Construction and Building Materials xxx (xxxx) xxx

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Flexural toughness of sustainable ECC with high-volume substitution of

cement and silica sand
Kâzim Turk a, Moncef L. Nehdi b,⇑
a
Department of Civil Engineering, Inonu University, Malatya, Turkey
b
Department of Civil and Environmental Engineering, Western University, London, Ontario, Canada

h i g h l i g h t s

Effects of high-content FA and LSF on flexural toughness of ECC was explored.

LSF improved early-age cracking strength of ECC & enhanced its flexural strength.

LSF did not reduce flexural toughness up to deflection of L/150 (T150).

Increasing FA/OPC ratio decreased T150 of ECC beams.

RT,150 from ASTM C 1609 better distinguished pre-peak and post-peak behavior.

a r t i c l e i n f o a b s t r a c t

Article history: This study explores the effects of high-content fly ash and limestone filler partial replacement for port-
Received 20 June 2020 land cement and silica sand, respectively on the flexural toughness parameters of engineered cementi-
Received in revised form 12 October 2020 tious composites (ECC). Various groups of mixtures having variable fly ash/portland cement ratio and
Accepted 20 October 2020
different levels of limestone filler were prepared. ASTM C1609, JSCE-SF4 and the Post-Crack Strength
Available online xxxx
method were employed to appraise the flexural toughness parameters of the ECC mixtures at 3, 28
and 90-d. The results show that according to ASTM C1609, JSCE-SF4 and the Post-Crack Strength method,
Keywords:
limestone filler did not significantly affect the flexural toughness, while the flexural toughness of ECC
Engineered cementitious composite
Toughness
beams decreased when the fly ash content increased. Considering deflection capacity, specimens made
Ductility with a FA/OPC ratio of 1.2 without limestone filler achieved higher ductility at all curing ages. Owing
Modulus of rupture to its superior crack resistance and toughness compared to normal concrete, ECC with high fly ash con-
Limestone powder tent and limestone filler could be a sustainable alternative construction material in diverse civil engineer-
Silica sand ing applications. ECC with enhanced ductility compared to normal concrete could offer increased crack
Fly ash resistance, durability and better resilience.
Ó 2020 Elsevier Ltd. All rights reserved.

1. Introduction to global CO2 emissions. The fine silica sand used could also con-
tain breathable crystalline silica, which with long-term exposure
Engineered cementitious composites (ECCs) are typically com- can cause silicosis, a disabling lung disease. Therefore, developing
posed of cement, silica sand, water, fibers and tailored dosages of ECC with less cement dosage and a healthy substitution to fine sil-
chemical admixtures. ECC yields relatively high toughness and ica sand is desirable.
ductility in the range of 6–8% owing to micromechanical principles Some mineral admixtures including ground granulated blast
used in its mixture design. Its cementitious matrix, fibers and furnace slag, fly ash, metakaolin, rice husk ash, and silica fume
interfacial properties have to be designed properly to achieve the can be used as partial replacement for ordinary portland cement
unique engineering properties of ECC [1,2]. Under tension, ECC dis- (OPC) in ECC. Fly ash (FA) is widely used in cementitious compos-
plays strain-hardening via formation of micro-cracking, while the ites owing to its large volume production, low cost, physical filler
crack width remains typically narrow and not exceeding 60 lm. effect and pozzolanic reaction [3]. It enhances the early-age and
However, ECC formulation in cement intensive, which contributes hardened properties of cement-based composites, while reducing
the OPC dosage, thus mitigating the environmental footprint of
⇑ Corresponding author. OPC production [4,5]. Wang [3] tested specimens of ECC with dif-
E-mail address: [email protected] (M.L. Nehdi).
ferent ratios (0 to 1.5) of fly ash/cement and showed that the

https://doi.org/10.1016/j.conbuildmat.2020.121438
0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.

Please cite this article as: Kâzim Turk and M.L. Nehdi, Flexural toughness of sustainable ECC with high-volume substitution of cement and silica sand, Con-
struction and Building Materials, https://doi.org/10.1016/j.conbuildmat.2020.121438
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

Nomenclature

d1 deflection corresponding to first peak load FTF flexural toughness factor

b width of specimens h height of specimens
dfc first crack deflection in ASTM C1609 L length of specimens
dK/150 deflection of span at L/150 LOP limit of proportionality
dLOP first cracking deflection MOR modulus of rupture
dMOR mid-span deflection P load
dP peak force deflection in ASTM C1609 P1 first peak load
dpeak deflection at peak load RT,150 equivalent flexural strength ratio
Epost,m post-peak energy T Toughness
f150 flexural strength at dL/150 T150 flexural toughness up to dL/150
Ffc first crack force in ASTM C1609 TLOP flexural toughness up to dLOP
fLOP flexural strength at LOP TMOR flexural toughness up to dMOR
fMOR flexural strength
FP peak force in ASTM C1609

