Flexural strengthening of pre-cracked RC slabs with prestressed NSM CFRP laminates and evaluation of strain loss

Abstract

The strengthening intervention of RC structures often involves already cracked concrete. To evaluate the effect of the level of damage prior to the strengthening (pre-cracks) on the behavior of the flexurally strengthened RC slabs with prestressed NSM CFRP laminates, an experimental research was carried out. Two pre-cracking levels of damage were analyzed and, for each one, three levels of prestress were tested (0%, 20% and 40%). The obtained results showed that the strengthening of damaged RC slabs with prestressed NSM CFRP laminates results in a significant increase on the load carrying capacity at serviceability limit states. Pre-cracked RC slabs strengthened with prestressed NSM CFRP laminates presented a load carrying capacity almost similar to the corresponding uncracked strengthened slabs. To determine the effective prestress level in CFRP laminates, the variation of strain over the length of the CFRP and over time was experimentally recorded. The prestress transfer length was also evaluated. The experimental results revealed that the transfer length of CFRP laminates was less than 150 mm, and the maximum value of strain loss out of transfer length (around 14%) was measured close to the cracked section of the damaged RC slabs. Significant part of strain loss in CFRP laminates occurred during 24 h after releasing the prestress load.

Keywords

experimental results flexural strengthening pre-cracked RC slabs prestressed NSM CFRP laminates strain loss

Introduction

The use of fiber reinforced polymers (FRP) to retrofit and rehabilitate reinforced concrete (RC) structures is becoming a competitive technology over traditional approaches. The advantages of using these advanced composite materials for retrofitting, compared to steel plate bonding or jacketing, are their high strength-to-weight ratio and high durability resistance under environmental conditions. Due to their relatively high elasticity modulus and tensile strength and their high resistance in fatigue loading conditions and in aggressive mediums, carbon fiber reinforced polymers (CFRP) with very small cross section can introduce notable increment of load carrying capacity in RC structures with minimum interference with the original configuration of these structures (Aslam et al., 2015). CFRP has been used for the flexural strengthening, either as an externally bonded reinforcement (EBR technique) applied to the soffit tension surface of the RC elements (Barros et al., 2007; Bilotta et al., 2015; Esfahani et al., 2007; Pham and Al-Mahaidi, 2004), or as a near surface mounted (NSM technique) by installing CFRP bars (circular, square or rectangular cross sections) into the pre-cut slits on the concrete cover of the elements (Barros et al., 2007; Barros and Fortes, 2005; Bilotta et al., 2015; Costa and Barros, 2010; De-Lorenzis and Nanni, 2002; El-Hacha and Riskalla, 2004; Sharaky et al., 2015). Based on experimental research, strengthening effectiveness with the NSM technique is higher than the EBR technique, mainly when rectangular cross section CFRP laminates are used in the NSM technique (Barros et al., 2007; Bilotta et al., 2015; El-Hacha and Riskalla, 2004). In fact, strengthening RC structural elements with NSM CFRP provided higher strength capacity than EBR CFRP using the same material with the same axial stiffness (Barros et al., 2007; El-Hacha and Riskalla, 2004). Moreover, CFRP applied using the NSM technique is less prone to premature debonding, thus its high tensile strength is better exploited.

Although flexural strengthening of RC beams (Barros et al., 2007; Barros and Fortes, 2005; Bilotta et al., 2015; Costa and Barros, 2010; De-Lorenzis and Nanni, 2002; El-Hacha and Riskalla, 2004; Sharaky et al., 2015) and slabs (Bonaldo et al., 2008; Dalfré and Barros, 2013) with NSM CFRP laminates without any prestress level (passive reinforcement) can significantly increase the ultimate load carrying capacity of these structures, it does not remarkably change their behavior under service loads. By applying a certain level of prestress in the CFRP laminates for the flexural strengthening of RC beams (Badawi and Soudki, 2009; Barros, 2009; El-Hacha and Gaafar, 2011; Hajihashemi et al., 2011; Hong and Park, 2016; Peng et al., 2014; Rezazadeh et al., 2014) and slabs (Hosseini et al., 2014), their high tensile capacity is more efficiently used, causing a significant increase in the load bearing capacity of the strengthened members under both service and ultimate conditions. The strengthening intervention often involves RC elements already cracked. Although the level of damage due to pre-cracking in RC structural elements should be accounted for, only a few number of studies have assessed the influence of this type of damage when NSM prestressed CFRP laminates are adopted (Kotynia and Przygocka, 2018).

In the present study, the effect of the level of damage by pre-cracking on the behavior of the RC slabs flexurally strengthened with prestressed NSM CFRP laminates was assessed by executing an experimental program composed of the following 10 RC slabs: one reference slab; three slabs with 0% prestress (passive CFRP laminates); two groups of three slabs strengthened with prestressed CFRP laminates, one group at 20% and the other at 40% of the tensile strength of the laminates. Two pre-cracking levels of damage were analyzed for each level of prestress, and their effects on the behavior of the RC slabs were evaluated based on the load-deflection relationship up to failure, the type of failure modes, the mobilized stress levels of the CFRP, as well as the cracking pattern.

Furthermore, limited studies have taken into account the strain loss in prestressed NSM CFRP laminates for designing strengthening interventions (Badawi et al., 2011; Rezazadeh and Barros, 2015). For obtaining an accurate estimation of prestress level in CFRP laminates, parameters of strain loss and transfer length were investigated in the present study by executing an experimental program consisted of five series of tests including damaged and non-damaged RC slabs flexurally strengthened with prestressed NSM CFRP laminates. The values of strain in the CFRP laminates were recorded over time using strain gauges to obtain strain loss. The variation of strain loss was also assessed along the length of the slabs.

Experimental program about the effect of pre-cracks

Slab prototypes

The experimental program was composed of 10 one-way RC slabs with a rectangular cross section of 600 × 120 mm²: one reference RC slab without CFRP and nine RC slabs flexurally strengthened with two NSM CFRP laminates with the cross section of 1.4 mm thickness and 20 mm depth. In three RC slabs the strengthening was passive while in the remaining six slabs a certain level of prestress was imposed in the NSM CFRP laminates. This prestress level was 20% or 40% of the ultimate tensile strength of the CFRP laminates (three slabs each). The slabs had a total length of 2600 mm and a span length of 2400 mm (Figure 1(a)). The adopted longitudinal steel reinforcement consisted of 4 bars of 8 mm diameter (4ϕ8) in the tension zone and 3 bars of 6 mm diameter (3ϕ6) in the compression zone. The concrete clear cover of the longitudinal tensile steel reinforcements was 31 mm. Steel stirrups of 6 mm diameter spaced at 300 mm (ϕ6@300 mm) were adopted as transverse reinforcement to maintain the longitudinal steel reinforcement in the targeted position. The average value of the concrete compressive strength (f_cm) was 39.5 MPa and the reinforcement systems were designed in order that all slabs fail in flexural.

Figure 1.

(a) General information of the RC slabs and (b) position of the strain gauges on the monitored longitudinal tensile bars (SG-S1 and SG-S2) and on the NSM CFRP laminates (SG-L1, SG-L2, and SG-L3), dimensions in mm.

