Accepted Manuscript

CFRP-ECC hybrid for strengthening of the concrete structures

Chao Wu, Victor C. Li

PII: S0263-8223(16)32648-4
DOI: http://dx.doi.org/10.1016/j.compstruct.2017.07.034
Reference: COST 8689

To appear in: Composite Structures

Received Date: 25 November 2016

Revised Date: 1 May 2017
Accepted Date: 12 July 2017

Please cite this article as: Wu, C., Li, V.C., CFRP-ECC hybrid for strengthening of the concrete structures, Composite
Structures (2017), doi: http://dx.doi.org/10.1016/j.compstruct.2017.07.034

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 CFRP-ECC HYBRID FOR STRENGTHENING OF THE CONCRETE STRUCTURES

Chao Wu1,2 and Victor C. Li2,*

1
School of Transportation Science and Engineering, Beihang University, 37 Xueyuan Road,

Beijing 100191, China

wuchao@buaa.edu.cn
2
Department of Civil and Environmental Engineering, University of Michigan, Ann Arbor, MI

48109, USA

* Corresponding author E-mail: vcli@umich.edu; Tel.: +1 734 764 3368

ABSTRACT: The strengthening of concrete structures using FRP composite and polymer

adhesive has been challenged considering structural fire. This technology is limited to situations

with low fire hazards. Though the cementitious adhesive showed improved fire resistance than

the polymer counterpart, the former faces challenges of brittleness and high temperature spalling.

This paper proposes a CFRP-ECC hybrid system for strengthening of the concrete structures.

This hybrid consists of CFRP composite embedded in the ECC matrix. CFRP functions as the

main load-carrying element, while ECC acts as the adhesive layer for protecting the CFRP and

transferring the load between the concrete structure and CFRP. ECC is ductile in tension and has

been shown to have improved fire performance without spalling over normal cementitious

materials. In this paper, ECC was developed with a tensile strain capacity of 3%. Then direct

pull-out tests were conducted to quantify the interface behavior between CFRP and ECC, and to

obtain the effective bond length of CFRP. Finally a concrete beam with the CFRP-ECC hybrid

strengthening was tested under four-point bending. The experimental results showed that the

effective bond length of CFRP in ECC was longer than 170 mm. When the CFRP-ECC hybrid

was used for strengthening of the concrete beam, premature debonding occurred at the interface

1
between CFRP-ECC hybrid layer and the surface of the concrete beam, thus the flexural

behavior of the concrete beam was barely improved. Therefore, it is recommended that proper

surface treatment of the concrete beam should be prepared or innovative anchoring techniques

should be developed before the CFRP-ECC hybrid can be used for the strengthening of the

concrete structures.

Key words: Carbon fiber reinforced polymer (CFRP); Engineered cementitious composite

(ECC); Bond; Strengthening; Concrete; Beam

1. Introduction

There has been a considerable increase in the demand of repair and strengthening of the concrete

structures worldwide [1-3]. With high strength to weight ratio and desirable resistance to the

harsh environment, fiber reinforced polymer (FRP) composites have been widely accepted for

the strengthening of the concrete structures [1-8]. Polymer adhesives, like epoxy, are normally

adopted as the bonding agent [9-11]. One major issue with the polymer adhesives is that, when

the glass transition temperature (Tg) is exceeded, the mechanical properties of the adhesive is

reduced and the strengthening effect of the FRP cannot be maintained [12-17]. Unfortunately,

most of the polymer adhesives used for the structural strengthening have a Tg less than 100 oC

[18]. A study in [19] found that the efficiency of the CFRP strengthening of the concrete

structures would extensively reduce by 65% at 300 oC under elevated temperatures [20].

In order to solve this problem, cement based adhesives are introduced for the FRP

strengthening of the concrete structures [12, 21-23]. Although the bond behavior between the

FRP and the concrete is improved under the elevated temperatures, another problem of the

cement based adhesives has been observed [21, 24, 25]. Since most of the cement based

2
adhesives lack ductility, the load cannot be effectively transferred from the concrete to the FRP

composite. Also the failure is generally brittle and lack warning [21]. It was found that the FRP-

mortar system was generally less effective than the FRP-epoxy system in strengthening. For

example, it was concluded that the FRP-mortar system was 45% [26] and 50% [27] less effective

than the FRP-epoxy counterparts.

Therefore, it is necessary to develop a cementitious adhesive with desired ductility,

which can effectively distribute and transfer the load from the concrete substrate to the FRP

composite, without deterioration under elevated temperatures. Engineered Cementitious

Composites (ECC) is a promising candidate for such strengthening applications [28, 29]. ECC

was developed based on the micro fracture mechanics with a strain capacity in the range of 3~7%

[30], comparing to 0.1% for ordinary Portland cement (OPC) mortar. This ductility of ECC is

achieved by micro-cracking (with micro-crack width less than 100 µm [30]) in the strain

hardening stage [31]. More importantly, ECC can effectively resist spalling failure under

elevated temperatures up to 800 oC, due to the melting of PVA fibers [32].

Considering these advantages, ECC has great potential as adhesive for the FRP

strengthening of the concrete structures [33-35]. In a recent study [33], ECC was used as the

adhesive for bonding the basalt fiber reinforced polymer (BFRP) grid to the concrete beams for

the flexural strengthening. Uniform tiny cracks were observed in ECC during the loading process,

and the load carrying capacity of the concrete beam was increased by 35% maximum. The

authors of this paper investigated the CFRP-ECC hybrid behavior under elevated temperatures

up to 500 oC [35]. It was found that ECC showed desirable anti-spalling behavior with the

temperature up to 500 oC. The compression and tensile strengths of ECC showed little reduction

with the temperature up to 300 oC. It was found that the interface between ECC and the concrete

3
substrate was sensitive to the temperature, and premature failure may occur at the interface

during the pull-out tests of CFRP. However, the existing studies on the FRP-ECC hybrid for the

strengthening of the concrete structures only focused on the failure mechanism and the

improvement in the load-carrying capacity of the concrete beams. Very few work reported the

comprehensive characterization of the ECC material, and fundamental investigation on the

CFRP-ECC interface behavior (e.g. effective bond length).