increase in FA/OPC ratio improved strain capacity. Wang and Li [5] sand (SS) having average grain size of 425 lm and maximum par-
also studied the engineering properties of ECC made with high fly ticle size of 1.18 mm, respectively, were used as aggregate. Their
ash dosage(FA/OPC = 0.1–1.5) and found that the tensile strain of specific gravity was 2.65 and 2.70, respectively. Particle size grada-
ECC increased with larger fly ash content, but the ultimate tensile tion of LSF and SS is illustrated in Fig. 1. The PVA fiber used had
strength did not follow a clear trend. The influence of limestone fil- 1600 MPa in tensile strength, 8 mm in length, elastic modulus
ler (LSF) as an admixture in ECC has also been studied [6,7]. For equal to 40 GPa, 6.5% in maximum elongation, 1300 kg/m3 in den-
instance, Turk and Demirhan [8] explored the mechanical proper- sity and melting temperature of 225 °C, meeting requirements for
ties of ECC having LSF replacement (25%, 50%, 75%, and 100%) for ECC strain-hardening behavior. To control their interfacial bond
silica sand and found that ductility and flexural strength increased properties with the cementitious matrix, the surface of the PVA
and crack width decreased with an increase of LSF content. fibers was coated during manufacturing with a proprietary
Flexural toughness is a key parameter in structural and seismic hydrophobic oiling agent at 1.2% by weight [17]. To enhance the
design. ASTM C1018 [9] and JSCE [10] provisions are commonly ECC mixtures workability, a superplasticizer having specific gravity
adopted to appraise the flexural toughness parameters. Yet, of 1.06 g/cm3 was used. Details of the ECC mixtures having various
because of difficulties in identifying first-cracking in load–deflec- FA/OPC mass ratio (i.e. 1.2, 2.2, 3.2) are provided in Table 2. ECC
tion curves, ASTM C1018 was substituted for by ASTM C1609 mixtures having constant binder content and 0.26 water-to-
[11]. Nataraja [12] posited that the JSCE method was simple and cementitious materials (w/cm) ratio were made using a standard
the kind of deflection measurement method did not influence the Hobart mixer. The superplasticizer dosage was tailored so that
results. Yu et al. [13] also stated that the JSCE method could better the mixture achieves desired workability.
capture the flexural toughness behavior of steel fiber-reinforced To consider the impact of the ECC ingredients with regards to
concrete. Kim et al. [14] indicated that due to some difficulties in CO2 released into the atmosphere, the data reported in Table 3
using the ASTM C1609 standard, the first cracking point should was assumed [18]. The CO2 emission coefficient was calculated
be specified according to ASTM C1018. They also suggested an through the following formula as suggested by [19,20];
additional evaluation point (L/100) to better capture the effect of
fibers. Therefore, in order to better characterize the flexural tough-
EMSCO2 ¼ kf MOR;90 ð1Þ
ness, the present study compares three different approaches in Where EMSCO2 is the mass of CO2 emitted by manufacturing of
assessing the toughness parameters of ECC incorporating LSF and one cubic meter of ECC in kg, fMOR,90 is the flexural strength of
high fly ash content. ECC at 90 days in MPa, k is the CO2 emission coefficient in kg/
This research investigates the influence of LSF as partial replace- m3MPa and the value of k was calculated using the smallest
ment for silica sand and high-volume fly ash as partial substitute squares method in MS Excel. Based on the regression analysis in
for ordinary portland cement on the compressive strength, fracture terms of CO2 emission of the ECC mixtures, the lowest CO2 emis-
toughness and flexural toughness parameters of ECC. Within this sion coefficient (k) was obtained from the ECC mixture
scope, different FA/OPC mass ratios (i.e. 1.2, 2.2 and 3.2) were used FA3.2_LSF0.0, while FA1.2_LSF1.0 had the highest CO2 emission
and their effects were examined. Flexural toughness parameters coefficient. Therefore, the thoughtful choosing of components can
were evaluated according to ASTM C1609 [11], JSCE [10] and the yield the design of sustainable ECC that can produce favorable
Post Crack Strength (PCS) method [15]. Results obtained using EMS.
these different methods were analyzed and compared.
2.2. Specimen preparation
2. Materials and methods
Three control mixtures made with silica sand and having FA/
2.1. Materials and mixture proportions OPC ratio of 1.2, 2.2 and 3.2, respectively were prepared. Moreover,
nine ECC mixtures with 0%, 50% and 100% mass replacement of SS
Commercially available ASTM C150 Type I cement ordinary with LSF were made. The FA/OPC ratio for each mixture was varied
portland cement (OPC) was used. Class-C fly ash (FA) meeting (1.2, 2.2 and 3.2). All ECC mixture ingredients were maintained
ASTM C 618 [16] requirements was employed for partial cement constant with exception of the LSF and SS contents and the FA/
replacement. The physical and chemical properties of the FA and OPC ratio. A 20-liter capacity high-shear mixer was used to make
OPC are listed in Table 1. Limestone filler (LSF) and micro-silica the ECC mixtures. The OPC, LSF, FA and SS were first dry mixed
2
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

Table 1
Chemical and physical characteristics of FA and ECC.

Oxide PC 42.5(%) Fly Ash(%) Phase PC 42.5(%) Fly Ash(%) Physical Property PC 42.5 Fly Ash
CaO 61.50 14.92 C3S 55 – Loss of Ignition 1.9 0.80
SiO2 19.60 41.76 C2S 15 – Insoluble Residue 0.44 –
Al2O3 4.80 22.91 C3A 7 – SiO2 + Al2O3 + Fe2O3 27.70 73.90
Fe2O3 3.30 9.23 C4AF 10 – Autoclave expansion, % 0.09 0.00
MgO 3.00 2.95 Specific gravity 3.15 2.43
SO3 3.50 1.60 Surface area (m2/kg) 371 115
K2O + 0.66 Na2O 0.7 2.05 Amount retained on 45 mm, % 3 19

100
SS
90
LSP
80
Cumulative (%) Passing

70

60

50

40

30

20

10

0
0.106 0.15 0.212 0.3 0.425 0.6 0.85 1.18
Particle Size (mm)

Fig. 1. Particle size gradation of limestone filler and silica sand.

Table 2
Mixture properties of ECC and mortar.

Mixture ID Component, (kg/m3) Slump-

PC FA Water w/cma Sand FA/PC HRWR PVA Flow
SS LSF (cm)
FA1.2_LSF0.0 570 685 330 0.26 430 0 1.2 9.0 26 27.1
FA1.2_LSF0.5 570 685 330 0.26 220 218 1.2 10.1 26 27.6
FA1.2_LSF1.0 570 685 330 0.26 0 440 1.2 10.3 26 27.5
FA2.2_LSF0.0 390 860 330 0.26 400 0 2.2 6.5 26 27.0
FA2.2_LSF0.5 390 860 330 0.26 200 202 2.2 7.2 26 27.2
FA2.2_LSF1.0 390 860 330 0.26 0 408 2.2 7.3 26 27.1
FA3.2_LSF0.0 300 955 330 0.26 385 0 3.2 6.5 26 27.2
FA3.2_LSF0.5 300 955 330 0.26 195 195 3.2 6.6 26 27.9
FA3.2_LSF1.0 300 955 330 0.26 0 390 3.2 6.7 26 27.7
a
cm: binder (cement + fly ash).