Figure 1(a) displays the details of the cross section and reinforcement arrangement for the specimens, as well as the longitudinal geometry, loading configuration, and support conditions. The general information of the tested RC slabs is represented in Table 1. According to this table the group of strengthened slabs includes three slabs without pre-cracks (S2L-0, S2L-20, S2L-40 with 0%, 20%, and 40% prestress level, respectively), three slabs with the pre-cracking level I (S2L-0-PC1, S2L-20-PC1, S2L-40-PC1 with 0%, 20%, and 40% prestress level, respectively), and three slabs with the pre-cracking level II (S2L-0-PC2, S2L-20-PC2, S2L-40-PC2 with 0%, 20%, and 40% prestress level, respectively). SREF was the reference slab without CFRP. The tested slabs had a percentage of longitudinal tensile steel bars (ρ_sl) of about 0.39%, while the CFRP strengthening percentage (ρ_f) was 0.09% (Table 1).

Table 1.

General information of the tested RC slabs.

Slab	ρ_sl (%)^a	ρ_f (%)^b	Level of damage(pre-cracks)	Level of prestress (%)
SREF	0.39 (4ϕ8)	—	—	—
S2L-0		0.09 (2 CFRPlaminates with1.4 ×20 mm²of cross section)	0	0
S2L-20			0	20
S2L-40			0	40
S2L-0-PC1			I	0
S2L-20-PC1			I	20
S2L-40-PC1			I	40
S2L-0-PC2			II	0
S2L-20-PC2			II	20
S2L-40-PC2			II	40

The percentage of the existing flexural reinforcement was obtained from $ρ_{sl} = (A_{sl} / (b_{w} \times d_{s})) \times 100$ , where A_sl is the cross sectional area of the longitudinal tensile steel reinforcement (see Figure 1), b_w = 600 mm is the width of the slab’s cross section, and d_s is the distance from extreme compression fibre to the centroid of tensile reinforcement.

The CFRP percentage was obtained from $ρ_{f} = (A_{f} / (b_{w} \times d_{f})) \times 100$ , where $A_{f}$ is the cross sectional area of the NSM CFRP laminates and d_f is the distance from extreme compression fibre to the centroid of the NSM CFRP laminates.

The experimental program is composed of two types of four-point beam bending tests (Figure 1): pre-cracking test and failure test. In the pre-cracking tests, six RC slabs were loaded in order to develop a specific level of cracking in the concrete (pre-cracks) prior the application of the CFRP (three RC slabs with the level I of cracking and three RC slabs with the level II of cracking). In the failure tests, all RC slabs were loaded up to failure. The four point beam bending tests were carried out using a displacement control method with the rate of 0.02 mm/s at mid-span in the loading process. For pre-cracking test, after the loading phase until a certain level of deformation in the mid-span of the slabs, the unloading phase with the rate of 0.03 mm/s was imposed.

All slabs were instrumented to measure the applied load, deflections and strains in the CFRP laminates and in the longitudinal tensile steel reinforcement. The deflections of the specimens were measured at mid-span, under the applied loads, and at half of one of the shear spans, by using five Linear Variable Displacement Transducers (LVDTs). The values of strains on the tensile steel reinforcements were registered by two strain gauges (SG-S1 and SG-S2) (Figure 1(b)). The strains on the CFRP laminates of the strengthened slabs were also monitored and recorded using strain gauges (SG-L1 to SG-L3) installed on the longitudinal direction of the two CFRP laminates according to the distribution presented in Figure 1(b).

Materials properties

The compressive strength (EN 206-1, 2000) and the Young’s modulus (LNEC E397-1993, 1993) of the concrete were evaluated during the slab tests, by carrying out direct compression tests on seven cylinders of 150 mm diameter and 300 mm height. The main properties of the tensile steel reinforcements (with 6 and 8 mm diameter) used in the tested RC slabs were characterized by carrying out uniaxial tensile tests (five specimens of steel bars of both 6 and 8 mm of diameter were tested) according to the recommendations of EN 10002-1 (1990). The tensile properties of the CFK 150/2000 S&P laminates were also evaluated by performing uniaxial tensile tests in seven specimens according to ISO 527-5 (1997). The obtained results (average values and coefficient of variation) are shown in Table 2.

Table 2.

Average values of the properties of intervening materials.

Concrete	Compressive strength, f_cm (age of slab tests)		Young’s modulus, E_cm (age of slab tests)
	39.5 MPa (5.2)		32.6 GPa (5.2)
Steel	Diameter	Yield stress, f_sym		Tensile strength, f_sum
	ϕ6	527.6 MPa (3.5)		651.4 MPa (3.0)
	ϕ8	556.4 MPa (3.3)		679.9 MPa (1.5)
CFRP laminate	Tensile strength, f_fum	Elasticity modulus, E_fm		Maximum strain, ε_fu
	2733.5 MPa (4.5)	178.6 GPa (5.7)		15.4‰ (7.5)

The value in parentheses is the coefficient of variation.

The CFRP laminates was bonded to the concrete substrate by S&P Resin 220 epoxy adhesive. The short and long term tensile behavior of this adhesive was experimentally assessed by Costa and Barros (2015). On 3 day the elasticity modulus (E_{0.5‰–2.5‰}) attained a stabilized value and the obtained values of the tensile strength and the E_{0.5‰–2.5‰} were about 20 MPa and 7 GPa, respectively.

Pre-cracking tests

As mentioned above, prior to the application of the CFRP strengthening systems, in six RC slabs a certain cracking level in the concrete (level type I in the slabs S2L-0-PC1, S2L-20-PC1, and S2L-40-PC1; and level type II in the slabs S2L-0-PC2, S2L-20-PC2, and S2L-40-PC2) was imposed. The criterion for the level I of pre-cracking was to load the slabs until to obtain in the mid-span the maximum allowed deflection for the serviceability limit states according to the EN 1992-1-1 (2004) (u_Fserv = l/250 = 2400/250 = 9.6 mm, where l is the slab span length). Then, the slabs were unloaded. Taking into account the behavior of the SREF slab (this slab was previously tested up to failure), the criterion for the level II of pre-cracking was to load the slabs until to obtain, in the mid-span, the deflection corresponding to 80% of the yielding load of the SREF slab (that was around 15 mm according to Figure 2(a)) and then it was imposed the unloading phase. This deflection target at mid span for the second level of pre-cracking is in accordance to the recommendation of ACI 440.2R-08 (2008) about the maximum stress in steel reinforcement for the serviceability limit states that should be lower than 80% of the yielding strength of the steel reinforcement. The points in correspondence to the level I and II of pre-cracking tests are depicted in the load versus mid span deflection curve of SREF slab (Figure 2(a)). Figure 2(c) shows the application of pre-cracks in S2L-20-PC2 slab, as an example.

Figure 2.

(a) Definition of the criterion for the pre-cracking tests, (b) crack of unknown reason in S2L-0-PC2 slab before pre-cracking, (c) pre-cracking test of S2L-20-PC2 slab, (d) load versus mid-span deflection of the pre-cracking tests (level I), (e) load versus mid-span deflection of the pre-cracking tests (level II), (f) crack patterns of the pre-cracked RC slabs (level I), and (g) crack patterns of the pre-cracked RC slabs (level II).

