Repairing of reinforced concrete continuous beams by CFRP sheets

Although continuous beams may be found in a range of projects such as garages, bridges, and multi-story structures, studies have been still restricted to a limited field. Flexural cracks are a common issue in continuous beams; therefore, this article outlines an investigation study used to assess the performance of two-span reinforced concrete beams repaired by attaching (CFRP) sheets. The program included seven beam specimens with a length of 2800mm and rectangular cross sections of 150*250mm. All beams were strengthened externally on the tension zones via CFRP-sheet considering changing the ratio of sheet length to beam span which is 0.5, 0.7, and 0.9 except one was chosen as a reference beam. Repaired beams were preloaded with damage ratios of 45% and 65% with respect to the reference beam. The findings showed that using style 0.7L for both positive and negative regions achieve an appropriate restored percentage of ultimate capacity of about 101.7% and 98.2%. In addition, eliminating the sheet length in the positive moment regions gives higher stiffness. it is also found that when the CFRP sheet length in the tension part is increased, the tensile rupture of the sheet was the dominant failure mode.


Introduction
Although the continuous beam can be found in a variety of projects such as parking garages, bridges, and multi-story buildings, the research has been still limited in this area [1] [2].Predominantly over the last twenty years, there has been a substantial increase in the practical application of FRP strips in the flexural reinforcement of concrete beams, which has resulted in a positive impact on their performance [3].It is a beneficial method to bond carbon-fibre polymer sheets externally on the surface of the beam that undergoes bending stresses for repairing and strengthening purposes [4].
Repairing RC structures has grown more and more crucial, particularly in the last ten years.Cracks and deflections are two common deterioration types that cause damage to many RC constructions.These are impacted by a variety of causes including earthquakes, vibrations, corrosion of reinforced bars, and alterations in the environment [5].In the other words, repairing damaged concrete buildings is fundamental not just to extend their life span but also to ensure their serviceability and security.
A structural element's failure mode is often more brittle after it has been strengthened with FRP technologies.The interfacial debonding failure was dominated by the strengthening beams followed by the concrete crushing failure [6].Several studies obtained increasing in loading capacity and reducing in ductility for FRP-strengthened continuous beams, even though ductility is more important, particularly in statically indeterminate structures such as continuous beams, because it allows moment redistribution via rotations of plastic hinges [7] [8].Because of their lightweight, high tensile strength, high durability, corrosion resistance, and easy installation, external bonding of FRP composites has become a new structural strengthening technology in response to the growing need for retrofitting RC structures [9].
Wight and Erki [10] found that the using of prestressed sheets was more successful in load vs deflection than non-prestressed sheets.As well as, the dominant failure mode in the strengthened beam with nonprestressed FRP materials was debonding interfacial, but in the beam that was strengthened by prestressed FRP materials was tensile rupture of the CFRP sheet.Al-Khafaji and Salim [11] proved that as the ratio of the width of CFRP to the width of concrete increased, the ultimate load and stiffness for continuous Tbeams strengthened with CFRP enhanced.Strengthening by CFRP for preloading 100% and 90% beams can be enhanced the load capacity by 51.5% and 68.5% respectively [12].Hamrat et al. [13] showed experimentally that adding more than the layer of FRP laminates/sheets achieved an increase in the ultimate load capacity for the repaired beams of about (from 13 -100) %, with decreasing in deflection at mid-span to about (15 -35) %.Furthermore, Li et al. [14] deduced that the debonding failure has a higher effect on the early cracking loads than on ductility, stiffness, and ultimate loads.
The present study aimed to provide a streamlined approach for evaluating the flexural behavior of RC continuous beams repaired with externally bonded CFRP sheets as well as studying the impact of changing the sheet length in positive and negative regions, and the damage level in the continuous beams through measuring of ductility, stiffness, crack pattern and load versus deflection curve.Another target that choosing the best style of external bonding beams that may give a high ultimate capacity with a small value of deflection with acceptable ductility.