Table 3 included in the final stage and mixing resumed an additional 3 min
The environmental impact of the components of ECC [18] at 150 rpm for. Specimens were prepared by direct casting into
Components Unit Global warming potential(GWP) CO2 molds and compaction using a vibrating table for two minutes.
Cement Type I kg 0.832 Specimens were removed from melds after 24 h and stored at
Limestone powder kg 0.0191 25 ± 2 °C for 7-d in airtight plastic bags without moisture supply.
Fly Ash kg – Specimens were then maintained in laboratory conditions at rela-
Aggregates kg 0.00246 tive humidity of 50 ± 5% and Temperature of 25 ± 2 °C until testing
Water kg 0.000318
at 3, 28 and 90 days.
HRWR kg 0.720
PVA kg 2.06
2.3. Testing methodology

for one minute at 100 rpm. Water and the superplasticizer were To evaluate the ECC mixtures flexural strength, three
then added and mixing at 150 rpm resumed for two minutes. Sub- 100x100x380 mm prismatic specimens were prepared for each
sequently, mixing continued for 3 min at 300 rpm. PVA fibers were mixture and testing age. ECC prisms were subjected to four-point
3
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

bending using an MTS testing machine with tri-axial cell capability PL

f ¼ 2
ð2Þ
with a load rate of 0.003 mm/s (Fig. 2). The span length of ECC bh
specimens subjected to flexural loading was arranged as 324 mm
where f represents the flexural strength (MPa), P denoted the
with a 108 mm central span length according to guidelines of the
load corresponding to specific deflection point (N), L, b and h are
ASTM C1018, ASTM C1609 and PCS testing methods. The applied
the length, width and height of specimens (mm), respectively.
force and mid-span deflection were recorded throughout the flex-
Toughness values were obtained from the load–deflection curve
ural strength tests using a computerized data recording system.
area up to the specified deflection point and named as T (N-m).
The evaluation of flexural toughness test results was made as spec-
As per ASTM C1609, (RT,150) which is the equivalent flexural
ified in ASTM C1018, ASTM C1609 and PCS.
strength ratio is calculated per Eq. (3):
2.3.1. Flexural performance of ECC beam specimens 150T 150
RT; 150 ¼ 2
ð3Þ
Toughness is a key parameter in appraising the contribution of f LOP bh
fibers to the post-peak behavior of cement-based materials. The
flexural toughness parameters for the various ECC mixtures were where T150 is the area under the load–deflection curve up to the
evaluated as per the provisions of ASTM C1609 [11], JSCE [10], as L/150 (2 mm) of the span, fLOP is the flexural strength at LOP.
well as the PCS method [15]. According to the ASTM C1609 proce- According to the JSCE Method [10], the flexural toughness factor
dure (Fig. 3), d1 is the deflection associated with first peak load (P1) (FTF) was calculated as:
 
and dL/150 is the deflection of span at L/150 (mm). Although ASTM T 150 L
C1609 uses the first peak point, in this study, first cracking is des- FTFJSCE ¼ ð4Þ
dL=150 bh2
ignated as a point where the curve gets clearly nonlinear. The use
of first peak point is inadequate for materials characterized by where L is the length of span (mm), dL/150 is the deflection of
deflection hardening behavior and is hard to identify, as indicated span at L/150 (mm), b and h represent the width and depth of spec-
in previous studies [14,21]. This point is defined in ASTM C1018 [9] imens (mm), and T150 is the flexural toughness up to dL/150 (N.mm).
as the limit of proportionality (LOP). The point at which the soften- The load–deflection curves were further analyzed according to the
ing part of the curve starts after LOP is recognized as the modulus PCS method as reported in [22,23]. The curve was separated into
of rupture (MOR). Therefore, residual strength values correspond- post-peak and pre-peak parts from the peak point as shown in
ing to deflection of 2 mm was calculated in this study. Flexural Fig. 4. Then, the area under the curve was obtained from the origin
strength at LOP, MOR and L/150 were calculated by Eq. (2): to the peak load and was named as pre-peak energy. The curve for
the post-peak portion was considered till deflection value L/m
(where L is the length of the beam and m was taken as 150) and
the area under the curve from the peak point to deflection of L/m
was designated as post-peak energy (Epost,m). For a beam having
depth d and width b, PCS at deflection of L/m was calculated as:
 
Epost;m l
FTFPCS ¼
ð5Þ
l
m
 dpeak bh 2

3. Experimental results and discussion

3.1. Flexural load versus deflection behavior

Figs. 5-7 show the 3, 28 and 90-d load–deflection curves for ECC
beam specimens incorporating high-content fly ash and various
levels of limestone filler used as partial substitution for silica sand.
ECC specimens at various test ages achieved a ductile plastic
behavior and first peak strength values were higher than that of
Fig. 2. Experimental setup for flexural toughness evaluation. the peak strength. Due to difficulties in detecting the first peak

peak force
Fp
Ffc
Fp

first crack force

Load

Load
peak force deflection
first crack deflection

pre-peak toughness

post-peak toughness

0
L/150 0
fc p
p
Mid-span Deflection Mid-span Deflection

Fig. 3. Parameters in ASTM C1609. Fig. 4. Post-Crack Strength Analysis.

4
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

30 30

25 25

20 20

Load (kN)
Load (kN)
15 15

10 10

5 5

0 0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Deflecon (mm) Deflecon (mm)

LSF_0 LSF_0.5 LSF_1 LSF_0 LSF_0.5 LSF_1

(a) (b)

25

20
Load (kN)

15

10

0
0 1 2 3 4 5 6 7
Deflecon (mm)
LSF_0 LSF_0.5 LSF_1

(c)
Fig. 5. Load-deflection curves for ECC beam specimens at 3-d (a) FA/OPC = 1.2, (b) FA/OPC = 2.2, and (c) FA/OPC = 3.2.

