Properties of Engineered Cementitious Composites with

Raw Sugarcane Bagasse Ash Used as Sand Replacement

Sujata Subedi 1; Gabriel A. Arce, A.M.ASCE 2; Hassan Noorvand, S.M.ASCE 3; Marwa M. Hassan, F.ASCE 4;
Michele Barbato, F.ASCE 5; and Louay N. Mohammad, F.ASCE 6
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

Abstract: This study investigates the effects of raw sugarcane bagasse ash (SCBA) used as a partial or complete replacement for silica sand
in engineered cementitious composites (ECCs). ECC mixtures with five different replacement levels of sand with SCBA were produced
(i.e., 0%, 25%, 50%, 75%, and 100% by volume). The SCBA utilized in this study was comprehensively characterized. Furthermore,
the fresh and hardened properties of the produced ECC materials were thoroughly evaluated. The characterization of SCBA revealed that
raw SCBA consisted mainly of small (i.e., 256 μm average particle size), porous, and irregularly shaped particles with carbon and silica as the
main constituents. Furthermore, the SCBA met the pozzolanic component requirement for Class C pozzolans; however, it did not meet the
minimum strength activity index requirement to be classified as a pozzolan; however, it did not meet the minimum strength activity index
requirement to be classified as a pozzolan. In terms of ECC fresh properties, the incorporation of SCBA produced an important loss in
workability, which was mitigated with increasing dosages of high-range water reducer for increasing amounts of SCBA, which also resulted
in a substantial increase in the air content. In the case of hardened material properties, the incorporation of SCBA as sand replacement
produced a slight decrease (up to 11%) of the compressive strength of ECCs. However, the tensile strength and especially the tensile ductility
of the composites were substantially enhanced (up to 22.3% and 311%, respectively). The tensile strength improvements were credited to the
pozzolanic and/or filler effect of SCBA. On the other hand, enhancements in the tensile ductility were associated to the combined effect of
the reduction of crack-tip matrix toughness (credited to the decrease in aggregate particle size), reduction in ECC cracking strength (due to
increase in air content), and increase in the complementary energy (attributed to the potential decrease in the chemical bond of the fiber/matrix
interface and an enhanced fiber dispersion). The surface resistivity of ECC materials was negatively affected by the addition of SCBA.
Furthermore, the length change of all SCBA-ECC materials, at all ages of curing, was higher in comparison with that of the control mixture,
except for the mixture with 25% replacement at 28 days of curing. DOI: 10.1061/(ASCE)MT.1943-5533.0003892. © 2021 American
Society of Civil Engineers.
Author keywords: Strain-hardening cementitious composites (SHCC); Sugarcane bagasse ash (SCBA); Engineered cementitious
composites (ECC); Polyvinyl alcohol (PVA) fiber; Concrete.

