −- 1.1 (21.4) 1.1 (11.9) 1.0 (25.5) 0.9 (15.3) −- 15.5 (2.9) 16.0 (12.5) 15.5 (6.1) 14.9 (1.3) E3 −- 0.3 (11.1) 0.3 (19.1) 0.3 (10.1) −- −- 15.8 (3.8) 15.1 (5.1) 14.6 (8.1) 14.3 (7.0) E4 −- 0.3 (11.3) 0.3 (12.8) 0.3 (24.6) 0.3 (24.8) −- 15.78 (2.3) 15.0 (4.5) 13.8 (4.3) 15.4 (1.1) E5 −- 0.3 (11.2) 0.4 (13.1) 0.3 (11.5) 0.4 (10.5) −- 15.1 (4.6) 14.9 (2.5) 13.5 (7.0) 15.4 (3.1) E6 −- 0.3 (4.3) 0.3 (12.9) 0.3 (17.4) 0.4 (14.2) −- 15.6 (2.9) 16.0 (1.9) 15.7 (2.8) 15.9 (5.0) Note: all values in parentheses express coefficient of variation in percent; REF: Reference values from the specimens tested at the beginning (before exposure) i.e., at T0. A. Dushimimana et al. Composite Structures 340 (2024) 118174 16 strength in E1 generally improved, showing an increase of approximately 10 % in the last year. Similarly, the bond strength in outdoor environments (E5 and E6) generally improved, indicating that exposure to high temperatures (with presence of carbonation) and airborne chlorides (with presence of both carbonation and high humidity) can generally be beneficial to the bond properties in NSM technique. Also, change of concrete properties played a major role in improving bond properties in E6. The effects of atmospheric CO 2 were minimal in E3, with very insignificant changes in the bond strength. On the other hand, the bond stiffness generally showed a decrease (with approximatively 0.7 retentions in all environments) during the first year, followed by a general increase in the following years, remarkable for the specimens in laboratory-controlled environments. 4. Failure modes: visual examination of failure modes in NSM showed that the failure mode at initial time (T0) changed after exposure. In general, the failure at adhesive-CFRP interface (F/A) dominated in E1 and E5; the combination of adhesive failure (A) with concrete cohesive failure (CC) or with concrete splitting (CS) dominated in E2; F/A dominated in E3, a combination of F/A and CS dominated in E4 mainly because of the long-lasting negative effects of freeze–thaw on both adhesive and CFRP; F/A or a combination of F/A and CS dominated in E6. 5. The bond strength retentions from the AAT data (from existing studies) vary significantly, ranging approximately between 0.8 and 1.2, which shows a remarkable dispersion. Also, most of AAT bond strength retentions from literature did not agree with those of natural aging test (NAT) from the present work. Attempts were made to derive the ECF. Considering 10 % of the non-conservative estimates from the AAT and NAT data as the maximum allowable percentage, the conversion factors of 0.88 and 0.93 are suggested for the former and the latter, respectively. Comparing these two factors, it can be seen that the existing accelerated aging test data tends to overestimate the bond strength degradation rate for up to 4 years. As a main recommendation, the authors suggest that studies with NAT data for longer periods (e.g., more than 4 years) of exposure are required to be able to appropriately predict the conversion factors. CRediT authorship contribution statement Aloys Dushimimana: Writing – original draft, Methodology, Investigation, Formal analysis, Conceptualization. Jos´ e Sena-Cruz: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Investigation, Funding acquisition. Luís Correia: Investigation, Formal analysis, Methodology, Supervision, Validation, Writing – review & editing. Jo˜ ao Miguel Pereira: Supervision, Validation, Writing – review & editing. Susana Cabral-Fonseca: Investigation, Supervision, Validation, Writing – review & editing. Ricardo Cruz: Writing – review & editing, Validation, Investigation. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data availability Data will be made available on request. Acknowledgements This work was carried out in the scope of the project FRPLongDur POCI-01-0145-FEDER-016900 (FCT PTDC/ECM-EST/1282/2014) and DURABLE-FRP (PTDC/ECI-EGC/4609/2020 – DOI 10.54499/PTDC/ ECI-EGC/4609/2020) funded by national funds through the Foundation for Science and Technology (FCT) and co-financed by the European Fund of the Regional Development (FEDER) through the Operational Program for Competitiveness and Internationalization (POCI) and the Lisbon Regional Operational Program and, partially financed by the project POCI-01-0145-FEDER-007633 and by FCT/MCTES through national funds (PIDDAC) under the R&D Unit Institute for Sustainability and Innovation in Structural Engineering (ISISE), under reference UIDB/ 04029/2020 (DOI 10.54499/UIDB/04029/2020), and ARISE under reference LA/P/0112/2020. Furthermore, this work is financed by national funds through FCT under grant agreement DFA/BD/08403/2021 attributed to the first author. The authors also acknowledge all the involved companies: S&P Clever Reinforcement Iberica Lda., Portuguese Institute for Sea and Atmosphere, I.P. (IPMA, IP), Sika Portugal – Produtos Construç˜ ao e Indústria, S.A., Hilti Portugal – Produtos e Serviços, Lda., Artecanter – Indústria Criativa, Lda., Tecnipor – Gomes&Taveira Lda., Vialam – Indústrias Metalúrgicas e Metalomecˆ anicas, Lda., Laborat´ orio Nacional de Engenharia Civil (LNEC, IP), EDP – Energias de Portugal and APDL, SA. Annex A. Single-lap shear test results for all the specimens tested from T0 to T4. Time Specimen F lmax [kN] s lmax [mm] F lmax [kN] s lmax [mm] F lmax [kN] s lmax [mm] T0 No.1 24.47 0.59 − − − − No.2 27.55 0.53 − − − − No.3 28.81 0.53 − − − − No.4 28.99 0.50 − − − − E1 E1 E2 E2 E3 E3 T1 No.1 28.30 0.65 29.05 0.51 26.98 0.57 No.2 28.26 0.63 25.53 0.57 29.54 0.71 No.3 29.16 0.60 24.77 0.49 28.40 0.55 No.4 29.50 0.60 25.36 0.49 26.06 0.43 T2 No.1 28.07 0.56 24.18 0.42 25.68 0.57 No.2 28.48 0.63 25.41 0.61 27.11 0.51 No.3 28.30 0.57 24.62 0.56 27.94 0.48 No.4 27.27 0.47 24.18 0.35 27.11 0.60 T3 No.1 29.12 0.56 24.11 0.43 28.78 0.64 (continued on next page) A. Dushimimana et al. Composite Structures 340 (2024) 118174 17 (continued) No.2 28.76 0.63 24.99 0.61 26.28 0.64 No.3 28.77 0.57 24.20 0.56 26.83 0.40 No.4 29.43 0.47 24.76 0.47 28.07 0.68 T4 No.1 − − 25.94 0.59 26.16 0.24 No.2 30.66 0.63 25.36 0.50 27.70 0.55 No.3 31.01 0.70 24.30 0.39 28.76 0.44 No.4 31.18 0.43 24.20 0.44 27.96 0.55 E4 E4 E5 E5 E6 E6 T1 No.1 27.00 0.61 29.29 0.68 30.26 0.59 No.2 28.28 0.68 29.98 0.60 29.77 0.66 No.3 27.39 0.63 30.04 0.56 30.96 0.75 No.4 28.16 0.58 29.39 0.73 31.17 0.54 T2 No.1 26.70 0.54 29.04 0.58 29.90 0.65 No.2 24.86 0.46 27.01 0.55 27.11 0.47 No.3 26.50 0.55 27.98 0.58 30.99 0.65 No.4 24.66 0.52 28.40 0.56 29.65 0.54 T3 No.1 26.86 0.60 28.63 0.82 28.78 0.57 No.2 26.59 0.65 29.26 0.60 30.11 0.57 No.3 25.53 0.54 29.66 0.65 28.43 0.41 No.4 26.30 0.52 25.02 0.49 30.01 0.65 T4 No.1 26.46 0.65 28.36 0.19 28.78 0.57 No.2 25.68 0.41 27.65 0.41 30.11 0.57 No.3 24.61 0.44 − − 28.43 0.41 No.4 25.73 0.53 28.86 0.54 30.01 0.65 Notes: F lmax : maximum pull-out force; s lmax : loaded end slip at F lmax ; E1-E6: studied environments; T0: initial time before exposure; T1-T4: exposure time in years. Annex B. Single-lap shear test results: Pull-out force vs. loaded end slip curves for all specimens tested up to 4 years. 