point, the first cracking point (LOP) was defined as the point at 3.2. Flexural results at LOP and MOR
which nonlinearity of the curve became evident, as recommended
in Kim et al. [14]. It can be observed in Figs. 5-7 that the original The initial strength at cracking (fLOP) and flexural strength (fMOR)
linear elastic stage of the curves increased with longer age. In the were obtained according to Eq. (2). The first cracking deflection,
specimens with 1.2 FA/OPC ratio, the slope of the descending por- strength and toughness values are shown in Table 4. The mid-
tion was generally higher than that of the other specimens. This span deflection, flexural strength (peak strength) and toughness
indicates that increasing the FA content enhanced the post- (area under load–deflection curve up to peak strength) are provided
cracking toughness of specimens. All ECC beam specimens at 3 days in Table 5. It can be observed in Table 4 that at early age, LSF inclu-
exhibited deflection hardening behavior, while at 90 days, other sion into the mixture as partial replacement for SS enhanced fLOP
ECC specimens showed no multiple cracking behavior unlike typi- likely owing to the formation carbo-aluminates. In contrast, as
cal of conventional ECC. This can be explained by the fact that at the ECC matrix gained stiffness with time and higher maturity,
90 days, the matrix gained more strength and thus, more PVA the initial cracking strength of specimens decreased with increasing
fibers ruptured during crack bridging. Similarly, the centre-span LSF content. The hardness of LSF is low compared to that of SS. Thus,
deflection of ECC beam specimens was higher at 3 days and its large particles were easily broken and crossed by cracks [25] and
decreased as the curing age increased. Because of stronger interfa- thus, it induced a decrease in first cracking strength. In terms of
cial bond between matrix and fiber at later ages likely due to deflection capacity, as the LSF content increased from 0% to 50%,
advanced pozzolanic reactions of fly ash, PVA fibers did not expe- the first cracking deflection (dLOP) generally increased, but
rience pull-out behavior but rather ruptured. Moreover, the ECC decreased in ECC specimens with 100% LSF, excluding specimens
beam specimens mid-span deflection capacity at 90-d decreased having FA/OPC ratio of 1.2 at all curing ages. This implies that spec-
as the proportion of LSF partial replacement for SS increased. LSF imens having 50% replacement of SS with LSF achieved higher first
caused a decrease in fracture toughness of the matrix and frictional cracking deflection capacity. The addition of LSF decreased both
bond between the fibers and ECC matrix. Thus, the deflection toughness of the matrix and fLOP. Moreover, an increase of the FA
capacity also decreased, which is consistent with findings of Turk content decreased the first cracking strength at all curing ages. At
and Demirhan [8,24]. At 90 days, the smallest mid-span deflection early age, FA inclusion had a positive influence on deflection capac-
was 1.53 mm in specimen FA1.2_ LSF1.0, while the highest was ity, while at 90 days, dLOP values of specimens having FA/OPC ratio
2.97 mm for specimen of FA2.2_ LSF0. of 2.2 were higher than that of other specimens.

5
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

35
35
30
30
25
25
Load (kN) 20

Load (kN)
20
15
15
10
10
5
5
0 0
0 1 2 3 4 0 1 2 3 4
Deflecon (mm) Deflecon (mm)
LSF_0 LSF_0.5 LSF_1 LSF_0 LSF_0.5 LSF_1

(a) (b)
35

30

25
Load (kN)

20

15

10

0
0 1 2 3 4
Deflecon (mm)
LSF_0 LSF_0.5 LSF_1

(c)
Fig. 6. Load-deflection curves for ECC beam specimens at 28-d (a) FA/OPC = 1.2, (b) FA/OPC = 2.2, and (c) FA/OPC = 3.2.

As expected, the modulus of rupture was higher at 90-days for 3.3. Flexural toughness of ECC
all ECC beam specimens, while the mid-span deflection of these
specimens was the lowest. As the ratio of FA/OPC increased, the The flexural toughness of fiber-reinforced materials is indicative
flexural strength at all testing ages decreased due to cement dilu- of the strain energy stored in the composite and can be obtained
tion and strength gain delays because of the high FA content through the area under the load–deflection curve. Due to the brit-
(Table 5). Conversely, ECC beam specimens had increased flexural tle behavior of concrete, discrete fibers are added in the cementi-
strength at early-age with increasing LSF content, similar to previ- tious matrix to improve its energy absorption capacity and
ous studies [26,27], which indicates that LSF accelerated the OPC resistance to cracking. Thus, flexural toughness is a central param-
hydration rate. However, at 90 days, LSF replacement for SS eter to quantify the effect of fibers on the post-peak behavior of
decreased the fMOR and toughness values. This is likely because concrete. In this study, three different methods, namely ASTM
LSF caused weaker fiber–matrix frictional interface. Li [28] sug- C1609 [11], JSCE [10] and the PCS method [15] were used to assess
gested that weaker bonding within the matrix hinders the strain- toughness parameters (Table 6).
hardening behavior in ECC. Toughness values were higher at 3 days
and decreased with curing age. This is attributed to the gain in 3.3.1. Flexural toughness evaluation using ASTM C1609
stiffness of ECC with time and decreased deformation. Also, at early The flexural strength (fLOP), flexural toughness (T150) values at
age, PVA fibers generally underwent pull-out behavior enhancing the L/150 span, and the equivalent flexural strength ratio (RT,150)
toughness, but at 90 days, PVA fibers tended to rather rupture at 3, 28 and 90-d for ECC specimens having high-content fly ash
because of stronger interfacial bond between the matrix and fibers, along with different levels of limestone filler as partial substitution
thus compromising toughness. The LSF dosage and curing age for silica sand were calculated according to ASTM C1609 (Fig. 8).
affected the mid-span deflection (dMOR). LSF partial replacement Regardless of the LSF content, the highest flexural toughness values
for SS enhanced early-age deflection capacity. This can be attribu- up to the deflection of span at L/150 were achieved by ECC mix-
ted to better fiber dispersion owing to enhanced rheology of the tures with FA/OPC ratio of 1.2. Moreover, T150 of ECC beams
matrix [29]. However, at 90 days, with advanced pozzolanic reac- decreased with higher FA/OPC ratio at all curing ages. This could
tions, the matrix gained stiffness, and LSF caused a reduction in be ascribed to lower cement content due to high FA replacement
deflection capacity as the failure mode shifted to fiber rupture and PVA fibers rupturing due to stronger bond with the matrix.
rather than pull-out. According to Eq. (3), the evaluation of RT,150 depended on the
6
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

35 35
30 30
25 25
Load (kN)

Load (kN)
20 20
15 15
10 10
5 5
0 0
0 2 4 0 1 2 3 4 5
Deflecon (mm) Deflecon (mm)
LSF_0 LSF_0.5 LSF_1 LSF_0 LSF_0.5 LSF_1

(a) (b)

35
30
25
Load (kN)

20
15
10
5
0
0 2 4
Deflecon (mm)
LSF_0 LSF_0.5 LSF_1

(c)
Fig. 7. Load-deflection curves for ECC beam specimens at 90-d (a) FA/OPC = 1.2, (b) FA/OPC = 2.2, and (c) FA/OPC = 3.2.