Introduction substantial progress has been achieved to enhance the ductility

and tensile strength of concrete materials through the development
Early deterioration of concrete structures is commonly attributed to of high-performance fiber-reinforced cementitious composites
the inherent brittle nature of concrete and its low tensile strength, (HPFRCCs). In contrast to regular concrete or fiber reinforced
which allows for cracking to occur. Over the last 2 decades, concrete, after first-cracking, HPFRCCs exhibit a pseudo-strain-
hardening (PSH) behavior after first cracking, which enhances the
1
Graduate Assistant, Dept. of Construction Management, Louisiana tensile ductility of these composites (Parra-Montesinos 2005). PSH
State Univ., Baton Rouge, LA 70803. Email: ssubed5@lsu.edu is a process of multiple microcrack formations associated with an
2
Research Assistant Professor, Dept. of Construction Management, increase in load-carrying capacity and substantial amounts of plas-
Louisiana State Univ., Baton Rouge, LA 70803. Email: garcea1@lsu.edu tic deformation (Parra-Montesinos 2005).
3
Graduate Assistant, Dept. of Construction Management, Louisiana
Engineered cementitious composites (ECCs) are an emerging
State Univ., Baton Rouge, LA 70803. Email: hnoorv1@lsu.edu
4
Construction Education Trust Fund Distinguished Professor, Dept. of
class of HPFRCCs designed to exhibit high tensile ductility
Construction Management, Louisiana State Univ., 3116A Patrick F. Taylor, (i.e., usually between 1% and 8% tensile strain) at relatively low
Baton Rouge, LA 70803 (corresponding author). ORCID: https://orcid.org fiber contents (i.e., typically 2% volume fraction) (Li 2008, 2019;
/0000-0001-8087-8232. Email: marwa@lsu.edu Yu et al. 2017, 2020). Due to their low fiber content, ECC materials
5
Professor, Dept. of Civil and Environmental Engineering, Univ. of outperform traditional HPFRCCs in terms of practicality and cost,
California, Davis, CA 95616. ORCID: https://orcid.org/0000-0003-0484 thus making this novel class of concrete materials promising for
-8191. Email: mbarbato@ucdavis.edu both repair and new construction applications. Moreover, the PSH
6
Professor and Engineering Materials Characterization and Research behavior of ECCs is characterized by the development of steady-
Facility Manager, Louisiana Transportation Research Center, 4101 state multiple tight cracks (usually around 60 μm), which in turn
Gourrier Ave., Baton Rouge, LA 70808. ORCID: https://orcid.org/0000
gives rise to a high tensile ductility (Li 2019). Furthermore, autog-
-0003-4802-459X. Email: louaym@ltrc.lsu.edu
Note. This manuscript was submitted on March 28, 2020; approved on enous healing of such tight cracks in ECC material enhances the
February 17, 2021; published online on July 2, 2021. Discussion period durability of these composites (Yang et al. 2009).
open until December 2, 2021; separate discussions must be submitted In cementitious composites, aggregate is used as an economical
for individual papers. This paper is part of the Journal of Materials in Civil filler component, which assists with dimensional stability (Turk
Engineering, © ASCE, ISSN 0899-1561. 2013). Aggregate generally occupies a significant volume fraction

© ASCE 04021231-1 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 of cementitious composites; as such, they notably influence the In contrast, ECC materials require very fine aggregate to exhibit
properties of the cementitious product (e.g., workability, strength, PSH behavior, which is achieved by using manufactured microsilica
and shrinkage). In particular, aggregate in ECC materials can sand, which is expensive and not readily available. Therefore, several
greatly affect fiber dispersion, matrix fracture properties, and researchers have investigated the possibility of implementing locally
matrix/fiber interface; therefore, they are of significant relevance available fine sands in ECC compositions (Arce et al. 2018; Ma et al.
for the performance of ECCs. For instance, the use of coarse ag- 2015; Şahmaran et al. 2009; Noorvand et al. 2019; Wu et al. 2019).
gregate can produce difficulty in the dispersion of fibers, leading to These studies revealed that it is feasible to produce ECC-like mate-
fiber clumping; thus, it can reduce the effective number of fibers at rials by utilizing more conventional types of sands that are coarser
crack planes and negatively affect the tensile performance of the than microsilica sand. However, when implementing these sands,
material (Perdikaris and Romeo 1995). As the maximum particle achieving high tensile ductility becomes more challenging.
size of the aggregate increases, the difficulty in fiber dispersion Based on these considerations, the present study evaluates the
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