As part of a project on the development of the CFRP-ECC hybrid for the strengthening

of the concrete structures, this paper presents comprehensive experimental results of this hybrid

under room temperature. The thermal-mechanical behavior of this hybrid under elevated

temperatures has been reported in [35]. This paper firstly focused on the development of ECC as

the adhesive for the CFRP strengthening. Then the interface behavior between CFRP and ECC

was investigated through direct pull-out tests. Various CFRP bond lengths were used in the pull-

out tests in order to find out the effective bond length of the CFRP-ECC hybrid. Finally, four-

point bending tests were conducted on steel reinforced concrete beams strengthened with the

CFRP-ECC hybrid. The failure modes and the load-displacement responses were reported. For

comparison purpose, the corresponding mortar specimens were also prepared and tested under

the same condition as ECC specimens in all tests. The advantages of ECC as the adhesive

comparing to the cement mortar were discussed. This paper contributes to the fundamental

understanding on the interface behavior between CFRP and ECC, and provides comprehensive

experimental data and necessary precautions for the future design of the CFRP-ECC hybrid for

the strengthening of the concrete structures.

2. Experimental Program

4
The proposed method using the CFRP-ECC hybrid for the strengthening of the concrete

structures are tested with an experimental program of three steps. Firstly, ECC was developed

and the material properties were characterized through compressive and tensile tests. Then direct

pull-out tests were conducted to study the interface behavior between CFRP and ECC. Finally,

four-point bending tests were carried out on the steel reinforced concrete beams to investigate the

strengthening effect of the CFRP-ECC hybrid system. The three testing steps will be described

respectively in the following sections.

2.1 Compression and tension tests of ECC

The mixture proportion of ECC used in the current study is presented in Table 1. The major

matrix ingredients of ECC include Type I ordinary Portland cement, Class F fly ash, fine

aggregate (F-75 silica sand), water and water reducer (ADVA-190). The fly ash was provided by

the DTE Monroe Power Plant in the State of Michigan. The chemical properties of the fly ash

provided by the manufacturer are listed in Table 2. The F-75 silica sand has a maximum and an

average grain size of 250 µm and 110 µm, respectively. The Poly-Vinyl Alcohol (PVA) fibers

are 12 mm long with a diameter of 39 µm. The PVA fibers are coated with 1.2% by weight of oil

to control the fiber/matrix interfacial properties. The volume content of the PVA fiber in ECC

was 2%. The detailed information of the PVA fiber has been well documented in the publications

by the ACE-MRL group at the University of Michigan [32] as summarized in Table 3.

The ECC mixture was prepared in a Hobart type mixer of 10 L capacity. Firstly, cement,

sand and fly ash were dry mixed for 5 mins at the low speed of the mixer. Then the mixture of

water and water reducer was added and mixed for another 4 mins under the same speed. Then the

mixture was checked and the blocks were broken to ensure a homogeneous paste status. The

5
mixing continued for another 2 mins until the paste reached a good fluidity. Then the PVA fibers

were added slowly into the mortar and mixed for 5 mins under the medium speed of the mixer.

Then the mixture was checked and the glomeration of the fibers was broken to ensure a uniform

dispersion of the fibers. Finally, the mixing continued for another 3 mins until the fibers were

well dispersed.

The fresh ECC was cast into the molds that were moderately vibrated using a vibration

table. The molds were sealed with plastic bags and subsequently demolded after 1 day curing.

All specimens were cured for another 27 days in air (23±3 oC; 30 ± 10% RH) before testing.

Both cube specimens and dogbone-shaped specimens were prepared for compressive and tensile

testing respectively. The cube specimens had a dimension of 50×50×50 mm3 according to

ASTM C190/C190M [38]. The dimensions of the dogbone-shaped specimens can be found in

[39], which were recommended by the Japan Society of Civil Engineers (JSCE) [40] for

standardized tensile testing of ECC.

The compression tests were conducted on a FORNEY F-50F-F96 machine under load

control at a speed of 20 MPa/min. The peak load was recorded to determine the compressive

strength. The tension tests were carried out on a servo-hydraulic Instron 5969 testing system with

a capacity of 50 kN, using displacement control at a speed of 0.5 mm/min. The experimental

setup of the tensile testing is presented in Fig. 1. Two linear variable displacement transducers

(LVDT), with a gage length of approximately 100 mm, were attached to the specimen. Stress-

strain curves were then recorded to determine the behavior of the specimens under direct tension.

For comparison purpose, mortar cube and dogbone specimens (ECC without PVA fibers)

were also prepared. The mix proportion, preparation process and testing setup were all the same

as ECC specimens except that mortar specimens were without PVA fibers.

6
2.2 Pull-out test of CFRP-ECC hybrid system

Direct pull-out tests were conducted in order to investigate the interface behavior between CFRP

and ECC of the hybrid system. Since this system was proposed for the strengthening of the

concrete structures, a concrete block was also included when preparing the pull-out specimen.

The experimental setup and dimensions of the specimen are presented in Fig. 2. The specimen

seated on a steel base plate and was secured in position by the top steel plate which was fastened

to the base plate through two steel rods. A stopper was mounted on the base plate to prevent any

lateral movement of the specimen during the loading process. The stopper can also be adjusted

when installing the specimen so that the CFRP sheet was in the vertical direction. A spirit level

was used to check the straightness of the CFRP sheet before testing. The CFRP sheet was 50 mm

wide and was embedded in the middle of the ECC layer which was attached to the concrete

block. Two LVDTs were used to record the relative movement between the CFRP sheet and the

ECC. The LVDTs were the same as those in Fig. 1. The gauge length of each specimen was

recorded and used for the calculation of the relative slip between CFRP and ECC (see the

detailed discussions in Section 3.2).