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 Specimen No.1 Specimen No.2 Specimen No.3 Specimen No.4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T1-E1 Pull-out force [kN]Pull-out force [kN] 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T1-E2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T1-E3 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T0 T1-E4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T1-E5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T1-E6 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T2-E1 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T2-E2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T2-E3 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T2-E4 Pull-out force [kN] 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T2-E5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T2-E6 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T3-E1 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T3-E2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T3-E3 Pull-out force [kN] 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T3-E4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T3-E5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T3-E6 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T4-E1 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T4-E2 Pull-out force [kN] 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T4-E3 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T4-E4 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T4-E5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 T4-E6 Loaded end slip [mm]Loaded end slip [mm] Loaded end slip [mm] Loaded end slip [mm]Loaded end slip [mm] . A. Dushimimana et al. Composite Structures 340 (2024) 118174 18 References [1] T. Westphal, P. Bortolotti, R.P.L. Nijssen, Carbon glass hybrid materials for wind turbine rotor blades, 2013. http://resolver.tudelft.nl/uuid:f214ffc8-60c5-4855 -a4b9-db3be9991193. [2] Abbood IS, Odaa SA, Hasan KF, Jasim MA. Properties evaluation of fiber reinforced polymers and their constituent materials used in structures - A review. Mater Today Proc 2021;43:1003–8. https://doi.org/10.1016/j.matpr.2020.07.636. [3] CEB-FIP, fib90 Bulletin:Externally Applied FRP reinforcement for Concrete structures, 2019. [4] Cruz JS, Barros J. Modeling of bond between near-surface mounted CFRP laminate strips and concrete. Comput Struct 2004;82:1513–21. https://doi.org/10.1016/j. compstruc.2004.03.047. [5] De Lorenzis L, Teng JG. Near-surface mounted FRP reinforcement: An emerging technique for strengthening structures. Compos Part B Eng 2007;38:119–43. https://doi.org/10.1016/j.compositesb.2006.08.003. [6] Coelho MRF, Sena-Cruz JM, Neves LAC. A review on the bond behavior of FRP NSM systems in concrete. Constr Build Mater 2015;93:1157–69. https://doi.org/ 10.1016/j.conbuildmat.2015.05.010. [7] Fernandes P, Sena-Cruz J, Xavier J, Silva P, Pereira E, Cruz J. Durability of bond in NSM CFRP-concrete systems under different environmental conditions. Compos Part B Eng 2018;138:19–34. https://doi.org/10.1016/j.compositesb.2017.11.022. [8] Cruz R, Correia L, Cabral-Fonseca S, Sena-Cruz J. Durability of bond between NSM CFRP strips and concrete under real-time field and laboratory accelerated conditioning. J Compos Constr 