becomes more prevalent (De Koker and Van Zijl 2004). use of raw SCBA as sand replacement in ECC due to its fine
Moreover, in ECC mixtures, it is important to use smaller particle size, which more closely resembles that of microsilica
aggregate particles size lower than the average distance between sand. It is worth mentioning that, to the best of the authors’ knowl-
fibers to enhance fiber dispersion and prevent fiber clumping edge, the effects on the ECC properties of using raw SCBA as sand
(De Koker and Van Zijl 2004). In addition, the tortuosity of the replacement have not been previously investigated in the literature.
fracture path of a propagating crack decreases when smaller aggre- The implementation of raw SCBA as sand replacement in ECC
gate particle sizes are used, leading to a reduction in the matrix materials presents also a significant relevance because it could help
fracture toughness (Perdikaris and Romeo 1995). This, in turn, reduce the ECC cost and improve its practicality, mechanical prop-
facilitates crack development and propagation, thus enhancing erties, and greenness. Furthermore, the implementation of SCBA
the PSH behavior of ECC materials (Turk 2013). To guarantee a in ECC materials would valorize an underutilized agricultural
by-product and reduce the environmental impacts of its disposal.
robust PSH behavior of ECC materials, only very small particle
size aggregate is utilized in their composition. Typically, ECCs
are produced using an artificially manufactured microsilica sand, Objective and Scope
with an average particle size of 110 μm (Kim et al. 2007; Wang
and Li 2006; Zhou et al. 2010). The objective of this study is to evaluate the effect of raw SCBA on
Sugarcane bagasse ash (SCBA) is an abundant agricultural the fresh and hardened properties of ECC materials when utilized
by-product of the sugar industry and represents a material with as a partial or complete replacement of microsilica sand. Further-
potential as a replacement of microsilica sand for the production more, this study aims to understand the influence of raw SCBA on
of ECC mixtures. In 2018, the global production of sugarcane the fiber-bridging relation of ECC materials.
was 1.91 billion tons (FAO 2020). In the same year, the total sugar-
cane production in the US was 32 million tons, with Louisiana
and Florida being the biggest producers. In Louisiana alone, it Background
is estimated that approximately 16.1 million tons of sugarcane were
harvested in 2018, which produced about 1.8 million tons of sugar The design fundamentals for ECC were introduced in the early
(American Sugar Cane League 2019). Sugarcane bagasse fiber 1990s (Li 1993; Li and Leung 1992). These criteria are based on
(SBF) is the fibrous residue left after sugar production and amounts micromechanics and fracture mechanics concepts (Li 1992). The
to nearly 40%–45% of the total sugarcane production (Dhengare main requirement for the tensile PSH behavior of ECC to take place
et al. 2015). Nearly 90% of the produced SBF is burned at the sugar is that, upon crack initiation in the cementitious matrix, there
mill to produce energy (Chang 2020). Depending on the burning should exist a suitable fiber-bridging capacity and a subsequent
steady-state flat crack propagation, in which the cracks grow to a
process, SBF yields about 3%–9% of SCBA, which contains high
controlled crack width that is typically 60–100 μm (Li et al. 2002).
amounts of amorphous silica. SCBA has a great potential to be
For this phenomenon to occur, both strength and energy criteria
implemented as a supplementary cementitious material (SCM)
need to be met.
due to the presence of high amount of pozzolanic components.
The strength criterion requires, at any possible crack plane, the
However, previous studies have reported that the raw SCBA as
cracking strength of the cementitious matrix (σfc ) to be lower than
directly collected from sugar mills does not exhibit substantial
the fiber-bridging capacity (σ0 ) of the composite, i.e., (Li 1993) as
pozzolanic activity due to the presence of coarse impurities (i.e., un- follows:
burnt fibers and gravel) and high carbon content (Bahurudeen and
Santhanam 2015; Payá et al. 2002; Subedi et al. 2019). Currently, σfc ≤ σ0 ð1Þ
SCBA in Louisiana has no economic value and produces signifi-
cant landfill and containment problems. where σfc is controlled by the matrix fracture toughness (K m ) and
Different studies have investigated the use of SCBA as a the pre-existing initial flaw size. Failure to meet this criterion would
replacement of fine aggregate in regular concrete (Modani and result in fiber rupture or pullout upon crack initiation (Li 1993).
Vyawahare 2013; Sales and Lima 2010; Sua-Iam and Makul The energy criterion describes the necessary conditions for
2013). These studies revealed that the implementation of SCBA steady-state flat crack propagation as first established by Marshall
in regular concrete is feasible, but that its effect on mechanical per- and Cox (1988) using J-integral analysis, and is described by
formance, practicality, and cost is marginal. In particular, the use Li (2008) as follows:
of SCBA did not significantly enhance the properties of concrete, Z δ Z δ
0 ss
0
J b ≡ σ0 δ 0 − σðδÞdδ ≥ J tip ¼ σss δ ss − σðδÞdδ ð2Þ
i.e., SCBA-admixed concrete exhibited similar or marginally
0 0
higher compressive strength than the control concrete (Modani
and Vyawahare 2013; Sua-Iam and Makul 2013). This is the case where σðδÞ = fiber-bridging relation; J b0
= complementary energy;
because the fine aggregate in ordinary concrete does not have δ 0 = crack opening at σ0 ; J tip = crack-tip matrix toughness; σss =
strong requirements in terms of fineness. steady-state cracking stress; and δ ss = crack opening at σss . Eq. (2)

© ASCE 04021231-2 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Fiber-bridging behavior: (a) schematic fiber-bridging stress (σ) versus crack opening (δ) curve; and (b) schematic tensile stress versus strain
curves of cementitious materials. (Adapted from Noorvand et al. 2019.)