CFRP sheet was Tyfo SCH-41 provided by MTC Corporation in St. Louis with a

measured thickness of 0.55 mm. The tensile tests were conducted according to the ASTM D3039

[41]. The tensile strength was not achieved due to the difficulty of clamping the CFRP sheets.

The tested modulus was 79.7 GPa with the applied stress up to 833.5 MPa. Six repeating tests

were conducted to measure this modulus. The nominal tensile strength of CFRP sheet was 986

MPa. The mixture proportion of ECC was the same as that in Table 1. The concrete mixture is

presented in Table 4. The compression strength of the concrete was tested using a concrete

7
cylinder of 75 mm in diameter according to ASTM C873/C873M [42]. The 28-day compressive

strength was 40.5 MPa.

The preparation procedure of the pull-out specimens is presented in Fig. 3. Firstly, the

surface of the concrete block was prepared after 28 days of curing. The cement on the surface

was removed and the course aggregates were exposed. The purpose of the surface treatment was

to ensure the bonding quality between ECC and concrete, so that this interface would not be a

problem during the pull-out tests. Then the concrete blocks were placed in a mold, where the

CFRP sheets can be secured in position. The first layer of the fresh ECC of 12.5 mm thick was

applied on top of the concrete block and underneath the CFRP sheet. Finally another layer of

ECC of 12.5 mm thick was casted on top of the CFRP sheet. The thickness of the ECC layer was

25 mm with the CFRP sheet embedded in the middle. The specimens were demolded after 1 day

and cured for another 27 days in air (23±3 oC; 30 ± 10% RH). Because the concrete block was

cured for 56 days before pull-out tests, its 56-day compressive strength was also tested which

was 41.7 MPa (only 3% higher than the 28-day strength which was 40.5 MPa).

In order to study the effect of the embedded length of CFRP on the interface behavior

between CFRP and ECC, five embedded lengths were selected as 25 mm, 50 mm, 75 mm, 100

mm, 135 mm and 170 mm. For each embedded length, at least three identical specimens were

prepared and tested for repeating purpose. The pull-out tests were conducted on a MTS 810

material testing system with a loading capacity of 100 kN. Displacement control was used at a

loading speed of 0.5 mm/min. The pull-out test was continued until the load dropped

dramatically or the CFRP sheet was pull out by 30 mm from LVDT reading.

For comparison purpose, the pull-out specimens were also prepared with CFRP

embedded in a mortar layer (ECC without PVA fiber, see Table 1). The mixture proportion of

8
the mortar was the same as that of ECC but without PVA fibers. The specimen preparation

process and the loading scheme were kept the same as the ECC specimens.

2.3 Four-point bending tests on concrete beams

Three identical steel reinforced concrete beams were prepared. The first beam was used as

reference. The second beam was strengthened with CFRP-ECC hybrid system. The third beam

was the same as the second beam except that ECC was replaced with mortar (ECC without PVA

fiber, see Table 1). The dimensions of the beam and the four-point bending test setup are

presented in Fig. 4. The total length of the beam was 1524 mm (60 in) with a span length of 1372

mm (54 in). As can be seen in Fig. 4, the strengthening layer was 1067 mm (42 in) long and was

not extended under the supports. This was because in real applications, the concrete beams are

normally rigidly connected to the columns, which makes it difficult to extend the strengthening

layer into the column.

The longitudinal steel rebar was 12.7 mm (0.5 in) in diameter and the stirrup was 6.4

mm (0.25 in) in diameter. The stirrups had a spacing of 76 mm (3 in). Both rebar and stirrup

were Grade 60 according to ASTM A615/A615M [43] with a nominal yielding strength of 469

MPa. The concrete mix proportion was the same as in Table 4. The strengthening layer (25 mm

thick) was either CFRP-ECC hybrid or CFRP-mortar hybrid for comparison purpose. The mix

proportion of ECC or mortar can be found in Table 1.

Fig. 5 shows the preparation process of the strengthened concrete beam. It should be

noted that, the concrete beam was casted in the mold in an upside down position, which made it

easier to apply the hybrid strengthening layer. Firstly, the bottom surface of the concrete beam

was treated after 28 days curing with the coarse aggregates exposed. This was to ensure a better

9
interface bonding between concrete and ECC. Then a layer of ECC of 12.5 mm thick was

applied on the concrete surface. The CFRP sheet of 100 mm was placed on top of the first ECC

layer and finally the second layer of ECC of 12.5 mm thick was applied. The CFRP sheet was

thus embedded in the middle of ECC layer.

The four-point bending tests were conducted on a servo-hydraulic testing system with a

loading capacity of 500 kN. The tests were under displacement control with a loading speed of 1

mm/min. The mid-span displacement was recorded with an optical tracking system, Optotrak

Certus, from NDI Measurement Sciences. The experimental setup of the four-point bending tests

is shown in Fig. 6.

3. Experimental Results and Discussions

3.1 Compressive and tensile properties of ECC

For the compressive strength of ECC, six cube specimens were tested and the average

compressive strength was 50.9 MPa. The ECC mortar without PVA fiber was also tested and the

compressive strength was 52.6 MPa. Five specimens were tested for the tensile properties of

ECC and mortar, respectively. The tensile stress-strain curves of ECC at the age of 28 days are

plotted in Fig. 7, with a typical multiple-cracking image of ECC. The tensile properties of ECC

and mortar are listed in Table 5.

In Table 5, the first crack strength was determined from the point where the stress-strain

curve becomes non-linear [40]. Peak stress and the corresponding strain were defined as “tensile

strength” and “tensile strain”, respectively. In the case when slight stress decrease after the peak

stress was observed, the strain at 95% of the peak stress was used as the “tensile strain” [39]. The

strain hardening behavior of ECC is obvious in Fig. 7, with the tensile strain almost 252 times

10
higher than that of the mortar (see Table 5).