Table 4
Results at LOP.

Mix Code dLOP (mm) fLOP (MPa) TLOP (N-M)

3 day 28 day 90 day 3 day 28 day 90 day 3 day 28 day 90 day
FA1.2_LSF0.0 0.14 0.23 0.23 4.69 6.95 7.42 3.39 5.48 5.97
FA1.2_LSF0.5 0.26 0.30 0.28 5.50 6.58 6.87 5.79 7.43 6.53
FA1.2_LSF1.0 0.19 0.22 0.26 5.50 6.64 6.78 4.12 6.56 6.23
FA2.2_LSF0.0 0.24 0.30 0.31 4.54 6.44 6.95 4.47 7.20 7.46
FA2.2_LSF0.5 0.23 0.30 0.36 4.74 6.00 6.37 3.85 6.71 7.43
FA2.2_LSF1.0 0.18 0.26 0.31 4.73 6.19 6.16 2.73 5.88 6.20
FA3.2_LSF0.0 0.26 0.21 0.24 4.31 5.46 5.70 4.84 3.73 5.24
FA3.2_LSF0.5 0.29 0.19 0.25 4.53 4.54 4.99 7.20 3.47 3.76
FA3.2_LSF1.0 0.21 0.12 0.12 4.40 4.50 4.93 4.56 2.92 2.44

flexural strength at LOP (fLOP) and the area beneath the load–deflec- 3.3.2. Flexural toughness evaluation using JSCE and PCS
tion curve up-to the L/150 (2 mm) of the span (T150). The difficulty The flexural toughness values of ECC beam specimens at 3, 28
in determining the first cracking point could be the reason of some and 90-d were also calculated according to the PCS method [15]
inconsistent trends in the results. Due to immaturity in the matrix and JSCE [10]. It can be observed in Table 6 that the PCS and JSCE
at 3-days, it was observed that the inclusion of LSF had variable toughness values for all mixtures were comparable, with PCS val-
effect on toughness. However, at 28 and 90-d, especially in mix- ues being somewhat larger. The similarity of PCS and JSCE values
tures involving FA/OPC ratio of 2.2 and 3.2, RT,150 values increased were more notable for 3-d specimens. As shown in Fig. 9, when
compared to that of mixtures without LSF. It can be observed in the FA content increased, PCS and JSCE toughness values at all cur-
Fig. 8 that ECC with FA/OPC ratio of 3.2 and 50% LSF at 3 days ing ages decreased, regardless of the LSF content.
had lowest RT,150. However, the highest RT,150 value was obtained Moreover, as LSF increased, the PCS and JSCE toughness values
for 28-days ECC with FA/OPC ratio of 3.2 and 100% LSF for 28 days. for 3-d ECC specimens with 1.2 and 2.2 FA/OPC ratio increased,

7
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

Table 5
Results at MOR.

Mix Code dMOR (mm) fMOR (MPa) TMOR (N-M)

3 day 28 day 90 day 3 day 28 day 90 day 3 day 28 day 90 day
FA1.2_LSF0.0 3.02 2.04 2.03 7.76 10.22 10.89 74.30 65.88 64.13
FA1.2_LSF0.5 3.16 1.96 1.65 8.34 9.05 10.54 80.99 54.74 50.53
FA1.2_LSF1.0 2.59 1.42 1.53 8.55 9.65 9.77 69.01 41.38 41.45
FA2.2_LSF0.0 3.49 2.15 2.97 6.68 8.40 9.43 100.93 57.02 85.97
FA2.2_LSF0.5 3.60 2.20 2.15 6.91 7.77 8.96 77.48 55.12 58.59
FA2.2_LSF1.0 4.47 2.26 2.09 7.63 8.46 8.67 104.46 60.66 53.65
FA3.2_LSF0.0 3.18 1.90 2.62 6.02 8.03 8.76 73.71 45.53 63.55
FA3.2_LSF0.5 3.21 1.96 2.15 5.55 6.75 7.98 53.66 40.63 47.52
FA3.2_LSF1.0 3.72 1.83 2.02 6.07 7.28 7.40 70.76 38.69 41.36

Table 6
Flexural toughness parameters according to ASTMC1609, PCS and JSCE.