2022;26:1–15. https://doi.org/10.1061/(asce) cc.1943-5614.0001262. [9] Correia L, Teixeira T, Michels J, Almeida JAPP, Sena-Cruz J. Flexural behaviour of RC slabs strengthened with prestressed CFRP strips using different anchorage systems. Compos Part B 2015;81:158–70. https://doi.org/10.1016/j. compositesb.2015.07.011. [10] El-Hacha R, Wight R, Green M. Prestressed fibre-reinforced polymer laminates for strengthening structures. Prog Struct Eng Mater 2001;3:111–21. https://doi.org/ 10.1002/pse.76. [11] Cruz R, Correia L, Dushimimana A, Cabral-Fonseca S, Sena-Cruz J. Durability of epoxy adhesives and carbon fibre reinforced polymer laminates used in strengthening systems: Accelerated ageing versus natural ageing. Materials (Basel) 2021;14. https://doi.org/10.3390/ma14061533. [12] Tatar J, Brenkus NR. Performance of FRP-strengthened reinforced concrete bridge girders after 12 years of service in coastal florida. J Compos Constr 2021;25. https://doi.org/10.1061/(asce)cc.1943-5614.0001134. [13] Huang NM, Chang JJ, Liang MT. Effect of plastering on the carbonation of a 35year-old reinforced concrete building. Constr Build Mater 2012;29:206–14. https://doi.org/10.1016/j.conbuildmat.2011.08.049. [14] Talakokula V, Bhalla S, Ball RJ, Bowen CR, Pesce GL, Kurchania R, et al. Diagnosis of carbonation induced corrosion initiation and progression in reinforced concrete structures using piezo-impedance transducers. Sensors Actuators, A Phys 2016; 242:79–91. https://doi.org/10.1016/j.sna.2016.02.033. [15] Han Shen X, Feng Liu Q, Hu Z, Qiang Jiang W, Lin X, Hou D, Hao P. Combine ingress of chloride and carbonation in marine-exposed concrete under unsaturated environment: A numerical study. Ocean Eng 2019;189:106350. https://doi.org/ 10.1016/j.oceaneng.2019.106350. [16] Medvedev V, Pustovgar A. A review of concrete carbonation and approaches to its research under irradiation. Buildings 2023;13. [17] Roy SK, Northwood DO, Pohs KB. Effect of plastering on the carbonation of a 19year-old reinforced concrete building. Constr Build Mater 2012;29:206–14. https://doi.org/10.1016/j.conbuildmat.2011.08.049. [18] Al Fuhaid AF, Niaz A. Carbonation and corrosion problems in reinforced concrete structures. Buildings 2022;12:1–20. https://doi.org/10.3390/buildings12050586. [19] Mi Z, Hu Y, Li Q, An Z. Effect of curing humidity on the fracture properties of concrete. Constr Build Mater 2018;169:403–13. https://doi.org/10.1016/j. conbuildmat.2018.03.025. [20] Zhao J, Cai G, Cui L, Si Larbi A, Tsavdaridis KD. Deterioration of basic properties of the materials in FRP-strengthening RC structures under ultraviolet exposure. Polymers (Basel) 2017;9. https://doi.org/10.3390/polym9090402. [21] Wang Y, Nanukuttan S, Bai Y, Basheer PAM. Influence of combined carbonation and chloride ingress regimes on rate of ingress and redistribution of chlorides in concretes. Constr Build Mater 2017;140:173–83. https://doi.org/10.1016/j. conbuildmat.2017.02.121. [22] Kuosa H, Ferreira RM, Holt E, Leivo M, Vesikari E. Effect of coupled deterioration by freeze-thaw, carbonation and chlorides on concrete service life. Cem Concr Compos 2014;47:32–40. https://doi.org/10.1016/j.cemconcomp.2013.10.008. [23] P. Fernandes, P. Silva, L. Correia, J. Sena-cruz, Durability of an epoxy adhesive and a CFRP laminate under different exposure conditions, in: Third Conf. Smart Monit. Assess. Rehabil. Civ. Struct. 