3.2 Interface behavior of the CFRP-ECC hybrid system

Pull-out tests were conducted to investigate the interface behavior between CFRP and ECC.

Various CFRP embedded length were selected as 25 mm, 50 mm, 75 mm, 100 mm, 135 mm and

170 mm. For each embedded length, at least three specimens were tested for repeating purpose.

The detailed experimental setup can be found in Section 2.2. The stress versus the relative slip

between CFRP and matrix are presented in Fig. 8. The stress, σ, is equal to the applied load, P,

divided by the section area of CFRP. The relative slip, δ, can be calculated by the following

equation:

୔ୈ
δ = Δ − ୛୲୉ (1)

where ∆ is the LVDT readings; D is the gauge length of LVDT (see Fig. 2); W and t are the

width and thickness of CFRP sheet which are 50 mm and 0.55 mm respectively; E is the

modulus of CFRP which was measured as 79.7 GPa.

As can be seen in Fig. 8, the ECC and mortar specimens showed similar development of

the pull-out stress against the relative slip between CFRP and the matrix (either ECC or mortar).

Firstly, the stress increased with the slip up to a peak point after which a fast decrease of the

stress was observed. Finally the stress became stabilized at a very low level with the increase of

the slip. The last stable stage of the stress was attributed to the friction between CFRP and the

matrix. It is obvious in Fig. 8 that the peak stress of ECC specimen seems proportionally increase

with the embedded length of CFRP, while it is not the case for the mortar specimens. The peak

stress stopped increasing for mortar specimens when the embedded length exceeded 100 mm

(Fig. 8b).

11
 The peak stress of each specimen was extracted from the stress-slip curve and plotted

against the embedded length in Fig. 9(a) for ECC specimens and in Fig. 9(b) for mortar

specimens. In Fig. 9, the symbol represents the average peak stress of repeating specimens and

the error bar refers to the standard deviation. It is clear that the pull-out peak stress of ECC

specimens increased with the CFRP embedded length. For example, the average peak stress of

ECC-25 was 71.6 MPa, and it increased to 174.2 MPa (143.3%) when the embedded length was

170 mm. Similarly, the peak stress of the mortar specimens increased from 37.2 MPa to 154.3

MPa when the CFRP embedded length changed from 25 mm to 100 mm. However, the

increasing trend stopped when the CFRP embedded length exceeded 100 mm. For example, the

peak strength of Mortar-135 and Mortar-170 was 133.5 MPa and 142.6 MPa, respectively.

The different increasing trends of the peak stress between ECC and mortar specimens

can be attributed to their different failure modes. All ECC specimens failed by CFRP being

pulled out from ECC (see Fig. 10a). However, two different failure modes were observed for the

mortar specimens. When the embedded length was shorter than 100 mm, CFRP was pulled out

from the mortar matrix (similar to Fig. 10a). When the embedded length exceeded 100 mm, the

mortar layer failed by shear cracking under the tensile loading transferred by the CFRP sheet.

Then the CFRP was pulled out from the rest of the mortar layer (see Fig. 10b). The failure modes

indicated that the CFRP-mortar hybrid may hardly sustain the load transferred from the CFRP

when the embedded length was longer than 100 mm. Therefore, the effective bond length of

CFRP with ECC as adhesive may be longer than 170 mm, whereas the effective bond length of

CFRP with mortar as adhesive was in the range of 100 mm to 135 mm. In a similar study by

Hosseini and Mostofinejad [44], CFRP sheet (SikaWrap230C) was bonded to the concrete prism

using external bonded reinforcement (EBR) technique using the adhesive of Sikadur 330. Single

12
shear tests were conducted and the corresponding effective bond length was identified as 35 mm.

The effective bond length reported by Hosseini and Mostofinejad was based on the conventional

EBR technique and was much smaller than the effective bond length of the CFRP-mortar or

CFRP-ECC hybrid in the current study.

3.3 Load-displacement curves of the concrete beams

Four-point bending tests were conducted on the concrete beams with either CFRP-ECC hybrid

strengthening or CFRP-mortar hybrid strengthening. The detailed experimental plan can be

found in Section 2.3. The failure modes of all beams are presented in Fig. 11.

For the reference beam, it was observed that the flexural cracks firstly extended from

the bottom and the beam finally failed by the concrete crushing on the top (Fig. 11a). As for the

beam strengthened with the CFRP-mortar hybrid, the flexural cracks initially developed from the

mortar layer and extended to the concrete beam. With the mortar layer cracking, the internal

CFRP sheet was exposed bridging adjacent mortar blocks (Fig. 11b). When loaded further, due

to the tensile load transferred from the CFRP sheet, the mortar layer either debonded from the

concrete beam or cracked due to the same mechanism as shown in Fig. 10(b). The beam finally

failed by concrete crushing on the top. As for the beam strengthened with the CFRP-ECC hybrid,

the strengthening layer started pealing off from the concrete beam at both ends (Fig. 11c) when

the load was around 15 kN. This debonding was also indicated in the load-displacement curve in

Fig. 12 (see the circled region). With the increasing flexural load, the debonding horizontally

extended from the end of the hybrid layer toward the mid-span of the beam. At the same time,

flexural cracks were observed in the concrete beam. Due to debonding, the load could not be

effectively transferred to the CFRP-ECC hybrid layer and therefore no cracking damage was

13
observed in this layer. Finally, the beam failed by the concrete crushing on the top, similar to the

other two beams. At the end of the four-point bending test, the CFRP-ECC layer almost detached

from the concrete beam. The failure modes indicated that the interface between ECC and the

concrete may become a problem when this CFRP-ECC hybrid is used for the strengthening of

the concrete structures.