Mix Code Age ASTM C1609 FTFPCS FTFJSCE

f150 (kN) T150 (Nm) RT,150
FA1.2_LSF0.0 3 7.50 48.44 1.55 7.832 7.848
FA1.2_LSF0.5 7.76 48.93 1.33 8.05 7.93
FA1.2_LSF1.0 8.06 50.65 1.38 8.31 8.20
FA2.2_LSF0.0 5.95 37.73 1.25 6.23 6.11
FA2.2_LSF0.5 6.59 43.19 1.37 7.17 7.00
FA2.2_LSF1.0 6.91 43.31 1.37 7.04 7.02
FA3.2_LSF0.0 5.71 36.58 1.27 6.62 6.39
FA3.2_LSF0.5 5.31 33.70 1.12 5.59 5.46
FA3.2_LSF1.0 5.77 36.02 1.23 5.84 5.84
FA1.2_LSF0.0 28 10.18 64.67 1.40 10.81 10.48
FA1.2_LSF0.5 9.05 55.89 1.27 9.26 9.05
FA1.2_LSF1.0 9.54 61.96 1.40 10.08 10.04
FA2.2_LSF0.0 8.30 52.99 1.23 8.71 7.95
FA2.2_LSF0.5 7.69 49.63 1.24 8.17 7.44
FA2.2_LSF1.0 8.38 52.91 1.28 8.79 7.94
FA3.2_LSF0.0 7.99 49.88 1.37 8.37 7.48
FA3.2_LSF0.5 6.73 42.60 1.41 6.94 6.39
FA3.2_LSF1.0 7.13 47.74 1.59 7.73 7.16
FA1.2_LSF0.0 90 10.18 63.22 1.28 10.45 9.44
FA1.2_LSF0.5 8.96 65.38 1.43 10.65 9.69
FA1.2_LSF1.0 8.32 55.52 1.23 9.18 8.33
FA2.2_LSF0.0 9.10 55.53 1.20 9.22 8.33
FA2.2_LSF0.5 6.97 50.24 1.26 8.54 7.54
FA2.2_LSF1.0 8.12 51.50 1.25 8.67 7.73
FA3.2_LSF0.0 7.76 47.36 1.25 7.78 7.10
FA3.2_LSF0.5 6.95 44.17 1.33 7.24 6.63
FA3.2_LSF1.0 8.68 51.83 1.58 8.34 7.77

80 2.4
FA1.2 FA2.2 FA3.2 FA1.2 FA2.2 FA3.2 FA1.2 FA2.2 FA3.2
70 2.1
60 1.8
50 1.5
T 150 (Nm)

RT,150

40 1.2
30 0.9
20 0.6
10 0.3
0 0.0
LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5
LSF_0
LSF_0.5
LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0
LSF_0.5
LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1

3 days 28 days 90 days

LSF content
Flexural Toughness-T150 (Nm) Equivalent flexural strength rao-RT,150

Fig. 8. T150 and RT,150 values at (a) 3 days, (b) 28 days, (c) 90 days.

8
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

14 14
FA1.2 FA2.2 FA3.2 FA1.2 FA2.
FA2.2 FA3.2 FA1.2 FA2.2 FA3.2
12 12

10 10

FTFJSCE
FTFPCS 8 8

6 6

4 4

2 2

0 0
LSF_0
LSF_0.5
LSF_1
LSF_0
LSF_0.5
LSF_1
LSF_0
LSF_0.5
LSF_1
LSF_0
LSF_0.5
LSF_1
LSF_0

LSF_0

LSF_1
LSF_0

LSF_1
LSF_0
LSF_0.5
LSF_1

LSF_0.5

LSF_0.5

LSF_0.5
LSF_1
LSF_0
LSF_0.5
LSF_1
Axis Title 90 days
3 days 28 days
LSF content
FTF-PCS FTF-JSCE

Fig. 9. FTFPCS and FTFJSCE values at (a) 3 days, (b) 28 days, (c) 90 days.

while values for specimens made with FA/OPC ratio of 3.2 ratio 3, 28 and 90-d and having FA/OPC ratio of 2.2 and 3.2, regardless
decreased compared to corresponding values for the control mix- of the LSF content. Thus, it is important to identify the most
ture without LSF. At 2-d, PCS and JSCE toughness values for all appropriate method for the measurement of flexural toughness
FA/OPC ratios decreased compared to that of specimens without parameters. RT,150 values based on ASTM C1609 depend on both
LSF, except for specimens FA2.2_LSF1.0. Moreover, PCS and JSCE the flexural strength at LOP (fLOP) and the area beneath the load–
90-d toughness of ECC specimens with 1.2 and 3.2 FA/OPC ratio deflection curve up to L/150 (2 mm) of the span (T150). However,
increased compared to that of the control mixture without LSF. It the flexural toughness factor (FTFJSCE) based on JSCE was a linear
can be observed in Fig. 9 that 3-d ECC specimens with 3.2 function of T150 [30]. This implies that FTFJSCE values were calcu-
FA/OPC ratio and 50% LSF achieved lowest PCS and JSCE toughness lated only according to the area under the load–deflection curve
values. Moreover, highest PCS toughness values (10.81 N/mm2 at up to the L/150 (2 mm). Hence, the JSCE method failed to differen-
28 days) were obtained for FA1.2_LSF0.0 specimens, followed by tiate the pre-peak and post-peak behavior in the load–deflection
FA1.2_LSF0.5 specimens (10.65 N/mm2 at 90 days), while curves.
specimens FA1.2_LSF0.0 achieved highest JSCE toughness
(10.48 N/mm2), followed by specimens FA1.2_LSF1.0
3.4. Ductility
(10.04 N/mm2 at 28 days).

Ductility is an important criterion to describe the post-crack

3.3.3. Comparing ASTM C1609 and JSCE toughness behavior of fiber-reinforced cementitious composites in terms of
Comparison of the flexural toughness parameters, which deflection capacity. It is determined through the load–deflection
include the equivalent flexural strength ratio (RT,150) based on curves as the elongation from the point of first crack deflection
ASTM C1609 and the flexural toughness factor (FTFJSCE) based on to the point where ECC specimens cannot withstand further load-
JSCE is shown in Fig. 10. It can be observed that these two flexural ing. As suggested by other researchers [e.g. 31], the ductility index
toughness values had different trends for ECC beam specimens at for fiber-reinforced cementitious composites was calculated by

14 2.1
FA1.2 FA2.2 FA3.2 FA1.2 FA2.2 FA3.2 FA1.2 FA2.2 FA3.2
12 1.8

10 1.5
FTFJSCE

8 1.2
RT,150

6 0.9

4 0.6

2 0.3

0 0
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0

LSF_1
LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

LSF_0.5

3 days 28 days 90 days

LSF content
FTF-JSCE RT,150

Fig. 10. Comparison of FTFJSCE and RT,150 values at (a) 3 days, (b) 28 days, (c) 90 days.

9
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

4. Conclusions

This present study explores the flexural performance of ECC

beams having different fly ash content used as partial replacement
for OPC and different levels of limestone filler as partial replace-
ment for silica sand. Toughness parameters at 3, 28 and 90-d were
assessed as per the ASTM C1609 and JSCE provisions, as well as the
Post Crack Strength Method. Based on the experimental findings,
the following conclusions can be drawn:

LSF inclusion into ECC mixtures as partial replacement for SS

improved the early-age cracking strength of ECC beams and
enhanced its flexural strength. However, at 28 and 90-d, both
the flexural strength and toughness at LOP decreased with
increased LSF.