Figure 2(d) and (e) show the applied load versus mid-span deflection curves for the RC slabs that were pre-cracked according to the level I and II, respectively. After cracking the concrete, some unloading-reloading cycles in the load-deflection curves of the slabs were observed due to the formation of the new cracks. There was a crack of unknown reason in S2L-0-PC2 slab before application of the pre-cracking level II. This crack was in the mid-span section, as shown in Figure 2(b) and caused the lower stiffness of S2L-0-PC2 slab compared to the other slabs during the application of the pre-cracking load. Table 3 displays the main results of the pre-cracking tests in terms of cracking force (F_cr), maximum force (F_max-precrack), mid-span deflection at F_max-precrack (u_{Fmax-precrack}), and mid-span deflection after the unloading phase (u_{final-precrack}).

Table 3.

Summary of the results in terms of loads and deflections of the pre-cracking tests.

Level of damage	First level of pre-cracking (I)			Second level of pre-cracking (II)
Slab	S2L-0-PC1	S2L-20-PC1	S2L-40-PC1	S2L-0-PC2	S2L-20-PC2	S2L-40-PC2
F_cr (kN)	10.0	11.0	13.0	–^a	8.9	9.5
F_max-precrack (kN)	14.6	15.8	17.1	16.4	16.8	17.1
u_{Fmax-precrack} (mm)	9.7	9.6	9.3	16.3	14.9	15.0
u_{final-precrack} (mm)	4.8	4.9	4.3	7.8	6.5	6.8

Before the pre-cracking test, this slab had, due to an unknown cause, a crack at mid span visible in all faces of the slab (see Figure 2(b)).

Figure 2(f) and (g) show the crack pattern of the pre-cracked RC slabs for the level I and II, respectively. When the crack pattern of the pre-cracked RC slabs is compared, it seems that the length of the slab cracked band increases as the level of damage increases. The number of cracks in the slabs after the second level of damage (between 6 and 9) was higher compared with that at the end of the first level of damage (between 4 and 6). At the end of the first level of damage, cracks were restricted to the pure bending zone (between the loaded sections), while at the end of the second level of damage, cracking was extended to the shear span of the slabs, and the width of the previous ones in the pure bending zone has increased.

Flexural strengthening procedures

In order to apply the CFRP laminates (passive or prestressed) using the NSM technique, a diamond cutter was used to open slits of about 5 mm wide and 25 mm deep on the concrete cover of the tension face considering the arrangement of the laminates shown in Figure 1. Then, for the non-prestressed slabs, the procedure for the application of the strengthening was as following: (i) the slits were cleaned by compressed air and the laminates (supplied in rolls of 150 m, with a cross-section of 1.4 × 20 mm²) were cut with a length of 2300 mm and then cleaned with acetone; (ii) the epoxy adhesive was prepared according to the supplier recommendations and used to fill the slits; (iii) an adhesive layer was applied on the faces of the laminates, and the laminates were inserted into the slits, iv) the adhesive in excess was removed.

S2L-20, S2L-40, S2L-20-PC1, S2L-40-PC1, S2L-20-PC2, and S2L-40-PC2 slabs were strengthened with two prestressed NSM CFRP laminates. Figure 3 illustrates the application of this strengthening technique step by step. After installing the slab in the right position of the prestressing line, the CFRP laminates were placed into the slits and passed through the hydraulic jacks and load cells, and were anchored in both extremities by using an active and a passive anchor. Each laminate was installed in the middle of the slit, as close as possible to the external surface of the slab. Then, the prestressing load was applied to one extremity of the laminate by the hydraulic jack (active anchor), at a load rate of 0.5 kN/min, while the other extremity of the CFRP laminate remained fixed to the main frame of the prestressing line by using a steel anchor (passive anchor). When the prestressing load was completely applied, an epoxy adhesive was applied into the slits using a spatula. Special care was taken in order to avoid the formation of voids in the concrete-adhesive-CFRP interfaces, as well as in the adhesive layer. About 6 days after the application of the adhesive, the prestressing load was released slowly and simultaneously in both CFRP laminates at a load rate of about 0.3 kN/min.

Figure 3.

Flexural strengthening steps of a pre-cracked RC slab by applying prestressed NSM CFRP laminates (slab S2L-40-PC2).

Test up to failure

Behavior of RC slabs with pre-cracking

Figure 4 shows the load versus mid span deflection curves of the damaged RC slabs before and after strengthening. The behavior of the damaged RC slabs flexurally strengthened with NSM CFRP laminates with or without prestress has three important phases: up to the beginning of the pre-cracks propagation; between the beginning of the pre-cracks propagation and yield initiation of the steel reinforcement; and between steel yield initiation and ultimate load. The maximum load was obtained immediately before the CFRP laminates rupture, after which the load dropped to the reference slab capacity (SREF slab). In the phase before CFRP laminates rupture, whose increase of slab’s load carrying capacity is governed by the contribution of the CFRP laminates, the strengthened slabs’ curve had almost linear behavior (due to the linear behavior of CFRP laminates, while steel reinforcement was yielded with concrete in its cracking stabilized phase).

Figure 4.

Load versus deflection at mid-span of the tested RC slabs with: (a) first level of damage and (b) second level of damage.

In the first part of the monotonic test until failure, the reopening of the pre-cracks was verified. During the test, these cracks started to develop in the direction of the compression face of the slab. With the increase of the load, new cracks were also progressively formed in the shear spans toward the supports, enlarging the slab length in the cracked stage. This process of the development of the pre-cracks and the formation of new cracks was started for load levels above F_cr-damage (see Table 4 and Figure 4).

Table 4.

Prestress level versus pre-cracking level.

Level of damage	First level of pre-cracking (I)			Second level of pre-cracking (II)
Slab	S2L-0-PC1	S2L-20-PC1	S2L-40-PC1	S2L-0-PC2	S2L-20-PC2	S2L-40-PC2
F_cr-damage (kN)	17.9	24.6	29.7	19.2	28.3	35.4
(F_cr-damage−F_max-precrack)/F_max-precrack) × 100 (%)	22.6	55.7	72.7	45.9	68.5	107.0
u_{Fmax-precrack} − u_{final-precrack} (mm)	4.9	4.7	5.0	8.5	8.4	8.2
u_Fcr-damage (mm)	6.5	6.9	6.4	11.7	11.5	12.0

The results of F_max-precrack, u_{Fmax-precrack} and u_{final-precrack} are indicated in Table 3.

The values of F_cr-damage for the slabs with the first level of pre-cracking, S2L-0-PC1, S2L-20-PC1, and S2L-40-PC1, were 17.9, 24.6, and 29.7 kN, respectively. The values of F_cr-damage for the slabs with the second level of pre-cracking, S2L-0-PC2, S2L-20-PC2, and S2L-40-PC2, were 19.2, 28.3, and 35.4 kN, respectively. Considering the values of F_cr-damage and the values of F_max-precrack, it is possible to conclude that the difference in these forces for each slab increased with the prestress level, which confirms the benefit of the level of applied prestress on the strengthening effectiveness of damage RC slabs. Regarding the level of prestress, the differences between F_cr-damage and F_max-precrack were (apart from S2L-0 slab) higher for the second level of damage (slabs with more pre-cracks and consequently with more damage before the strengthening).