derives from the energy balance between the external work (due specific gravity of 1.3. The chemical compositions of OPC and
to tensile load application), and the energy consumed to break fly ash obtained from X-ray fluorescence analysis are presented
the crack-tip (i.e., J tip ) and open the crack to δ ss (Li 2012). in Table 1.
Fig. 1(a) presents the schematic σðδÞ curve with J b0 and J tip repre-
sented by the areas limited by the dashed and dotted lines, respec- Raw SCBA
tively. J b0 is primarily affected by the matrix/fiber interface
properties, whereas J tip is a property of the cementitious matrix. The SCBA utilized in this study was collected from Alma Planta-
J tip is dependent upon the water-to-binder (W/B) ratio, the content tion, a sugar mill located in Louisiana. For the production of raw
of cement and SCMs such as fly ash, and aggregate type and size SCBA, the collected SCBA was initially dried at 65°C for 10–12 h
(Li 2012). to remove moisture and subsequently sieved using a No. 20 sieve
From Eqs. (1) and (2), the PSH strength (σ0 =σfc ) and PSH en- (i.e., 841-μm-opening sieve) to eliminate impurities (i.e., gravels
ergy (J b0 =J tip ) indexes emerge as indicators of the PSH performance and unburnt fibrous particles, among others). The specific gravity
of ECC materials. The PSH behavior of ECCs is only possible of the processed raw SCBA was 2.32. The raw SCBA used in this
if both the PSH indexes are greater than one. By contrast, if either study was thoroughly characterized by using scanning electron
of the PSH indexes is less than one, then the strain-softening behav- microscopy (SEM), energy dispersive X-ray spectroscopy (EDS),
ior of regular concrete and fiber-reinforced concrete will occur X-ray diffraction (XRD), and laser diffraction particle-size analy-
[Fig. 1(b)]. In addition, considering the nonhomogeneous nature sis, Moreover, because SCBAs can exhibit substantial pozzolanic
of the composites (due to the variability of, e.g., material properties, activity depending on the level of processing, the pozzolanic
initial flaw size, and fiber distribution), Kanda and Li (2006) rec- reactivity was evaluated via the strength activity index (SAI)
ommended that the PSH strength and energy indexes should be method as per ASTM C109/109M (ASTM 2016a).
greater than 1.3 and 2.7, respectively, to ensure robust strain-
hardening behavior. Preparation of ECC Mixtures
To investigate the effect of replacing silica sand with raw SCBA on
Materials and Methods the properties of ECC, a total of five ECC mixtures were prepared.
ECC mixtures were designated as M-X, where X is the percent
replacement by volume of sand with raw SCBA. The dosages of
Materials Used in ECC Mixtures
sand replacement investigated were 0%, 25%, 50%, 75%, and
Type I ordinary portland cement (OPC), Class F fly ash, and silica 100%. In this study, the reference mixture M-0 utilized similar pro-
sand (i.e., fine river sand with a mean particle size of 474 μm and portions as those of the ECC mixture design denoted as M-2.2% in
D90 , 90% passing size of 786 μm) with a specific gravity of 3.15, Noorvand et al. (2019); however, the W/B content was increased
2.29, and 2.62, respectively, were utilized for all ECC mixtures. from 0.27 to 0.32 to enhance workability, and the fiber content was
Furthermore, a polycarboxylate-based high-range water-reducing reduced from 1.75% to 1.5% to produce more cost-effective com-
admixture (HRWR), water, and polyvinyl alcohol (PVA) fibers posites. The fly ash-to-cement ratio (F/C), W/B ratio, and the fiber
were utilized in the production of ECC. This study used RECS-15 content of the M-0 ECC mixture were 2.2 (by weight), 0.32 (by
fibers, which are non-oil-coated PVA fibers and were provided by weight), and 1.5% (by volume), respectively. The F/C ratio, fiber
NYCON (Fairless Hills, Pennsylvania). The RECS-15 fibers have a content, and W/B ratio were kept the same for all ECC materials
Young’s modulus of 40 GPa, maximum elongation of 6%, diameter evaluated in this study. The details of five ECC mixture proportions
of 38 μm, length of 8 mm, tensile strength of 1,600 MPa, and are illustrated in Table 2.