2015, 2015: p. 9. [24] Cui E, Jiang S, Wang J, Zeng X. Bond behavior of CFRP-concrete bonding interface considering degradation of epoxy primer under wet-dry cycles. Constr Build Mater 2021;292:123286. https://doi.org/10.1016/j.conbuildmat.2021.123286. [25] Grammatikos SA, Jones RG, Evernden M, Correia JR. Thermal cycling effects on the durability of a pultruded GFRP material for off-shore civil engineering structures. Compos Struct 2016;153:297–310. https://doi.org/10.1016/j. compstruct.2016.05.085. [26] Dong Hu D, Xun Lyu J, Liu T, Dong Lang M, Zhao L. Solvation effect of CO2 on accelerating the curing reaction process of epoxy resin. Chem Eng Process - Process Intensif 2018;127:159–67. https://doi.org/10.1016/j.cep.2018.01.027. [27] Silva P, Fernandes P, Sena-Cruz J, Xavier J, Castro F, Soares D, et al. Effects of different environmental conditions on the mechanical characteristics of a structural epoxy. Compos Part B Eng 2016;88:55–63. https://doi.org/10.1016/j. compositesb.2015.10.036. [28] Liu S, Pan Y, Li H, Xian G. Durability of the bond between CFRP and concrete exposed to thermal cycles. Materials (Basel) 2019;12. https://doi.org/10.3390/ ma12030515. [29] Jiang F, Han X, Wang Y, Wang P, Zhao T, Zhang K. Effect of freeze-thaw cycles on tensile properties of CFRP, bond behavior of CFRP-concrete, and flexural performance of CFRP-strengthened concrete beams. Cold Reg Sci Technol 2022; 194:103461. https://doi.org/10.1016/j.coldregions.2021.103461. [30] Helbling C, Karbhari V. Durability of composites in aqueous environments. In: Durab. Compos. Civ. Struct. Appl. Amsterdam, The Netherlands: Elsevier; 2007. p. 31–71. [31] J. Sena-Cruz, P. Fernandes, P. Silva, J. Barros, M.R.F. Coelho, Bond behaviour of concrete elements strengthened with NSM CFRP laminate strips under wet-dry cycles, in: J.W. Cairns, G. Metelli, G. A.Plizzari (Eds.), New Mater. under Sev. Cond., 2012 Publisher creations, 2012. [32] Garz´ on-Roca J, Sena-Cruz JM, Fernandes P, Xavier J. Effect of wet-dry cycles on the bond behaviour of concrete elements strengthened with NSM CFRP laminate strips. Compos Struct 2015;132:331–40. https://doi.org/10.1016/j. compstruct.2015.05.053. [33] Al-Mahmoud F, Mechling JM, Shaban M. Bond strength of different strengthening systems - Concrete elements under freeze-thaw cycles and salt water immersion exposure. Constr Build Mater 2014;70:399–409. https://doi.org/10.1016/j. conbuildmat.2014.07.039. [34] A. Dushimimana, L. Correia, R. Cruz, S. Cabral-fonseca, J.M. Pereira, J. Sena-Cruz, Durability of CFRP-concrete bond and corresponding involved materials under different natural environmental exposures for a period of four years, in: FRPRCS-15 and APFIS-2022, Shenzhen, China, 2022: pp. 10–14. [35] A. Dushimimana, L. Correia, R. Cruz, J.M. Pereira, J. Sena-Cruz, Durability of CFRP-Concrete Bond in EBR and NSM Sysytems under Natural Ageing for a Period of Three Years, in: Proc. 20th Eur. Conf. Compos. Mater., CC BY-NC 4.0, Lausanne, Switzerland, 2022: pp. 1–8. 10.5075/epfl-298799_978-2-9701614-0-0. [36] Committee 440, Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening Concrete Structures, Farmington Hills, MI, USA, 2017. [37] CNR-DT 200 R1, Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening Existing Structures; CNR-DT 200 R1/2013;, Rome, Italy, 2013. [38] IPQ, IPQ (Instituto Portuguˆ es da Qualidade). 