The difference between the maximum deflection of the RC slabs during the application of pre-cracking and final value of deflection after unloading of the slabs (u_{Fmax-precrack} − u_{final-precrack}), as well as the values of u_Fcr-damage are indicated in Table 4. The values of u_{Fmax-precrack} − u_{final-precrack} and u_Fcr-damage for each RC slab are almost the same. The difference between these values is justified by the slabs being turned over before the strengthening application. This operation closed the pre-cracks and decreased the deflection, therefore, during the monotonic test until failure of the strengthened slabs, larger deflection was necessary to reopen the pre-cracks.

Comparison of the general behavior of strengthened RC slabs with prestressed CFRP laminates with or without pre-cracks

Figure 5(a) indicates the behavior of the slabs with different levels of prestress (0%, 20%, and 40%) for each level of damage. Figure 5(a-i)–(a-iii) correspond to the slabs without damage, with the level of damage I, and with the level of damage II, respectively. Table 5 shows the summary of the results of the tested RC slabs in terms of service (F_serv), yielding (F_sy), and maximum (F_max) load. The values of the deflection at mid-span for the maximum load F_max (u_Fmax) are also indicated in the Table 5. The service load (F_serv) is the load corresponding to the maximum allowed deflection for serviceability limit states (u_Fserv). The yielding load is assumed the load at which a considerable decay of stiffness has occurred in the post cracking stage of a tested slab.

Figure 5.

Load versus deflection at mid-span of the tested RC slabs: (a) effect of the level of prestress in RC slabs without or with pre-cracks and (b) effect of the level of damage (pre-cracks).

Table 5.

Summary of the results in terms of loads and deflections.

Slab	Level ofdamage	Service		Yielding		Maximum
		F_serv(kN)	$Δ F_{serv} / F_{serv}^{Ref}$ (%)	F_sy (kN)	$Δ F_{sy} / F_{sy}^{Ref}$ (%)	F_max(kN)	$Δ F_{\max} / F_{\max}^{Ref}$ (%)	u_Fmax(mm)	$Δ u_{F_{\max}} / u_{F_{\max}}^{Ref}$ (%)
SREF	0	15.2	–	24.7	–	28.9	–	81.1	–
S2L-0	0	18.5	21.7	34.8	40.9	60.7	110.0	89. 8	10.7
S2L-20	0	26.8	76.3	46.9	89.9	66.1	128.7	65.4	−19.4
S2L-40	0	33.7	121.7	52.8	113.8	66.0	128.4	50.3	−38.0
S2L-0-PC1	I	21.6	42.1	36.5	47.8	66.0	128.4	87.1	7.4
S2L-20-PC1	I	27.6	81.6	41.0	66.0	61.0	111.1	65.9	−18.7
S2L-40-PC1	I	35.3	132.2	49.5	100.4	63.5	119.7	46.8	−42.3
S2L-0-PC2^a	II	12.5	−17.8	35.0	41.7	58.7	103.1	78.1	−3.7
S2L-20-PC2	II	25.3	66.4	43.5	76.1	64.2	122.1	69.2	−14.7
S2L-40-PC2	II	31.2	105.3	50.0	102.4	61.9	114.2	49.0	−39.6

Slab with the unknown crack (see Figure 2(b)).

The values of $F_{serv}^{Str}$ , $F_{sy}^{Str}$ , and $F_{\max}^{Str}$ , and deflection relative to $F_{\max}^{Str}$ ( $u_{F_{\max}}^{Str}$ ) of the prestressed strengthened slabs are compared in Table 5 with the corresponding values of the reference slab ( $F_{serv}^{Ref}$ , $F_{sy}^{Ref}$ , $F_{\max}^{Ref}$ , and $u_{F_{\max}}^{Ref}$ ). The parameters $Δ F_{serv} / F_{serv}^{Ref}$ , $Δ F_{sy} / F_{sy}^{Ref}$ , $Δ F_{\max} / F_{\max}^{Ref}$ , and $Δ u_{F_{\max}} / u_{F_{\max}}^{Ref}$ were evaluated and included in Table 5, where $Δ F_{serv} = F_{serv}^{Str} - F_{serv}^{Ref}$ , $Δ F_{sy} = F_{sy}^{Str} - F_{sy}^{Ref}$ , $Δ F_{\max} = F_{\max}^{Str} - F_{\max}^{Ref}$ , and $Δ u_{F_{\max}} = u_{F_{\max}}^{Str} - u_{F_{\max}}^{Ref}$ .

As shown in Figure 5(a-i), the load versus deflection curve of the non-damaged slabs has three important phases: up to concrete cracking initiation; between concrete cracking and yield initiation of the steel reinforcement; and between steel yield initiation and ultimate load. Considering the general behavior of the slabs with pre-cracks discussed in the previous section, the differences between the strengthened slabs with and without pre-cracks were verified in terms of stiffness for loads lower than the value of F_cr-damage (see Table 4). As expected, until this level of loading, the stiffness of the slabs without damage was higher than the stiffness of the slabs with damage regardless of the level of prestress (Figure 5(b)).

According to Figure 5 and the results shown in Table 5, the application of prestressed NSM CFRP laminates for the flexural strengthening of pre-cracked and non-pre-cracked RC slabs increased the load-carrying capacity of slabs in terms of serviceability limit state conditions. In fact, the average value of F_serv of prestressed slabs with pre-cracks (S2L-20-PC1, S2L-20-PC2, S2L-40-PC1, and S2L-40-PC2) and the average value of prestressed slabs without pre-cracks (S2L-20 and S2L-40) were, respectively, 1.96 and 1.99 times the value of F_serv of the SREF reference slab. These values indicate that by strengthening the already cracked slabs with prestressed CFRP laminates, the load carrying capacity of the corresponding uncracked prestressed strengthened slabs was recuperated, which means that the load carrying capacity at serviceability limit state conditions is not affected by the cracking damage adopted as long as the slabs are strengthened with prestressed laminates.

The service load (F_serv) was also increased with the level of prestress: the service load of the slabs with 20% of prestress S2L-20, S2L-20-PC1, and S2L-20-PC2 is, respectively, 1.76, 1.82, and 1.66 times the value of F_serv of the SREF reference slab; the service load of the slabs with 40% of prestress S2L-40, S2L-40-PC1, and S2L-40-PC2 is, respectively, 2.22, 2.32, and 2.05 times the value of F_serv of the SREF reference slab (Figure 6(a)). In terms of the increase in the service load, the level of prestress had a higher beneficial effect on the prestressed slabs with the level I of pre-cracking than in those with level II of pre-cracking.

Figure 6.

Influence of the level of damage in the effectiveness of the prestressed NSM CFRP laminates in terms of: (a) service load, (b) yielding load, (c) maximum load, and (d) maximum deflection.

Strengthening the damaged RC slabs with NSM CFRP laminates resulted also in higher yielding loads than the F_sy of the reference slab (SREF). As observed for F_serv, F_sy was also increased with the prestress level (Figure 6(b)). The values of F_sy of non-prestressed, 20% and 40% prestressed slabs with first level of damage were, respectively, 36.5, 41.0, and 49.5 kN, while for RC slabs with the second level of damage were, respectively, 35, 43.5, and 50 kN. These values were almost similar to the values of F_sy that were obtained in the strengthened slabs without pre-cracking (34.8, 46.9, and 52.8 kN for non-prestressed, 20% and 40% prestressed slabs, respectively).