Table 1. Oxide composition of OPC and fly ash

Oxide
Materials CaO SiO2 Al2 O3 Fe2 O3 K2 O Na2 O MgO SO3 CO2
OPC (%) 63.77 19.06 4.55 3.00 0.37 0.00 2.27 3.43 —
Fly ash (%) 7.97 57.94 20.03 3.67 1.24 2.14 2.02 0.49 —

© ASCE 04021231-3 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 Table 2. ECC mixture proportions
OPC Fly ash Water Sand Raw SCBA Raw SCBAa HRWRb Vf c
Mix ID (kg=m3 ) (kg=m3 ) (kg=m3 ) (kg=m3 ) (kg=m3 ) (%) (%) (%)
M-0 358.9 789.5 367.5 416.2 0 0 0 1.5
M-25 358.9 789.5 367.5 312.2 92.2 25 0.13 1.5
M-50 358.9 789.5 367.5 208.1 184.3 50 0.45 1.5
M-75 358.9 789.5 367.5 104.1 276.4 75 0.80 1.5
M-100 358.9 789.5 367.5 0 368.6 100 1.5 1.5
a
Sand replacement by volume (%).
b
HRWR by weight of OPC (%).
c
Fiber content by volume (%).
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

In this study, a planetary mixer was utilized to prepare all ECC to exhibit multiple cracks, each dog-bone-shaped specimen was
mixtures. Initially, powder components (OPC, fly ash, sand, and notched at its middle section around the entire specimen [Fig. 3(a)]
raw SCBA) were mixed for 3 min at low speed (i.e., 60 rpm). Next, to induce the development of the desired single crack. The notch
water and HRWR were slowly added in one minute and mixed for thickness adopted in this study was 1 mm. The SCTT was con-
an additional 3 min at medium speed (i.e., 110 rpm). Subsequently, ducted at 28 days of hydration on four specimens for each mixture
a modified marsh funnel test was used to evaluate the flowability design. The notched specimens were loaded under a displacement-
of the mix. Then, PVA fibers were added in 2 min while mixing at controlled constant axial displacement rate of 0.5 mm=min using
a medium speed. Finally, the material was mixed at high speed the same equipment utilized for the uniaxial tensile test. The crack
(i.e., 200 rpm) for another 5 min. After completion of the mixing opening, δ, was measured by attaching one LVDT at each side of
process, the air content of the ECC materials was measured accord- the specimen, for a total of two LVDTs, as shown in Fig. 3(b). The
ing to ASTM C231/C231M (ASTM 2017c). gauge length utilized in the SCTT was 20 mm. This small gauge
Subsequently, 3 cylinder, 10 dog-bone-shaped (6 for uniaxial length was utilized to minimize the contribution from the elastic
tensile test and 4 for a single-crack tensile test), and 4 prismatic deformation of the cementitious matrix on the crack opening
specimens (for length change measurements) were cast for each measurements.
mixture design. All the cylindrical and dog-bone-shape specimens
were covered with plastic sheet and left in the mixing room for Surface Resistivity
24 h. By contrast, the prismatic specimens were placed in a moist To gain insight into the durability of ECC materials, a surface re-
room for 230.5 h according to ASTM C157/C157M (ASTM sistivity test was performed as per AASHTO T 358 (AASHTO
2017d). Finally, after 24 h, all specimens were demolded and were 2017). Because this test is a nondestructive test, the cylindrical
cured in a lime-saturated water tank for 28 days, as per ASTM specimens utilized for this test were the same utilized for the com-
C511 (ASTM 2019). pressive strength tests. The electrical resistivity was measured at
28 days of curing by using a Wenner four-pin array [with a
Experimental Tests 38-mm (1.5-in.) probe spacing]. The moisture condition of the

Compressive Strength
The compressive strength of all mixtures was assessed as per
ASTM C39/C39M (ASTM 2018) on cylinder specimens measur-
ing 101.6 mm ð4 in:Þ × 203.2 mm (8 in.). A total of three speci-
mens per mixture were tested at 28 days of curing by using
hydraulic pressure.

Uniaxial Tensile Test

The uniaxial tensile test of HPFRCCs based on the recommendations
of the Japan Society of Civil Engineers (JSCE) was adopted in this
study to characterize the tensile performance of all ECC mixtures
(JSCE 2008). At 28 days of curing, six dog-bone-shaped specimens
of 330 × 30 × 13 mm3 (length × width × thickness) were tested for
each mixture. The geometrical details of the specimens are presented
in Fig. 2(a). The test was performed by applying a displacement-
controlled constant axial displacement rate of 0.5 mm=min by
utilizing a servohydraulic machine with a capacity of 250 kN. Fur-
thermore, to evaluate the material deformation, a LVDT was attached
at each side of the specimen (for a total of two LVDTs), as shown in
Fig. 2(b). The gauge length utilized was 80 mm.