Continuous beams specimens
Seven reinforced concrete two equal span beams were involved in this study with full-scale rectangular sections (150*250) mm and (2800mm) total length.All the beams were designed with extra reinforcement in shear to ensure the occurrence of flexural failure whereas they have the same dimensions and compressive strength.The first beam was chosen as a reference beam which was without bonding by CFRP sheets and tested until failure.The width and thickness of the sheets were constant in all the samples that were identified by 50mm, and 0.167 mm respectively.Figure 1 explains the details of the continuous beam in terms of reinforcement details in the rectangular section, supporting conditions, the applied load, as well as the bending moment diagram.
However, figure 2 presents the positions of gluing CFRP sheets in positive and negative moment zones.Firstly, beams B2 and B7 were attached by CFRP sheet 0.7L for both regions specifically under the center of both spans and up the middle support part.Moreover, B3 and B8 were glued by the longer sheets, 0.9L in the sagging zone and 0.7L in the hogging zone.Finally, in B4 and B9, the CFRP sheet length was eliminated to 0.5L in the negative moment area and kept at 0.9L for the positive moment area.The procedure that was followed to bond the samples by CFRP sheets: 1.
Grinding the beam surface by flap disk to remove the outside weak surface of the concrete.

2.
Wash the surface with water to clean it from the dust and dry it.
Mixing two components of epoxy and painting them on the beam surface and the sheet.

5.
Place the sheets on the beam by roller to remove the air voids.6.
Painting another layer of epoxy on the sheets.7.
Waiting until the epoxy curing.

Material properties
The continuous beams were made from resistant Portland cement with (19mm) maximum size of aggregate and 28-day and target compression strength of 35 MPa.Three cubes, cylinders, and prisms were manufactured from the same mixture for the beams.Then, all the specimens were subjected to different tests, and the average strength values were recorded as shown in table 1

crack pattern
Firstly, it was noticed the crack pattern for the seven continuous beams that are shown in figure 3, as the crack development rate in the control beam and the repaired specimens.Furthermore, the flexural crack was the first crack that was noticed clearly at 85KN upward in the middle support in the tension zone of the control beam.Then, the second cracks were propagated in tension regions diagonally, that classified as shear cracks and they had a large number with a small width in repaired beams.Finally, the yielding of steel bars began and the crushing concrete followed them.

load-deflection curve
Table 4 and figure 4 illustrate the relationship between the applied load at each span and the observed deflection at the midspan.Because the measured deflections in each beam's two spans were identical, one side midspan deflections are recorded via one dial gauge with 0.01 mm accuracy.In general, all repaired beams gave less deflection than the reference one.By discussing the curve (a) B2 and B7, both beams were corresponding with the control until the load 120KN with a nonlinear shape.After this stage, the CFRP sheet started to improve the capacity with the deflection of both beams gradually.However, in B2 which has a damaged ratio of 45%, the steel bars reached to yielding state and then the concrete crushing occurred, resulting in a failure in 290KN with the debonding sheet.In addition, B2 and B7 restored 101.7%, and 98.2% from the ultimate capacity of the control respectively.On the other hand, in figure 4 (b) and (c), B3 and B4 are conducted like a normal resistance of concrete because they have fewer cracks, so the concrete can resist more loading.Whilst in B8 and B9, it is clear to see the inducing the catenary action in the CFRP sheet behaved which means that the stiffness of the member may be increased at a higher load stage or in other words when the deflection increases.That is due to the inducing internal tensile force in the carbon sheets.When cracks begin to form and bar-reinforcement of steel yields, further cracks occur in the tension zone, concrete stresses are freed, and sheet fibres bear greater loads than concrete.Also, the effect of the sheet length might be considered because B8 and B9 had a long length of the sheet (0.9L) under a loading force.about 80% of the load in figure 4 (b) and (c), the sheet and concrete resistance tended to reach a similar damaged stage in the same style of strengthening.After that percentage, concrete and sheets started to work together to carry the applied load until the four beams failed.As a result, the continuous beams with 65% damaged percentage achieved an acceptable ratio of restoring the load, which may be regarded as the catenary action induced in the sheets.

Ductility
Ductility-index (∆ ) is predicted according to Park [15] and indicated in table 5, it is calculated from the equation: = (∆u /Δy), Whereas ∆u: the largest deflection in mid-span at ultimate load failure of the beam.
Δy: the deflection for yielding the reinforcement steel bar.It is clear to observe that in the table, most of the repaired specimens recorded less ductility index than the reference beam, this reason belongs to the external bonding of CFRP sheet techniques.However, repairing beams with a damage ratio of 45% resulted in a lower percentage of the loss in ductility than repairing beams with a damage ratio of 65%, due to the cracking in this stage being less.It's worth mentioning that increasing the Lsheet/Lspan causes a decrease in the ductility index because the sheets cover a greater length in the tensile area.