The values of maximum load of all strengthened RC slabs after strengthening varied between 58.7 and 66.1 kN, which is 2.0–2.3 times higher than the maximum load of the reference slab (28.9 kN). The average value of the maximum load for the prestressed slabs without pre-cracks, with the level I of pre-cracking and with the level II of pre-cracking was, respectively, 66.1, 62.3, and 63.1 kN. Considering these results, for the adopted level of pre-cracking, the pre-cracked RC slabs strengthened with prestressed NSM CFRP laminates have presented a load carrying capacity almost similar to that of the homologous uncracked strengthened slabs. As shown in Table 5 and Figure 6(c), by applying 20% of prestress in RC slabs with the level of damage of 0, I, and II, the maximum load increased, respectively, 129%, 111%, and 122%, while 40% of prestress has guaranteed an increase of 128%, 120%, and 114%, respectively.

Furthermore, it was verified that the existence of pre-cracks did not change the failure mode of the strengthened slabs. In fact, regardless of the level of damage applied to the RC slabs, all the strengthened slabs failed by rupture of the CFRP laminates after yielding of the tension steel reinforcement. This is in agreement with the maximum values of strain recorded in the CFRP laminates’ strain gauges at the maximum load (F_max) of the slabs as indicated in Table 6 as “Total.” These values were obtained with the addition of the strain registered at the end of the prestressing phase ( “Prestressing” values in Table 6) with the maximum strain recorded in the loading phase of the slab up to its F_max ( “Test” values in Table 6). In fact, the maximum values of strain measured in the CFRP, namely in SG-L1 or SG-L2 strain gauges, of the prestressed RC slabs with or without pre-cracks are quite close to the ultimate tensile strain of the CFRP laminates, justifying the failure mode of these RC slabs and the high effectiveness of the NSM technique for the flexural strengthening of the RC slabs.

Table 6.

Maximum strain values recorded in CFRP laminates’ strain gauges up to the maximum load of the slabs.

Slab	SG-L1 (‰)			SG-L2 (‰)			SG-L3 (‰)
	Prestressing	Test	Total	Prestressing	Test	Total	Prestressing	Test	Total
S2L-0	0.0	15.1	15.1	0.0	14.9	14.9	0.0	5.7	5.7
S2L-20	3.0	11.1	14.1	3.0	12.2	15.2	3.0	4.8	7.8
S2L-40	6.0	9.6	15.5	5.9	9.0	14.9	6.1	2.1	8.2
S2L-0-PC1	0.0	15.4	15.4	0.0	14.9	14.9	0.0	5.3	5.3
S2L-20-PC1	3.1	12.1	15.1	3.3	10.7	14.0	3.3	2.5	5.8
S2L-40-PC1	5.5	9.5	15.1	5.3	8.9	14.2	6.3	2.6	8.9
S2L-0-PC2	0.0	13.7	13.7	0.0	11.9	11.9	0.0	3.9	3.9
S2L-20-PC2	2.8	11.9	14.7	3.0	11.0	14.0	3.1	4.3	7.4
S2L-40-PC2	5.5	9.6	15.0	5.8	8.6	14.4	6.1	1.9	8.0

As shown in Figure 5(a), regardless of the level of damage, the higher the CFRP prestressed level the larger the performance of the NSM technique in the improvement of the behavior of the slabs at serviceability limit conditions. However, the deflection at the maximum load of the slabs decreased with the increase in the prestress level: between 14.7% and 19.4% in the slabs of 20% of prestress and between 38% and 42.3% in the slabs of 40% of prestress compared to the reference slab. The decrease in the ultimate deflection of prestressed slabs with the level of damage of 0 (without pre-cracking), I and II by applying 20% of prestress was, respectively, 27%, 24%, and 11%, while 40% of prestress provided a decrease of 44%, 46%, and 37% compared to the corresponding values of the 0% prestress strengthened slabs (Figure 6(d)).

The crack pattern of the tested slabs at their failure is presented in Figure 7. The values of the average distance between cracks (d_cr), measured in the tension face of the RC slabs, were 145, 89, 88, and 83 mm for the slabs SREF, S2L-0, S2L-20, and S2L-40, respectively. The values of d_cr for the slabs S2L-0-PC1, S2L-20-PC1, and S2L-40-PC1 were, respectively, 100, 93, and 93 mm, and for the slabs S2L-0-PC2, S2L-20-PC2 and S2L-40-PC2 were, respectively, 101, 92, and 100 mm. The comparison of the final crack pattern of the RC slabs for each level of damage indicates that by increasing the prestress level, the length of the slab’s cracked band is decreased. This can be justified due to the initial compressive strain introduced by the prestress. Furthermore, the values of d_cr for pre-cracked slabs were higher than the values of d_cr for the homologous uncracked strengthened slabs. This tendency was also verified for the values of the slab’s cracked band (l_cr,band). The values of l_cr,band were 1452, 1779, 1579, and 1333 mm for the slabs SREF, S2L-0, S2L-20, and S2L-40, respectively, and 1800, 1540, and 1408 mm for the slabs S2L-0-PC1, S2L-20-PC1, and S2L-40-PC1, respectively, and 1818, 1659, and 1403 mm for the slabs S2L-0-PC2, S2L-20-PC2, and S2L-40-PC2, respectively.

Figure 7.

Crack pattern of the tested slabs.

Experimental program for the evaluation of strain loss

Slab prototypes

The experimental program comprised 16 RC slabs, flexurally strengthened with two prestressed NSM CFRP laminates with the cross section of 1.4 mm thickness and 20 mm depth, which were divided in five series (series A, B, and E, each one with four slabs; series C and D, each one with two slabs). All of the RC slabs had a rectangular cross section of 600 mm × 120 mm, 2600 mm total length, 2400 mm between supports and a shear span of 900 mm (Figure 1). The specimens of the series A, B, D, and E had a percentage of tensile flexural reinforcement (ρ_sl) of 0.39% (4ϕ8 in the tension zone—see Figure 1), while the specimens of the series C had a ρ_sl = 0.62% (4ϕ10 in the tension zone). For all of the series of slabs the percentage of CFRP (ρ_f) was 0.09%. The average value of the concrete compressive strength (f_cm) for series A, B, C, and E was 39.5 MPa, and for series D was 15 MPa. The prestress load that was applied in the strengthened slabs was the following percentages of the ultimate tensile strength of the CFRP laminates: 20%, 30%, 40%, and 50% prestress level in series A; 20% and 40% prestress level in series B, C, D, and E. In the case of series E, prior to the application of the CFRP strengthening systems, the RC slabs were pre-cracked with a certain level of cracking in the concrete, while no pre-cracking was imposed in the slabs of series A, B, C, and D before the application of the prestressed NSM CFRP laminates.