Single-Crack Tension Test

In this study, the single-crack tension test (SCTT) was conducted to
evaluate the effects of utilizing raw SCBA as sand replacement on
the fiber-bridging relation, σðδÞ (Pereira et al. 2012; Pereira and
Fig. 2. Uniaxial tensile test: (a) dog-bone-shaped specimen dimensions
Fischer 2010). The main prerequisite of this test is that a single
(millimeters); and (b) uniaxial tensile test setup.
crack needs to develop during the test. Because ECCs are designed

© ASCE 04021231-4 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. SCTT: (a) notched dog-bone-shaped specimen dimensions (millimeters); and (b) SCTT setup.

samples was saturated. For each specimen, two sets of readings CRD − CRD0
ΔLx ¼ · 100 ð3Þ
were taken by placing the Wenner probe in the center of the lon- G
gitudinal side of the specimen at four different locations (i.e., 0°,
90°, 180°, and 270°). For each mixture design, the final resistivity where ΔLx = percent length change of any prism specimen at any
value was obtained by averaging the surface resistivity reading of given curing age; CRD = difference between the reading at any
three cylindrical specimens and multiplying by a factor of 1.1 as a given age of the comparator and reference bar and the specimen;
CRD0 = initial comparator reading; and G ¼ 250 mm (10 in.) =
curing condition correction for lime-saturated water curing.
gauge length.
Dimensional Stability
In order to investigate the dimensional stability of the ECC
mixtures during curing due to the inclusion of SCBA, this study Results and Discussion
measured the length change of all ECC mixtures during the curing
stage by following a procedure similar to that described in ASTM Raw SCBA Characterization
C157/C157M (ASTM 2017d), with the exception that no measure-
ments were taken after the end of the curing phase. Four prism SEM-EDS Analysis
specimens measuring 25.4 × 25.4 × 285 mm were cast for each For the SEM-EDS analysis, a Quanta three-dimensional (3D) Dual
ECC mixture design. After casting, specimens were kept in a moist Beam (Field Electron and Ion Company, Hillsboro, Oregon) field
room (at 23°C  2°C and relative humidity higher than 95%) for emission gun (FEG) focused ion beam (FIB)-scanning electron
23.50.5 h. Next, the specimens were demolded and immersed microscope (SEM) with EDAX Pegasus EDS/electron backscatter
in a lime-saturated water tank for 30 min. Subsequently, the initial diffraction (EBSD) detectors was utilized to investigate the mor-
length comparator dial reading was taken, and the prism specimens phology and chemical composition of raw SCBA. Fig 4(a) shows
were again immersed in the lime water tank. Finally, readings were the raw SCBA material used as a sand replacement in ECC mix-
taken at 7, 14, and 28 days of curing, and the length change at each tures. Backscattered electron (BSE) SEM imaging of raw SCBA,
specific curing age was computed as follows: presented in Fig. 4(b), shows that SCBA consists of a combination

Fig. 4. Characterization of raw SCBA: (a) raw SCBA; and (b) BSE SEM image of raw SCBA.

© ASCE 04021231-5 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 Table 3. Oxide composition of raw SCBA indication of the presence of crystalline silica. The quantitative min-
Oxide Amount in SCBA (%) eralogical composition and amorphous content of silica were deter-
mined by employing the Rietveld refinement technique by using
CaO 3.84
HighScore plus version 4.7 software (Degen et al. 2014) and based
SiO2 42.80
Al2 O3 5.54
on the external k-factor method (O’Connor and Raven 1988).
Fe2 O3 4.24 Furthermore, quartz (SiO2 ) powder was used as an amorphous
K2 O 3.63 standard and was analyzed in the same conditions as those used for
Na2 O 0.14 the raw SCBA. The analysis reported the presence of 52.9% of
MgO 0.56 amorphous silica, 38.9% of quartz, and 5.5% of albite in raw
SO3 0.15 SCBA. The quartz content of raw SCBA may be attributed to the
CO2 38.13 crystallization of amorphous silica during calcination of SBF, and
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

SiO2 þ Al2 O3 þ Fe2 O3 52.58 the presence of soil adhered to the SBF during sugarcane harvest.
Loss on ignition 11.4 The amorphous phase can be attributed to the presence of unburnt
SBF and amorphous silica.