The load-displacement curves of the three beams are plotted in Fig. 12. It can be seen

that, the trends of the three load-displacement curves are very similar, with the load linearly

increased with the displacement before a plateau was reached due to the yielding of the steel

rebar. It can be seen in Fig. 12, the stiffness (slope of the linear stage) of the strengthened beams

was higher than that of the reference beam (7.3×103 kN/m). This is mainly because of the

increased cross section area after strengthening, i.e. the height of the beam increased from 152

mm (6 in) to 178 mm (7 in) with the strengthening layer. However, this increase in stiffness is

marginal for CFRP-mortar hybrid (8.6×103 kN/m). This is because the mortar cracked very early

under flexural loading and the integrity of the increased cross section was compromised (see Fig.

11b). It is obvious that the initial stiffness of the beam with CFRP-ECC hybrid is much higher

than that of the reference beam, i.e. 25.5×103 kN/m, 2.5 times higher when the load was lower

than 15 kN (see Fig. 12). Unfortunately, the CFRP-ECC hybrid strengthening layer started

debonding at both ends in an early stage (see the circle notation in Fig. 12), and therefore the

average stiffness of the beam decreased to 8.4×103 kN/m, which is similar to that of the CFRP-

mortar strengthened beam. Since both strengthening systems experienced problems (i.e. cracking

of mortar and debonding of ECC), they hardly contributed to the load carrying capacity of the

concrete beam. For example, the ultimate load was 59.6 kN for the CFRP-mortar strengthened

beam and 61.3 kN for the CFRP-ECC strengthened beam. They were very similar to the ultimate

14
load of the reference beam, which was 60.3 kN.

3.4 Theoretical discussion on the strengthening potential of CFRP-ECC hybrid

It seems in Fig. 12 that both strengthening systems barely improved the flexural performance of

the concrete beam. However, when carefully examined, the underlying mechanism which caused

such reduced strengthening effect was totally different between the two strengthening systems.

For ECC-mortar hybrid, the mortar was subjected to tensile cracking and fractured into a serious

of mortar blocks. The blocks were connected by the internal CFRP sheet which was in tension.

When the tensile load was transferred from CFRP sheet to the mortar, similar shear cracking

occurred as observed in the pull-out tests (see Fig. 10b). Such a problem with CFRP-mortar

hybrid strengthening system could not be solved without changing the material characteristics.

This is also the reason why the mortar was replaced with ECC in the first place with desirable

multi-cracking and strain hardening behaviors.

On the other hand, the strengthening effect of the CFRP-ECC hybrid was mainly limited

by the debonding of the system from the concrete surface. The CFRP-ECC hybrid layer was

carefully examined after being removed from the concrete beam. It was found that this hybrid

layer was almost intact with no flexural cracks. This CFRP-ECC hybrid layer was also tested

under the same four-point bending setup as the concrete beam. The load-displacement curve and

the final failure mode are presented in Fig. 13. Though the hybrid layer was very thin (only 25

mm), the load still kept increasing with the displacement. This was attributed to the strain

hardening behavior of ECC as well as the geometric nonlinear behavior of the hybrid. The

CFRP-ECC hybrid layer showed extensive flexural curvature with a series of microcracks.

The results in Fig. 13 suggest that the strengthening potential of this CFRP-ECC hybrid

15
can be realized as long as the interface debonding problem can be solved. Similar debonding

issue was also encountered in [35], and a method to improve the interface between the CFRP-

ECC hybrid and the concrete beam is proposed in Fig. 14. Instead of the surface treatment as in

Fig 6(a), the concrete beam can be prepared with the bottom surface mounted with a series of

shear connectors.

With the careful arrangement of the shear connectors, the debonding problem can be

solved, and the CFRP-ECC hybrid can more effectively engage in the load carrying mechanism

of the concrete beam. Then the question is how much improvement in the moment capacity of

the concrete beam can be achieved, considering the contribution of the CFRP-ECC hybrid layer.

The moment capacity of the beam with CFRP-ECC hybrid strengthening (Mmax) can be

expressed by the following equation:

௛
‫ܯ‬௠௔௫ = ‫ܯ‬௥௘௙ + ‫ܯ‬௛௬௕௥௜ௗ + ܲ௛௬௕௥௜ௗ × ଶ (2)

ிೝ೐೑
‫ܯ‬௥௘௙ = ×݀ (3)
ଶ

ி೓೤್ೝ೔೏
‫ܯ‬௛௬௕௥௜ௗ = ×݀ (4)
ଶ

where Mref is the moment capacity of the reference beam; Mhybrid is the corresponding moment

capacity of the CFRP-ECC hybrid layer at the same displacement of the reference beam; Phybrid is

the pull-out force of the CFRP, and h (178 mm see Fig. 4) is the height of the strengthened beam.

It is assumed that the neutral axis is at the middle of the strengthened section; d is the distance

from the load to the closest support of the beam (d=533 mm see Fig. 4).

Fref can be found in Fig. 12 which is 60.3 kN with a corresponding displacement of 31.6

mm. In Fig. 13, Fhybrid is 1.4 kN corresponding to the same displacement as the reference beam.

When calculating the pull-out force Phybrid, it is assumed that the effective bond length of CFRP

in ECC is 170 mm (see Fig. 9a), then Phybrid is equal to the maximum pull-out stress 174.2 MPa

16
multipled by the CFRP section area of 100×0.55 mm2. Therefore Phybrid is equal to 9.6 kN.

According to the Eq. 2, the ultimate moment capacity of the concrete beam with the CFRP-ECC

hybrid strengthening is 17.3 kN.m, showing a 7.6% increase comparing to the reference beam

(16.1 kN.m using Eq. 3).