1992. NP EN EN 1992-1-1. Eurocode 2: design of concrete structures - Part 1-1: general rules and rules for buildings, Instituto Portuguˆ es da Qualidade (IPQ)., Caparica, Portugal, 2010. [39] S&P, S&P 220 Resin epoxy adhesive. Technical Data Sheet, Seewen, Switzerland, 2015. [40] S&P, S&P CFRP Laminates. Technical Datasheet, Seewen, Switzerland, 2014. [41] EN 1992-1-1 Eurocode 2, Design of concrete structures - Part 1-1 : General rules and rules for buildings, Brussels, Belgium:, 2004. EN 1992-1-1:2004 (E). [42] Cruz R, Correia L, Cabral-Fonseca S, Sena-Cruz J. Durability of bond of EBR CFRP laminates to concrete under real-time field exposure and laboratory accelerated ageing. Constr Build Mater 2023;377. https://doi.org/10.1016/j. conbuildmat.2023.131047. [43] IPQ, IPQ (Instituto Portuguˆ es da Qualidade). 2013. NP EN 12390-13. Testing Hardened Concrete. Part 13: Determination of Secant Modulus of Elasticity in Compression, Instituto Portuguˆ es da Qualidade (IPQ)., Caparica, Portugal, 2013. [44] IPQ, 2011. NP EN 12390-3. Testing Hardened Concrete. Part 3: Compressive Strength of Test Specimen, Instituto Portuguˆ es da Qualidade (IPQ), Caparica, Portugal, 2011. [45] BSI, BSI (British Standards Institution). Products and systems for the protection and repair ofconcrete structures – Test methods – Measurement ofbond strength by pull-off. EN 1542., London: BSI, 1999. [46] ISO, ISO 527-2:2012—Plastics—Determination of Tensile Properties—Part 2: Test conditions for Moulding and Extrusion Plastics, Gen` eve, Switzerland, 2012. [47] ISO, ISO 527-5:2009 Part 5: Test Conditions for Unidirectional Fibre-Reinforced Plastic composites. Plastic—Determ. Tensile Prop., Inte Gen` eve, Switzerland, 2009; Volume 1, 2009. [48] Hills TP, Gordon F, Florin NH, Fennell PS. Statistical analysis of the carbonation rate of concrete. Cem Concr Res 2015;72:98–107. https://doi.org/10.1016/j. cemconres.2015.02.007. [49] Bouzoubaˆ a N, Bilodeau A, Tamtsia B, Foo S. Carbonation of fly ash concrete: Laboratory and field data. Can J Civ Eng 2010;37:1535–49. https://doi.org/ 10.1139/L10-081. [50] Jiang L, Zhang Y. Analysis and calculation of the length of half – carbonated zone in concrete [in Chinese]. Ind Constr 1999;29:4–8. [51] Tseng H, Hsia J. Concrete carbonation and durability prediction[in Chinese], Hunan Commun. Sci Technol 2006;32:121–6. [52] Kabir MI, Shrestha R, Samali B. Effects of applied environmental conditions on the pull-out strengths of CFRP-concrete bond. Constr Build Mater 2016;114:817–30. https://doi.org/10.1016/j.conbuildmat.2016.03.195. [53] Dushimimana A, Niyonsenga AA, Nzamurambaho F. A review on strength development of high performance concrete. Constr Build Mater 2021;307:124865. https://doi.org/10.1016/j.conbuildmat.2021.124865. [54] Shoukry SN, William GW, Downie B, Riad MY. Effect of moisture and temperature on the mechanical properties of concrete. Constr Build Mater 2011;25:688–96. https://doi.org/10.1016/j.conbuildmat.2010.07.020. A. Dushimimana et al. Composite Structures 340 (2024) 118174 19 [55] Leone M, Matthys S, Aiello MA. Effect of elevated service temperature on bond between FRP EBR systems and concrete. Compos Part B Eng 2009;40:85–93. https://doi.org/10.1016/j.compositesb.2008.06.004. [56] Firmo JP, Correia JR, Pitta D, Tiago C, Arruda MRT. Bond behavior between nearsurface-mounted CFRP strips and concrete at high temperatures. J Compos Constr 2015;19:04014071. https://doi.org/10.1061/(asce)cc.1943-5614.0000535. [57] Cruz JS, Barros J. Bond between near-surface mounted carbon-fiberreinforced polymer laminate strips and concrete. J Compos Constr 2004;82:1513–21. https:// doi.org/10.1016/j.compstruc.2004.03.047. A. Dushimimana et al.