Series A was used to evaluate the effect of the prestress level in NSM CFRP laminates (20%, 30%, 40%, and 50%) on the CFRP strain losses during the first week after releasing the prestressing load in RC slabs of moderately high f_cm (39.5 MPa) and relatively low ρ_sl (0.39%), which is a little bit higher than the minimum ρ_sl according to the major part of design codes for RC slabs. Two of the prestress levels analyzed in series A (20% and 40%) were considered in the remaining series. In the slabs of series B (with the same ρ_sl and f_cm of the slabs of series A) the period of acquisition of CFRP strain variation was extended until around 1 month. In the slabs of series C (ρ_sl = 0.62% and f_cm = 39.5 MPa) and D (ρ_sl = 0.39% and f_cm = 15 MPa), the time of acquisition of CFRP stain variation was 7 days, as in the slabs of series A (the difference between the slabs of series A and C was the value of ρ_sl; the difference between the slabs of series A and D was the value of f_cm). In the slabs of series E (pre-cracking slabs with the same ρ_sl and f_cm of the slabs of series A and B), the data acquisition of the CFRP’s strain variation was used between 14 and 24 days.

Values of strains in the CFRP laminates were surveyed during the time period mentioned above (see also Table 7), which corresponds to the period between the release of the prestress load and the test until failure of the slabs. The reduction of strain values in terms of the percentage of initial prestress strain was quantified for all prestressed strengthened slabs and was considered as strain loss.

Table 7.

General information about experimental program of strain loss.

Series	Slab	Level of prestress (%)	Time recording ofCFRP strain variation(days)	ρ_sl (%)	ρ_f (%)	Compressivestrength,f_cm (MPa)	Damage(pre-cracking)
A	S2L-20-A	20	7	0.39 (4ϕ8)	0.09	39.5	No
	S2L-30-A	30	7
	S2L-40-A	40	7
	S2L-50-A	50	4
B	S2L-20-B1	20	28	0.39 (4ϕ8)	0.09	39.5	No
	S2L-40-B1	40	31
	S2L-20-B2	20	31
	S2L-40-B2	40	19
C	S2L-20-C	20	7	0.62 (4ϕ10)	0.09	39.5	No
	S2L-40-C	40	7
D	S2L-20-D	20	7	0.39 (4ϕ8)	0.09	15.0	No
	S2L-40-D	40	7
E	S2L-20-PC1	20	24	0.39 (4ϕ8)	0.09	39.5	Yes
	S2L-40-PC1	40	23
	S2L-20-PC2	20	18
	S2L-40-PC2	40	14

The arrangement of the strain gauges (SG-L1 to SG-L6) applied on the two prestressed NSM CFRP laminates is indicated in Figure 8. In order to obtain the strain loss in different parts of the laminate, the location of SG-L4 was changed in different series of the RC slabs in terms of the distance X (distance from SG-L5 to SG-L4 in the direction of longitudinal axis of slab), as indicated in Figure 8. This distance for series A was equal to zero (in this series, SG-L4 was not adopted), for series B and E was equal to 50 mm and for series C and D was equal to 100 mm. The distance from the left extremities of the CFRP laminates to SG-L1, SG-L2, SG-L3, SG-L5, and SG-L6 in the direction of longitudinal axis of slab was 1150, 850, 425, 100, and 25 mm, respectively. During the application of prestress load until the slab failure test, all strain gauges applied on the two prestressed NSM CFRP laminates were continuously monitored.

Figure 8.

Position of the six strain gauges installed on the two prestressed CFRP laminates (SG-L1 to SG-L6).

Strain loss over time

Table 8 shows the strain loss at the end of 1 day, 1 week, and immediately before the slab failure test (the final time of recording strains) after releasing the prestress load in the CFRP laminates. The strain loss was calculated using the following equation:

Strain loss = {(ε_{p} - ε_{t})}^{*} 100 / ε_{p}

(1)

where, $ε_{p}$ is the CFRP strain recorded before releasing the prestress load and $ε_{t}$ is the CFRP strain measured in the target periods, as indicated in Table 7.

Table 8.

Values of strain loss in the prestressed slabs.

Slab	Days afterreleasing	Strain loss (%)
		SG-L1	SG-L2	SG-L3	SG-L4	SG-L5	SG-L6
S2L-20-A	1	0.6	1.0	1.0	–	5.3	76.7^a
	7	0.6	1.0	1.0	–	6.7	79.2^a
S2L-30-A	1	0.8	0.5	1.0	–	4.4	51.2
	7	1.0	0.8	1.1	–	5.8	53.5
S2L-40-A	1	1.2	1.4	1.1	–	5.7	58.2
	7	1.8	2.3	1.6	–	8.9	62.7
S2L-50-A	1	1.5	1.5	1.7	–	7.9	65.7
	4	1.9	1.9	1.9	–	10.4	68.6
S2L-20-B1	1	2.0	1.4	1.4	1.8	2.6	36.1
	7	3.3	1.9	2.3	3.5	3.6	41.8
	28	6.2	5.4	3.4	6.1	6.1	44.9
S2L-40-B1	1	1.2	1.1	1.3	1.5	3.0	41.2
	7	1.8	1.5	2.0	2.2	3.9	44.8
	31	3.7	3.3	3.8	4.4	6.6	47.2
S2L-20-B2	1	0.6	0.6	0.7	0.6	1.4	30.6
	7	3.3	3.5	3.5	3.6	4.3	37.7
	31	7.9	7.7	7.8	8.1	7.7	44.7
S2L-40-B2	1	1.0	0.9	0.8	1.1	2.8	46.4
	7	1.8	1.4	1.4	1.5	4.1	50.9
	19	2.0	1.4	1.3	1.6	4.4	52.2
S2L-20-C	1	0.4	0.4	0.8	1.0	2.5^c	35.2^b
	7	0.7	0.6	1.1	1.6	4.0^c	42.3^b
S2L-40-C	1	0.0	0.0	0.0	0.2	3.5	46.0
	7	0.0	0.0	0.0	0.2	4.0	52.8
S2L-20-D	1	2.2	0.9	0.9	1.1	3.2	38.7
	7	7.4	5.0	5.0	5.9	8.8	48.2
S2L-40-D	1	1.8	1.8	1.7	2.2	10.0	47.1
	7	2.5	2.2	2.2	2.9	12.7	50.1
S2L-20-PC1	1	3.6	0.0	0.0	0.0	0.5	23.2
	7	3.7	0.0	0.0	0.0	0.5	26.8
	24	5.0	0.0	0.0	0.0	0.5	27.3
S2L-40-PC1	1	10.2	13.7	0.6	0.8	2.5	45.4
	7	10.2	13.8	0.6	0.8	3.0	50.5
	23	10.6	14.0	0.6	0.8	3.9	53.0
S2L-20-PC2	1	10.5	2.7	0.6	1.2	2.1	26.6
	7	11.2	2.7	0.6	1.2	2.1	32.2
	18	12.7	4.4	1.6	2.2	3.1	35.4
S2L-40-PC2	1	7.1	3.4	0.0	0.0	10.5	48.9
	7	7.1	3.4	0.0	0.0	10.5	51.9
	14	10.1	5.9	1.6	1.6	14.2	57.4

This strain gauge was at a distance of 20 mm from free extremity of the CFRP laminate.

This strain gauge was at a distance of 31 mm from free extremity of the CFRP laminate.

This strain gauge was at a distance of 106 mm from free extremity of the CFRP laminate.