It should be noted that, this calculated improvement in the moment capacity was based

on the assumption that the effective bond length of CFRP in ECC was 170 mm. As can be seen

in Fig. 9, the effective bond length can be longer and the pull-out stress may be higher than the

value at 170 mm bond length. Therefore, this estimation (7.6%) is conservative. In addition, only

one layer of 0.55 mm thick CFRP sheet was used in the current study. The moment capacity may

be improved further when more layers of CFRP sheets are adopted. The moment capacity of the

concrete beams with the CFRP-ECC hybrid and the shear coupler, and multiple layered CFRP,

will be the subject of a future experimental investigation.

4. Conclusions

In the current study, a CFRP-ECC hybrid was proposed for the strengthening of the concrete

structures. A comprehensive experimental program was conducted which consisted of three

steps. Firstly, ECC with a strain capacity of 3% was developed and the compressive and tensile

properties were calibrated. Then the interface between CFRP and ECC was investigated through

the direct pull-out tests. Finally, concrete beams were prepared and strengthened with the CFRP-

ECC hybrid. Four-point bending tests were carried out to study the performance of the

strengthening system. Based on the results in this paper, it could be concluded that:

(a) The tensile strain capacity of ECC in the current paper was tested as 2.8% (average), which

was almost 252 times higher than that of the cement mortar.

17
(b) In terms of the interface behavior, for the CFRP-ECC hybrid, the peak pull-out stress was

proportional to the CFRP embedded length. The peak stress increased from 71.6 MPa to

174.2 MPa when the embedded length increased from 25 mm to 170 mm. On the other hand,

as for the CFRP-mortar hybrid, the peak pull-out stress stopped increasing when the

embedded length exceeded 100 mm. With 25 mm embedded length, the pull-out stress was

37.2 MPa. After 100 mm, the pull-out stress became constant in the range of 133.5 MPa to

154.3 MPa.

(c) Through the pull-out tests, it was found that, the effective bond length of CFRP in ECC was

longer than 170 mm. As for mortar, the effective bond length of CFRP was in the range of

100 mm to 135 mm.

(d) In the four-point bending tests of the concrete beams, it was found that the CFRP-mortar

hybrid failed by the brittle cracking and debonding from the concrete beam. On the other

hand, the CFRP-ECC hybrid started pealing off from the concrete beam at an early stage.

Due to these premature failure mechanisms, the flexural behavior of the concrete beam was

barely improved by either hybrid system.

(e) Theoretical analysis indicated that, the moment capacity of the concrete beam with the

CFRP-ECC hybrid strengthening could increase by more than 7.6%, if the interface between

ECC and concrete could be secured.

From the theoretical analysis and the flexural testing of the CFRP-ECC hybrid alone

(see Fig. 13), CFRP-ECC hybrid exhibited a good potential for the strengthening of the concrete

structures. This paper pointed out for the first time the interface issue between ECC and concrete,

which provides precautions for the future application of this hybrid for the strengthening of the

concrete structures. The results indicated that the direct casting of the fresh ECC to the concrete

18
surface was not sufficient to ensure the bonding of the two, even though the surface of the

concrete was prepared with surface treatments beforehand. Therefore, it was highly

recommended that shear keys be designed and installed on the concrete surface before attaching

the CFRP-ECC hybrid layer. This would effectively secure the bond of ECC and concrete for not

only room temperature applications, but also for conditions concerning elevated temperatures.

Other techniques existing in the literature to postpone or totally prevent debonding of FRP sheets

from concrete substrate, i.e. grooving method (GM) governed through EBROG [45, 46] and

EBRIG [47] techniques, can also be borrowed and tested for the CFRP-ECC hybrid for the

strengthening of the concrete beams.

It is recommended that more four-point bending tests should be conducted to generate

useful information on the FRP-ECC hybrid for the beam strengthening. Unfortunately, this could

be only possible when the issue of the interface debonding between the CFRP-ECC hybrid and

the concrete can be solved. This can be achieved by focusing the next stage research on the

interface treatment methods, such as grooving method (GM) governed through EBROG and

EBRIG techniques. After existing interface treatment methods were proved, or other innovative

interface treatment methods were developed, can the CFRP-ECC hybrid be used for the

strengthening of the concrete beams.

Acknowledgments

The authors gratefully acknowledge the financial support provided by the Australian Ministry of

Education through the Endeavour Research Fellowship scheme (BR14-003378/44882015) and

the Thousand Talents Plan (Young Professionals) in China through the Fundamental Research

Funds for the Central Universities (YWF-16-RSC-013). This work was also partially funded by

19
the National Science foundation of China (51608020). Dr. Sukmin Kwon and Mr. Motohiro

Ohno are acknowledged for their assistance with the experimental tests.

References

[1] Teng JG, Chen JF, Smith ST, Lam L. FRP strengthened RC structures. John Wiley and Sons
Ltd, West Sussex, UK, 2002.
[2] Chen Y, Davalos J, Ray I, Kim HY. Accelerated aging tests for evaluations of durability
performance of FRP reinforcing bars for concrete structures. Composite Structures 2007;
78(1): 101-111.
[3] Abdalla JA, Harib F, Hawileh R, Mirghani A. Experimental investigation of bond slip
behavior of aluminum plates adhesively bonded to concrete. Journal of Adhesion Science
and Technology 2016; 30(25): 82-99.
[4] Ali AL, Abdalla JA, Hawileh R, Galal K. CFRP mechanical anchorage for externally
strengthened RC beams under flexure. Physics Procedia 2014; 55: 10-16.
[5] Larbi AS, Agbossou A, Hamelin P. Experimental and numerical investigations about textile-
reinforced concrete and hybrid solutions for repairing and/or strengthening reinforced
concrete beams. Composite Structures 2013; 99: 152-162.
[6] Michels J, Czaderski C, El-Hacha R, Brönnimann R, Motavalli M. Temporary bond strength
of partly cured epoxy adhesive for anchoring prestressed CFRP strips on concrete.
Composite Structures 2012; 94(9): 2667-2676.
[7] Hawileh R, Rasheed H, Abdalla JA, Tamimi A. Behavior of reinforced concrete beams
strengthened with externally bonded hybrid fiber reinforced polymer systems. Materials &
Design 2014; 53: 972-982.
[8] Hollaway LC, Leeming MB. Strengthening of reinforced concrete structures: Using
externally-bonded FRP composites in structural and civil engineering. Woodhead
Publishing, Cambridge, UK, 1999.
[9] Naser M, Hawileh R, Abdalla JA, AlTamimi A. Bond behavior of CFRP cured laminates:
experimental and numerical investigation. Journal of Engineering Materials and Technology
2012; 134(2): 021002.1021002.9.
[10] Kalfat R, Al-Mahaidi R. Development of a hybrid anchor to improve the bond performance
of multiple plies of FRP laminates bonded to concrete. Construction and Building Materials
2015; 94: 280-289.