At all curing ages, the LSF dosage did not significant affect flex-
Fig. 11. The calculation of ductility values by using load–deflection curves. ural toughness values up to a deflection of L/150 (T150), while
the increase in FA/OPC ratio decreased T150 of ECC beams.

FTFPCS and FTFJSCE values decreased as the FA content increased.
However, based on the PCS and JSCE methods, the toughness
dividing the mid-span deflection at MOR by the first crack deflec- parameters were not significantly influenced by the LSF content
tion at LOP, as given in Eq. (6) (see Fig. 11). at all test ages.

The flexural toughness factor (FTFJSCE) based on JSCE and the
equivalent flexural strength ratio (RT,150) based on ASTM C
Ductility index ¼ dMOR =dLOP ð6Þ 1609 were substantially different for ECC beam specimens with
FA/OPC ratio of 2.2 and 3.2, while those for ECC beam specimens
The obtained ductility index values for all ECC mixtures at all with 1.2 FA/OPC ratio showed similar tendency with change in
curing ages are presented in Fig. 12. Trends in this figure empha- LSF content.
size that the ductility of 3-day ECC specimens was highest for all
RT,150 values based on ASTM C 1609 appear to be more conve-
FA/OPC ratios. The maximum ductility index was measured for 3- nient to distinguish the pre-peak and post-peak behavior in
d ECC beam specimens with FA/OPC ratio of 2.2 and 100% LSF load–deflection curves since it depends on both the flexural
content. This was followed by 3-d ECC specimens made with strength at LOP (fLOP) and the area beneath the load–deflection
FA/OPC ratio equal to 1.2 and no LSF, specimens made with FA/ curve up to the L/150.
OPC ratio of 3.2 and 100% LSF, and specimens having FA/OPC ratio
In terms of deflection capacity and ductility, the optimum FA/
of 2.2 and no 50% LSF content, in that order. In terms of ductility, OPC ratio was 3.2 expect for 3-d ECC specimens, and mixtures
it was observed that the optimum FA/OPC ratio was 3.2 except for made with FA/OPC ratio of 1.2 having no LSF had the highest
3-d specimens. It was also observed that at all curing days for deflection capacity at all curing ages.
specimens made with FA/OPC ratio of 1.2, the mixtures having
Since ductility and toughness influence the long-term durability
no LSF had higher ductility. As the LSF content increased, ductility and resiliency of cement-based materials, the results underline
generally decreased. At 90 days, the highest mid-span deflection the importance of relying on 90-d of longer term toughness
value (2.97 mm) was obtained for specimens having FA/OPC ratio measurements rather than shorter term tests that can change
of 2.2 and no LSF. Since ductility and toughness influence the with advanced matrix maturity and shifting from fiber pull-
long-term durability and resiliency of cement-based materials, out to fiber rupture in flexural testing.
these results above underline the importance of relying on 90-d
of longer term toughness measurements rather than shorter term
CRediT authorship contribution statement
tests that can change with advanced matrix maturity and shifting
from fiber pull-out to fiber rupture in flexural testing.
Kâzim Turk: Conceptualization, Methodology, Validation, For-
mal analysis, Investigation, Data curation, Writing - original draft,
Visualization, Project administration. Moncef L. Nehdi: Conceptu-
alization, Methodology, Validation, Formal analysis, Investigation,
25.00
Resources, Writing - review & editing, Supervision, Project admin-
FA/C=1.2 FA/C=2.2 FA/C=3.2
20.00
istration, Funding acquisition.
Ductility index

15.00 Declaration of Competing Interest

10.00 The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
5.00
to influence the work reported in this paper.
0.00
LSF_0.5

LSF_0.5

LSF_0.5
LSF_0

LSF_1

LSF_0

LSF_1

LSF_0

LSF_1

Acknowledgments

LSP (%) This research received financial support from Turkey’s Scientific
and Technological Research Council (TÜBITAK),_ Grant Number:
3 days 28 days 90 days
TUBITAK-BIDEB-2219, through the (2012-2) International PDF
Fig. 12. Ductility values at 3 days, 28 days and 90 days. Scholarship Program.
10
Kâzim Turk and M.L. Nehdi Construction and Building Materials xxx (xxxx) xxx