Figure 9 shows the strain variations of the CFRP laminates during recording of strains after releasing the prestress loads on the prestressed RC slabs of the series A, C, and D (1 week with the exception of the slab S2L-50—see Table 7). As Table 8 and Figure 9 show, the highest strain losses occurred in the closest strain gauge to the free extremity of the laminate (SG-L6): 79.2%, 53.5%, 62.7%, and 68.6%, respectively, in the slabs with 20%, 30%, 40%, and 50% of the series A; 42.3% and 52.8% respectively, in the slabs with 20% and 40% of the series C; 48.2% and 50.1% respectively, in the slabs with 20% and 40% of the series D. The difference in the strain loss in SG-L6 strain gauge between 7 and 1 days after releasing the prestress was not more than 4.5%, 7.1%, and 9.5%, respectively in series A, C, and D. The rest of the strain gauges showed a strain loss less than 10.4% (series A), 4% (series C), and 12.7% (series D) and the difference in the strain loss in these strain gauges, between 7 and 1 days after releasing the prestress was no more than 1.4%, 1.5%, and 5.6%, respectively, in series A, C, and D. These values indicate that the maximum strain loss (in all strain gauges except SG-L6) and maximum difference in the strain loss (in all strain gauges) between 7 and 1 days after releasing the prestress occurred in series D (RC slabs with low strength concrete).

Figure 9.

Strains in the CFRP versus time after releasing the prestress load in the prestressed slabs of the series A, C and D during 1 week after the releasing.

Figure 10 shows the strain variations of the CFRP laminates during recording of strains after releasing the prestress loads on the RC slabs of the series B and E (in this series, the slabs were pre-cracked before the strengthening). The results shown in Table 8 and Figure 10 indicate that the maximum strain loss in both series was in SG-L6 and varied from 44.7% to 52.2% in series B, and 27.3% to 57.4% in series E. The maximum strain loss in the rest of the strain gauges was less than 8.1% in series B and 14.2% in series E, which indicated that the main part of the prestress load was transferred to the concrete. Furthermore, the difference in the strain loss between 7 and 1 days after releasing the prestress in the CFRP laminates was no more than 7.1% and 5.6%, respectively, in series B and E. The difference of the strain loss between 7 and 27 days after releasing the prestress load in the CFRP laminates of series B was no more than 7%, while the difference in the strain loss between 7 and 20 days after releasing the prestress in the CFRP laminates of series E was no more than 5.5%. Therefore, significant part of the strain loss in CFRP laminates occurred during 24 h after releasing the prestress.

Figure 10.

Strains in the CFRP versus time after releasing the prestress load in the prestressed slabs of the series B(non pre-cracked slabs) and E (pre-cracked slabs).

Strain loss along the length of the slab

Figure 11 shows the strain loss in the CFRP laminates along the length of the slabs for non-damaged (series A, B, C, and D) and damaged prestressed RC slabs (series E). As shown in Figure 11(a)–(d), in the non-damaged prestressed slabs, after a specific length (transfer length) from the free extremity of the CFRP laminates up to the middle of the slabs, the strain loss was almost constant along the length of the CFRP laminate. The transfer length of CFRP laminates was not higher than 150 mm.

Figure 11.

Strains loss in the CFRP laminate along the length of the slabs: (a) series A, (b) series B, (c) series C, (d) series D, and (e) series E.

Figure 12 indicates the relationship between strain loss and distance from the free extremity of CFRP bonded length in RC slabs. This figure was generated based on the nonlinear regression of the results of the all prestressed slabs (damaged and non-damaged) for each prestress level. The results indicate that the strain loss is increased with the prestress level (Figure 12 and Table 8).

Figure 12.

Loss of strains versus distance from the end of CFRP bonded length.

In the damaged RC slabs (series E—Figure 11(e)), the SG-L3, SG-L4, SG-L5, and SG-L6 strain gauges values were almost similar to the values in the non-damaged RC slabs (Figure 11(a)–(d)). However, SG-L1 and SG-L2 strain gauges in some RC slabs of series E showed higher strain loss values. In non-damaged RC slabs, the maximum value of the strain loss for SG-L1 and SG-L2 was, respectively, 7.9% and 7.7%, while in damaged RC slabs the corresponding values were 12.7% (SG-L1) and 14.0% (SG-L2).

In fact, the strain loss recorded by SG-L1 in S2L-20-PC2, S2L-40-PC1, and S2L-40-PC2 slabs and by SG-L2 in S2L-40-PC1 slab was more than 10%, while the strain loss recorded in the remaining SG-L1 (in slab S2L-20-PC1) and SG-L2 (in slabs S2L-20-PC1, S2L-20-PC2, and S2L-40-PC2) of series E was less than 6%. Figure 13 indicates that the strain gauges with a recorded strain loss higher than 10% were positioned closer to the pre-cracks, which demonstrates a tendency for an increase in strain loss in the sections of the laminate near the existing cracks (pre-cracks). In order to clarify this tendency, in the slab S2L-40-PC2 a new strain gauge (SG-L0) was installed slightly far from the pre-cracks. Namely, SG-L0 was positioned in the same laminate of SG-L2, but at a distance of 230 mm toward the mid-section, as indicated in Figure 13(d). The variation of strain in the SG-L0 strain gauge is indicated in Figure 10. The values of strain loss in this strain gauge after 1, 7, and 14 days were, respectively, 0%, 1.5%, and 1.5%. Final strain loss (after 14 days) in the SG-L0 (far from the pre-cracks), SG-L1 (near the pre-cracks) and SG-L2 (not as close as SG-L1 to the pre-cracks) strain gauges of S2L-40-PC2 slab was respectively, 1.5%, 10.1%, and 5.9%. These results confirm that the strain loss in the pre-cracked slabs was dependent on the relative distance between the strain gauges and the pre-cracks.

Figure 13.

Location of strain gauges on the RC slabs after application of pre-cracks: (a) S2L-20-PC1, (b) S2L-20-PC2, (c) S2L-40-PC1, and (d) S2L-40-PC2 (the pre-cracks are the blue lines).

Conclusions

To appraise the effect of the level of damage (pre-cracks) on the behavior of RC slabs flexurally strengthened with prestressed NSM CFRP laminates, an experimental program was carried out. From the obtained results, it can be concluded that the main difference on the behavior of prestressed NSM CFRP slabs with and without pre-cracks resides in an expected loss of initial stiffness in the pre-cracked slabs. The comparison of the flexural strengthening performance in damaged and non-damaged RC slabs indicates that the analyzed levels of damage has a small influence on the ultimate load carrying capacity of the structural elements. Regardless of the level of damage applied to the RC slabs, all the strengthened slabs failed by the rupture of the CFRP laminates after yielding of the tension steel reinforcement, showing that the level of damage did not change the failure modes of the RC slabs.

Strengthening damaged RC slabs with prestressed NSM CFRP laminates resulted in a significant increase of load carrying capacity at serviceability limit states. By applying a 20% prestress level in the NSM CFRP laminates, the service load increased between 66% and 82%, when compared to the corresponding values of the reference slab. A 40% prestress level resulted an increase that ranged between 105% and 132%. In terms of ultimate limit state, the increase of the maximum load (compared to the corresponding values of the reference slab) has ranged between 111% and 122% for 20% of prestress and between 114% and 120% for 40% of prestress. In the strengthened damaged RC slabs with prestressed CFRP laminates the deflection at maximum load was more than 1.8 times the deflection at yield initiation, with substantial plastic incursion of the steel bars, which guarantees the required level of ductility for the RC slabs. However, regardless of the level of damage, a decrease of the ductility level of these slabs was observed with the increase of the CFRP prestress level.