20
[11] Tamimi A, Hawileh R, Abdalla JA, Rasheed H. Effects of ratio of CFRP plate length to
shear span and end anchorage on flexural behavior of SCC R/C Beams. Journal of
Composites for Construction 2011; 15(6): 908-919.
[12] Gamage J, Al-Mahaidi R, Wong MB. Bond characteristics of CFRP plated concrete
members under elevated temperatures. Composite Structures 2006; 75(1): 199-205.
[13] Michels J, Widmann R, Czaderski C, Allahvirdizadeh R, Motavalli M. Glass transition
evaluation of commercially available epoxy resins used for civil engineering applications.
Composites Part B: Engineering 2015; 77: 484-493.
[14] Dong K, Hu K. Development of bond strength model for CFRP-to-concrete joints at high
temperatures. Composites Part B: Engineering 2016; 95: 264-271.
[15] Bai Y, Keller T. Effects of thermal loading history on structural adhesive modulus across
glass transition. Construction and Building Materials 2011; 25(4): 2162-2168.
[16] Al-Safy R, Al-Mahaidi R, Simon GP. A study of the use of high functionality-based resin
for bonding between CFRP and concrete under harsh environmental conditions. Composite
Structures 2013; 95(1): 295-306
[17] Gamage K, Al-Mahaidi R, Wong B. FE modelling of CFRP-concrete interface subjected to
cyclic temperature, humidity and mechanical stress. Composite Structures 2010; 92(4): 826-
834.
[18] Moussa O, Vassilopoulos AP, Keller T. Experimental DSC-based method to determine glass
transition temperature during curing of structural adhesives. Construction and Building
Materials 2012; 28(1): 263-268.
[19] Petkova D, Donchev T, Wen J. Experimental study of the performance of CFRP
strengthened small scale beams after heating to high temperatures. Construction and
Building Materials 2014; 68: 55-61.
[20] López C, Firmo JP, Correia JR, Tiago C. Fire protection systems for reinforced concrete
slabs strengthened with CFRP laminates. Construction and Building Materials 2013; 47:
324-333.
[21] Hashemi S, Al-Mahaidi R. Flexural performance of CFRP textile-retrofitted RC beams
using cement-based adhesives at high temperature. Construction and Building Materials
2012; 28(1): 791-797.
[22] Hashemi S, Al-Mahaidi R. Experimental and finite element analysis of flexural behavior of
FRP-strengthened RC beams using cement-based adhesives. Construction and Building
Materials 2012; 26(1): 268-273.
[23] Tetta ZC, Koutas LN, Bournas DA. Shear strengthening of full-scale RC T-beams using
textile-reinforced mortar and textile-based anchors. Composites Part B: Engineering 2016;
95: 225-239.

21
[24] D’Ambrisi A, Focacci F. Flexural strengthening of RC beams with cement-based
composites. Journal of Composites for Construction 2011; 15(5): 707-720.
[25] Elsanadedy HM, Almusallam TH, Alsayed SH, Al-Salloum YA. Flexural strengthening of
RC beams using textile reinforced mortar-Experimental and numerical study. Composite
Structures 2013; 97: 40-55.
[26] Triantafillou TC, Papanicolaou CG. Shear strengthening of reinforced concrete members
with textile reinforced mortar (TRM) jackets. Materials and Structures 2006; 39(285): 93-
103.
[27] Tzoura E, Triantafillou TC. Shear strengthening of reinforced concrete T-beams under
cyclic loading with TRM or FRP jackets. Materials and Structures 2016; 49(1): 17-28.
[28] Li VC. On engineered cementitious composites (ECC). Journal of Advanced Concrete
Technology 2003; 1(3): 215-230.
[29] Maalej M, Leong KS. Engineered cementitious composites for effective FRP-strengthening
of RC beams. Composites Science and Technology 2005; 65(7-8): 1120-1128.
[30] Li VC. Tailoring ECC for special attributes: A review. International Journal of Concrete
Structures and Materials 2012; 6(3): 135-144.
[31] Şahmaran M, Li VC. Durability of mechanically loaded engineered cementitious composites
under highly alkaline environments. Cement and Concrete Composites 2008; 30(2): 72-81.
[32] Yu JT, Lin JH, Zhang ZG, Li VC. Mechanical performance of ECC with high-volume fly
ash after sub-elevated temperatures. Construction and Building Materials 2015; 99: 82-89.
[33] Zheng YZ, Wang WW, Brigham JC. Flexural behaviour of reinforced concrete beams
strengthened with a composite reinforcement layer: BFRP grid and ECC. Construction and
Building Materials 2016; 115: 424-437.
[34] Dai JG, Wang B, Xu SL. Textile reinforced engineered cementitious composites (TR-ECC)
overlays for the strengthening of RC beams. Proceedings of the second Asia-Pacific
Conferenceon Fiber Reinforced Polymers in Structures. Seoul, Korea, 2009.
[35] Wu C, Li VC. Thermal-mechanical behaviors of CFRP-ECC hybrid under elevated
temperatures. Composites Part B Engineering, doi:10.1016/j.compositesb.2016.11.037,
2016.
[36] Kim SW, Yun HD. Crack-damage mitigation and flexural behavior of flexure-dominant
reinforced concrete beams repaired with strain-hardening cement-based composite.
Composites Part B: Engineering 2011; 42(4): 645-656.
[37] Şahmaran M, Li VC. De-icing salt scaling resistance of mechanically loaded engineered
cementitious composites. Cement and Concrete Research 2007; 37(7): 1035-1046.