Stiffness
Table 6 explains the initial stiffness which is estimated by the slope of the load versus deflection curve at (0.75Pu) according to Ngoc et al. [16].In B2 and B3, when the length of the sheets in the positive area is increased without changing the length in the negative area, the stiffness decreased from 9.3% to 0.8%, as well as the stiffness reduced from 18.1% to -5.6% in B7 and B8 which were damaged level is 65%.Also, it is clear that to note in B4 and B3 with keeping the positive region constant by 0.9L while the negative region increased from 0.5L to 0.7L, the stiffness decreased from about 10.1% to 0.8%.Furthermore, there was a decrement in stiffness in B9 and B8.The possibility of explaining this is attributed when reducing the length of the sheet, the spread of the tensile force induced by catenary action reduces, which leads to an increase in the stiffness.Generally, most of the beams had a higher percent of stiffness than the control beam except B8, and B9.

Mode (II).
Rupture tensile of CFRP-sheets was the dominant mode in beams (B3, B7, B8).The breaking of the CFRP sheets was abrupt and loud, signaling a speedy energy release in figure 6.

Mode (IV).
Peeling of concrete cover failure close to the end of CFRP-sheets happened for B9 as explained in figure 8. Furthermore, it was inflexible and swiftly followed by beam rupture.Peeling failure of the cover concrete was not avoided by lengthening the CFRP sheet to cover the full positive moment zone as in B9.
Figure 8. Peeling cover of concrete.

Conclusion
The majority of finding from this study could be summed up: x the effect of using 0.7 for Lsheet/Lspan was the most reasonable ratio restored of ultimate capacity by 101.7% and 98.2% for 45%, and 65% damage ratio respectively.x When elongation of the length of the CFRP sheet on tension parts in both damaged percentages, the loss of ductility index was large and the tensile rupture of the sheet was the dominant failure.x The effect of Carbon-FRP sheet performance like catenary action was very clear in the 65% damaged ratio.x it was found that by reducing the sheet length around the positive moment zone, the stiffness had a considerable enhancement.

Figure 1 .
Figure 1.(a) Longitudinal section for the beam with reinforcement details (b) Cross section for the beams (c) Bending moment diagram for general two-span beam has length L and load P in each span.

Figure 2 .
Figure 2. Beam strengthened by CFRP sheet: (a) 0.7L for positive moment and 0.7L for negative moment (b) 0.9L for positive moment and 0.7L for negative moment (c) 0.9L for positive moment and 0.5L for negative moment.Table 2. Mechanical properties of (CFRP) sheets.

Figure 4
Figure 4 load-deflection curve for the control beam with two beams have the same style of strengthening as shown in (a), (b), (c).

3. 5 . 1 .
Mode (I).The part of the control beam in figure5explained that the failure was flexural in a conventional manner.However, the crushing of concrete happened under plate loading in the centre of the spans and the middle support.This means the tension (steel bars) yielded, and then the compression (concrete) failed for the control beam (B1).

Figure 6 .
Figure 6.Tensile rupture of sheet).3.5.3.Mode (III).In beams B2 and B4, the separation of CFRP-sheet without concrete attachment was observed figure7.It is called debonding interfacial failure.Before the beam collapsed, there was a slight noise that suggested the glue was about to fail 90% from loading failure approximately.

Table 1 .
Test of averages three different molds.
. For the cylindrical ratio concrete strength test, fcu was multiplied by 0.8 as in relation (fc=0.8*fcu), it was 35MPa.As shown in figures 1 (a) and (b).Bars of 10 mm in diameter were put in the top and bottom specifically in tension regions.Moreover, the maximum yield and tensile strength were 603, and 670MPa respectively with an elongation percent of 15.2.However, the mechanical characteristics of both the CFPR sheets and bonding adhesive (epoxy) can be obtained from table 2 and table 3.

Table 3 .
Mechanical properties of epoxy paste.

Table 4 .
Ultimate capacity and failure modes of continuous beams.

Table 5 .
Ductility index for testing continuous beams.

Table 6 .
Stiffness for tested beams.