References [15] N. Banthia, J.F. Trottier, Test methods for flexural toughness characterization of
fiber reinforced concrete: some concerns and a proposition, ACI Mater. J. 92
(1995) 48–57. https://doi.org/10.14359/1176.
[1] A.W. Dhawale, Joshi V. P., Engineered Cementitious Composites for Structural
[16] Annu. B. ASTM Stand. (2010) 3–6, https://doi.org/10.1520/C0618.
Applications, in: 2013: pp. 198–205. www.ijaiem.org (accessed June 2, 2020).
[17] V.C. Li, C. Wu, S. Wang, A. Ogawa, T. Saito, Interface tailoring for strain-
[2] T. Li, Victor C., Kanda, INNOVATIONS FORUM: Engineered Cementitious
hardening polyvinyl alcohol-engineered cementitious composite (PVA-ECC),
Composites for Structural Applications | Journal of Materials in Civil
ACI Mater. J. 99 (2002) 463–472. https://doi.org/10.14359/12325.
Engineering | Vol 10, No 2, J. Mater. Civ. Eng. 10 (1998) 66–69. https://
[18] B. Chiaia, A.P. Fantilli, A. Guerini, G. Volpatti, D. Zampini, Eco-mechanical index
ascelibrary.org/doi/10.1061/%28ASCE%290899-1561%281998%2910%3A2%
for structural concrete, Constr. Build. Mater. 67 (2014) 386–392, https://doi.
2866%29 (accessed June 6, 2020).
org/10.1016/j.conbuildmat.2013.12.090.
[3] S. Wang, Micromechanics based matrix design for engineered cementitious
[19] G. Habert, N. Roussel, Study of two concrete mix-design strategies to reach
composites, 2005. http://proquest.umi.com/
carbon mitigation objectives, Cem. Concr. Compos. 31 (6) (2009) 397–402,
pqdweb?did=994237291&Fmt=7&clientId=36305&RQT=309&VName=PQD
https://doi.org/10.1016/j.cemconcomp.2009.04.001.
(accessed June 2, 2020).
[20] A.P. Fantilli, F. Tondolo, B. Chiaia, G. Habert, Designing reinforced concrete
[4] K. Turk, M. Bassurucu, R.E. Bitkin, Workability, strength and flexural toughness
beams containing supplementary cementitious materials, Materials (Basel). 12
properties of hybrid steel fiber reinforced SCC with high-volume fiber, Constr.
(2019). https://doi.org/10.3390/ma12081248.
Build. Mater. 266 (2021) 120944, https://doi.org/10.1016/
[21] J.J. Li, C.J. Wan, J.G. Niu, L.F. Wu, Y.C. Wu, Investigation on flexural toughness
j.conbuildmat.2020.120944.
evaluation method of steel fiber reinforced lightweight aggregate concrete,
[5] S. Wang, V.C. Li, Engineered cementitious composites with high-volume fly
Constr. Build. Mater. 131 (2017) 449–458, https://doi.org/10.1016/
ash, ACI Mater. J. 104 (2007) 233–241. https://doi.org/10.14359/18668.
j.conbuildmat.2016.11.101.
[6] T. Matschei, B. Lothenbach, F.P. Glasser, The role of calcium carbonate in
[22] N. Banthia, S.M. Soleimani, Flexural response of hybrid fiber-reinforced
cement hydration, Cem. Concr. Res. 37 (4) (2007) 551–558, https://doi.org/
cementitious composites, ACI Mater. J. (2005). https://doi.org/10.14359/
10.1016/j.cemconres.2006.10.013.
14800.
[7] N. Voglis, G. Kakali, E. Chaniotakis, S. Tsivilis, Portland-limestone cements.
[23] N. Banthia, M. Sappakittipakorn, Toughness enhancement in steel fiber
Their properties and hydration compared to those of other composite cements,
reinforced concrete through fiber hybridization, Cem. Concr. Res. 37 (9)
Cem. Concr. Compos. 27 (2) (2005) 191–196, https://doi.org/10.1016/j.
(2007) 1366–1372, https://doi.org/10.1016/j.cemconres.2007.05.005.
cemconcomp.2004.02.006.
[24] K. Turk, S. Demirhan, Effect of limestone powder on the rheological,
[8] K. Turk, S. Demirhan, The mechanical properties of engineered cementitious
mechanical and durability properties of ECC, European Journal of
composites containing limestone powder replaced by microsilica sand, Can. J.
Environmental and Civil Engineering 21 (9) (2017) 1151–1170, https://doi.
Civ. Eng. 40 (2) (2013) 151–157, https://doi.org/10.1139/cjce-2012-0281.
org/10.1080/19648189.2016.1150902.
[9] ASTM Stand. 04 (1997) 7, https://doi.org/10.1520/C1018-97.
[25] J. Zhou, S. Qian, M.G. Sierra Beltran, G. Ye, K. van Breugel, V.C. Li, Development
[10] JSCE, Test method for bending strength and bending toughness of steel fiber
of engineered cementitious composites with limestone powder and blast
reinforced concrete. Standard Specification for Concrete Structures, Test
furnace slag, Mater Struct 43 (6) (2010) 803–814, https://doi.org/10.1617/
Methods and Specifications, 2005. https://barchip.com/wp-content/uploads/
s11527-009-9549-0.
2016/08/JSCE-SF4-Method-of-Test-For-Flexural-Strength-and-Flexural-
[26] H. El-Didamony, T. Salem, N. Gabr, T. Mohamed, Limestone as a retarder and
Toughness.pdf.
filler in limestone blended cement, Ceram. - Silikaty. 39 (1995) 15–19.
[11] C.C. Test, T. Drilled, C. Concrete, S.T. Panels, C 1609/C 1609M-05 Standard Test
[27] M. Heikal, H. El-Didamony, M.S. Morsy, Limestone-filled pozzolanic cement,
Method for Flexural Performance of Fiber-Reinforced Concrete (Using Beam
Cem. Concr. Res. 30 (11) (2000) 1827–1834, https://doi.org/10.1016/S0008-
With Third-Point Loading) 1, Astm. i (2005) 1–8. https://doi.org/10.1520/
8846(00)00402-6.
C1609.
[28] V.C. Li, On Engineered Cementitious Composites (ECC), ACT 1 (3) (2003) 215–
[12] M.C. Nataraja, N. Dhang, A.P. Gupta, Toughness characterization of steel fiber-
230, https://doi.org/10.3151/jact.1.215.
reinforced concrete by JSCE approach, Cem. Concr. Res. 30 (4) (2000) 593–597,
[29] P. Soroushian, M. Nagi, J.W. Hsu, Optimization of the use of lightweight
https://doi.org/10.1016/S0008-8846(00)00212-X.
aggregates in carbon fiber reinforced cement, ACI Mater. J. 89 (1992) 267–276.
[13] R. Yu, P. Spiesz, H.J.H. Brouwers, Development of Ultra-High Performance Fibre
https://doi.org/10.14359/2587.
Reinforced Concrete (UHPFRC): Towards an efficient utilization of binders and
[30] J.-Y. Wang, N. Banthia, M.-H. Zhang, Effect of shrinkage reducing admixture on
fibres, Constr. Build. Mater. 79 (2015) 273–282, https://doi.org/10.1016/
flexural behaviors of fiber reinforced cementitious composites, Cem. Concr.
j.conbuildmat.2015.01.050.
Compos. 34 (4) (2012) 443–450, https://doi.org/10.1016/j.
[14] D.J. Kim, A.E. Naaman, S. El-Tawil, Comparative flexural behavior of four fiber
cemconcomp.2011.12.004.
reinforced cementitious composites, Cem. Concr. Compos. 30 (10) (2008) 917–
928, https://doi.org/10.1016/j.cemconcomp.2008.08.002.

11