22
[38] ASTM C190/C190M-16a. Standard test method for compressive strength of hydraulic
cement mortars (using 2-in. or [50-mm] cube specimens). ASTM International, West
Conshohocken, PA, 2016.
[39] Felekoglu B, Tosun-Felekoglu K, Ranade R, Zhang Q, Li VC. Influence of matrix
flowability, fiber mixing procedure, and curing conditions on the mechanical performance
of HTPP-ECC. Composites Part B: Engineering 2014; 60: 359-370.
[40] JSCE. Recommendations for design and construction of high performance fiberreinforced
cement composites with multiple fine cracks. Japan Society of Civil Engineers, Tokyo,
2008.
[41] ASTM D3039-14. Standard test method for tensile properties of polymer matrix composite
materials. ASTM International, West Conshohocken, PA, 2014.
[42] ASTM C873/C873M. Standard test method for compressive strength of concrete cylinders
cast in place in cylindrical molds. ASTM International, West Conshohocken, PA, 2015.
[43] ASTM A615/A615M-16. Standard Specification for Deformed and Plain Carbon-Steel Bars
for Concrete Reinforcement. ASTM International, West Conshohocken, PA, 2015.
[44] Hosseini A, Mostofinejad D. Effective bond length of FRP-to-concrete adhesively-bonded
joints: Experimental evaluation of existing models. International Journal of Adhesion and
Adhesives 2014; 48: 150-158.
[45] Hosseini A, Mostofinejad D. Effect of groove characteristics on CFRP-to-concrete bond
behavior of EBROG joints: Experimental study using particle image velocimetry (PIV).
Construction and Building Materials 2013; 49: 364-373.
[46] Mostofinejad D, Tabatabaei Kashani A. Experimental study on effect of EBR and EBROG
methods on debonding of FRP sheets used for shear strengthening of RC beams.
Composites Part B: Engineering 2013; 45(1): 1704-1713.
[47] Mostofinejad D, Shameli SM. Externally bonded reinforcement in grooves (EBRIG)
technique to postpone debonding of FRP sheets in strengthened concrete beams.
Construction and Building Materials 2013; 38: 751-758.

23
Figures

Fig. 1. Experimental setup of tensile testing

Fig. 2. Experimental setup of pull-out tests and specimen dimensions (unit in mm, not to scale)

24
Fig. 3. Preparation process of pull-out specimens: (a) surface treatment of concrete block; (b)

CFRP sheet positioning; (c) apply the first layer of ECC and (d) apply the second layer of ECC

Fig. 4. Dimensions of strengthened concrete beam and four-point bending setup (unit in inch)

25
 Fig. 5. Preparation process of the concrete beam with CFRP-ECC hybrid strengthening: (a)

bottom surface treatment of the concrete beam; (b) apply the first layer of ECC; (c) apply CFRP

sheet; (d) apply the second layer of ECC

Fig. 6. Experimental setup and instrumentations of the four-point bending tests

26
 Fig. 7. Tensile stress-strain curves of ECC

Fig. 8. Stress-slip curves of (a) ECC specimens and (b) mortar specimens

27
Fig. 9. Peak pull-out stress versus CFRP embedded length of (a) ECC specimens and (b) mortar

specimens

Fig. 10. Typical failure modes of (a) ECC specimen with CFRP pulled out and (b) mortar

specimen with mortar shear cracking and CFRP pulled out from the rest of mortar matrix

28
Fig. 11. Failure modes of concrete beams: (a) reference beam; (b) beam with CFRP-mortar

hybrid strengthening and (c) beam with CFRP-ECC hybrid strengthening

29
Fig. 12. Load-displacement curves of beams under four-point bending

30
 Fig. 13. Four-point bending of CFRP-ECC hybrid layer

Fig. 14. Proposed method to prevent interface debonding between CFRP-ECC hybrid layer and

concrete

31
Tables

Table 1. Mixture proportion of ECC in the current study (kg/m3)

Water Cement Fly ash Silica PVA Water

W/(C+FA) FA/C
(W) (C) (FA) sand fiber reducer
311 393 865 457 26 5 0.25 2.20

Table 2. Chemical composition and physical properties of fly ash provided by manufacturer

Chemical composition, % Fly ash

CaO 15.66
SiO2 42.20
Al2O3 22.51
Fe2O3 9.20
SO3 1.85
MgO 3.20
Na2O 0.98
K2O 1.53
Moisture 0.12
Loss on ignition 1.34
Physical properties
Fineness, % retained on #325 16.58
Water requirement, % 95
Density 2.53

32
Table 3. Properties of PVA fiber [31-33]

Diameter Length Nominal strength Modulus Density Melting point

(µm) (mm) (MPa) (GPa) (kg/m3) (oC)
39 12 1600 42 1300 230

Table 4. Mixture proportion of concrete (kg/m3)

Water Cement Stone Sand

200 476 1105 562

Table 5. Tensile properties of ECC and mortar

First crack strength Tensile strength Tensile strain

Materials
(MPa) (MPa) (%)
Mean 3.66 4.42 2.771
ECC SD 0.27 0.26 0.697
COV 0.07 0.06 0.282
Mean NA 3.54 0.011
Mortar SD NA 0.34 0.002
COV NA 0.13 0.210

33