The values of strain in CFRP laminates in prestressed slabs with or without damage (pre-cracks) were surveyed over time and along the length of the slabs during the period of time that ranged between 4 and 31 days. The maximum strain loss in the prestressed CFRP laminates of the tested slabs, which ranged between 27% and 79%, was observed in the closest strain gauge to the extremity of the CFRP laminates (at a distance of about 25 mm). The experimental results revealed that the transfer length of CFRP laminates is less than 150 mm. Out of this transfer length in the non-damaged RC slabs, the strain loss was almost constant along the length of the CFRP laminate (less than 8%), while the maximum value of strain loss out of transfer length in the damaged RC slabs was measured near the pre-cracks (around 14%). A significant part of strain loss in CFRP laminates occurred during 24 h after releasing. The difference of the strain loss between 7 and 1 days after releasing the prestress load in the CFRP laminates was no more than 9.5% and the maximum value of the additional strain loss after 7 days was 7%.

Footnotes

Acknowledgements

The authors wish to acknowledge the collaboration provided by the “Empreiteiros Casais” and S&P in the execution of the experimental program.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by FEDER funds through the Operational Competitiveness and Internationalization Programme (POCI) and National Funds through FCT—Portuguese Foundation for Science and Technology under the project POCI-01-0145-FEDER-030956. The authors greatly appreciate this support.

ORCID iDs

M.R. Mostakhdemin Hosseini

Salvador J.E. Dias

References

ACI 440.2R-08 (2008) Guide for the design and construction of externally bonded FRP systems for strengthening concrete structures.

Aslam

Shafigh

Jumaat

, et al. (2015) Strengthening of RC beams using prestressed fiber reinforced polymers - a review. Construction and Building Materials 82: 235–256.

Badawi

Soudki

(2009) Flexural strengthening of RC beams with prestressed NSM CFRP rods - experimental and analytical investigation. Construction and Building Materials 23: 3292–3300.

Badawi

Wahab

Soudki

(2011) Evaluation of the transfer length of prestressed near surface mounted CFRP rods in concrete. Construction and Building Materials 25: 1474–1479.

Barros

JAO

(2009) Pre-stress technique for the flexural strengthening with NSM-CFRP strips. In: Oehlers

Griffith

Seracino

(Eds). Proceedings of the 9th international symposium on fiber reinforced polymer reinforcement for concrete structures (FRPRCS-9), Sydney, Australia.

Barros

JAO

Dias

SJE

Lima

JLT

(2007) Efficacy of CFRP-based techniques for the flexural and shear strengthening of concrete beams. Cement and Concrete Composites 29(3): 203–217.

Barros

JAO

Fortes

(2005) Flexural strengthening of concrete beams with CFRP laminates bonded into slits. Cement and Concrete Composites 27(4): 471–480.

Bilotta

Ceroni

Nigro

, et al. (2015) Efficiency of CFRP NSM strips and EBR plates for flexural strengthening of RC beams and loading pattern influence. Composite Structures 124: 163–175.

Bonaldo

Barros

JAO

Lourenço

PJB

(2008) Efficient strengthening technique to increase the flexural resistance of existing RC slabs. Journal of Composites for Construction 12(2): 149–159.

10.

Costa

Barros

JAO

(2010) Flexural and shear strengthening of RC beams with composites materials - the influence of cutting steel stirrups to install CFRP strips. Cement and Concrete Composites 32: 544–553.

11.

Costa

Barros

JAO

(2015) Tensile creep of a structural epoxy adhesive: experimental and analytical characterization. International Journal of Adhesion & Adhesives 59: 115–124.

12.

Dalfré

Barros

JAO

(2013) NSM technique to increase the load carrying capacity of continuous RC slabs. Engineering Structures 56: 137–153.

13.

De-Lorenzis

Nanni

(2002) Bond between near surface mounted FRP rods and concrete in structural strengthening. ACI Structural Journal 99(2): 123–132.

14.

El-Hacha

Gaafar

(2011) Flexural strengthening of reinforced concrete beams using prestressed, near-surface-mounted CFRP bars. PCI Journal 56(4): 134–151.

15.

El-Hacha

Riskalla

(2004) Near-surface-mounted fiber-reinforced polymer reinforcements for flexural strengthening of concrete structures. ACI Structural Journal 101(5): 717–726.

16.

EN 10002-1 (1990) European standard CEN: Metallic materials - Tensile testing. Part 1: Method of test (at ambient temperature).

17.

EN 1992-1-1 (2004) Eurocode2: Design of concrete structures Parte 1-1: General rules for buildings>.

18.

EN 206-1 (2000) European standard CEN: Concrete - Part 1: Specification, performance, production and conformity.

19.

Esfahani

Kianoush

Tajari

(2007) Flexural behaviour of reinforced concrete beams strengthened by CFRP sheets. Engineering Structures 29(10): 2428–2444.

20.

Hajihashemi

Mostofinejad

Azhari

(2011) Investigation of RC beams strengthened with prestressed NSM CFRP laminates. Journal of Composites for Construction 15(6): 887–895.

21.

Hong

Park

(2016) Effect of prestress and transverse grooves on reinforced concrete beams prestressed with near-surface-mounted carbon fiber-reinforced polymer plates. Composites Part B: Engineering 91: 640–650.

22.

Hosseini

MRM

Dias

SJE

Barros

JAO

(2014) Effectiveness of prestressed NSM CFRP laminates for the flexural strengthening of RC slabs. Composite Structures 111: 249–258.

23.

ISO 527-5 (1997) ISO standard: Plastics - Determination of tensile properties - Part 5: Test conditions for unidirectional fibre-reinforced plastic composites.

24.

Kotynia

Przygocka

(2018) Preloading effect on strengthening efficiency of RC beams strengthened with non-and pretensioned NSM strips. Polymers 10(2): 145.

25.

LNEC E397-1993 (1993) Portuguese specification from LNEC: Concrete - Determination of the elasticity modulus under compression.

26.

Peng

Zhang

Cai

, et al. (2014) An experimental study on reinforced concrete beams strengthened with prestressed near surface mounted CFRP strips. Engineering Structures 79: 222–233.

27.

Pham

Al-Mahaidi

(2004) Experimental investigation into flexural retrofitting of reinforced concrete bridge beams using FRP composites. Composite Structures 66: 617–625.

28.

Rezazadeh

Barros

JAO

(2015) Transfer zone of prestressed CFRP reinforcement applied according to NSM technique for strengthening of RC structures. Composites Part B: Engineering 79: 581–594.

29.

Rezazadeh

Costa

Barros

JAO

(2014) Influence of prestress level on NSM CFRP laminates for the flexural strengthening of RC beams. Composite Structures 116(1): 489–500.

30.

Sharaky

Torres

Sallam

HEM

(2015) Experimental and analytical investigation into the flexural performance of RC beams with partially and fully bonded NSM FRP bars/strips. Composites Structures 122: 113–126.