Laser Diffraction Particle-Size Analysis

of particles with various shapes and sizes (i.e., irregular porous par-
The particle-size distribution of raw SCBA and silica sand were
ticles, large fibrous particles, spherical particles, and prismatic
obtained from a Beckman LS200 laser diffraction particle size ana-
particles).
lyzer (Beckman Coulter, Brea, California). The analysis was per-
Due to the uncontrolled burning of SBFs at the sugar mills, high
formed in a microvolume module where samples were suspended
temperature gradients exist in the boilers of cogeneration plants.
in water with a 60-s agitation. The particle-size distribution of the
Thus, some portions of the SBF experiences high temperatures,
raw SCBA and silica sand are presented in Figs. 5(b and c). It is
whereas some portions remain partially unburned. Although the
observed that the silica sand consists of coarser particles in com-
fibrous particles observed in the SEM images were attributed to
parison with those of the raw SCBA. For instance, the mean particle
unburned SBF, the spherical and prismatic particles were likely
size of the raw SCBA was 256 μm, and the D90 (90% passing size)
attributed to crystalline silica formed due to zones of high calcina-
was 539 μm, whereas the silica sand exhibited a mean particle size
tion temperature (usually above 700°C). A previous study has re-
of 474 μm and its D90 was 786 μm.
ported the presence of spherical silica particles in SCBA due to
melting when SBF burning temperature exceeds 600°C–800°C SAI Analysis
(Bahurudeen and Santhanam 2015). Moreover, prismatic crystal- The pozzolanic activity of raw SCBA was assessed by the SAI
line silica has also been reported due to SCBA contamination with method as per ASTM C109/109M (ASTM 2016a), similarly to
soil and crystallization of silica at temperatures beyond 700°C other previous studies available in the literature (Bahurudeen
(Arce et al. 2019; Cordeiro et al. 2009). It is also important to and Santhanam 2015; Cordeiro et al. 2008, 2009, 2016). This test
mention that the observed irregular porous particles were likely was selected to investigate the effects of the high amount of silica in
attributed to amorphous silica. the raw SCBA in order to allow a meaningful comparison with the
For the chemical composition of raw SCBA, EDS spectra were information already available in the literature. The 28-day compres-
collected by using an area mode with a current of 4 pA and a volt- sive strength of 50-mm control mortar cubes (designated as CO
age of 20 kV. The equipment used in the EDS analysis can detect all specimens) prepared with a W/B ratio of 0.48 and a sand-to-cement
elements ranging from Beryllium to Americium. The results of the ratio of 2.75 was compared with that of SCBA admixed mortar
EDS analysis are reported in Table 3 in terms of oxide composition, cubes that contained a 20% replacement by weight of OPC with
as customary for cementitious and pozzolanic materials (Noorvand raw SCBA (designated as BA specimens). Furthermore, to compare
et al. 2019; Wang and Li 2007; Yang et al. 2007), by using the the reactivity of raw SCBA with that of nonreactive silica sand,
element-to-stoichiometric oxide conversion method available in additional specimens (designated as SA specimens) were prepared
the TEAM version 4.4 software utilized for processing of the by using silica sand as a 20% replacement by weight of OPC. The
EDS data. sand utilized for all mortar mixtures was standard graded sand con-
The oxide composition of the raw SCBA indicates a prevalence forming to ASTM C778 (ASTM 2017b). At the end of 28 days of
of silicon dioxide (SiO2 ) and carbon dioxide (CO2 ). Whereas SiO2 hydration, the compressive strength of all mortar cubes was tested
content was indicative of silica content (amorphous and crystal- using six replicas for each mixture as per ASTM C311/C311M
line), the CO2 content was indicative of high carbon content (ASTM 2016b). The SAI value was calculated as follows:
due to unburnt SBF. It is important to mention that the loss on igni-  
tion for raw SCBA was 11.4%, which is consistent with high carbon X
SAI ¼ · 100 ð4Þ
content observed in the EDS analysis. Nevertheless, raw SCBA ex- Y
hibited a total pozzolanic component (SiO2 þ Al2 O3 þ Fe2 O3 ) of
52.6%, thus meeting the ASTM C618 (ASTM 2017a) minimum where X and Y = mean compressive strength of BA or SA cubes
pozzolanic component requirement to be classified as a Class C and control mixture cubes, respectively.
pozzolan (i.e., total pozzolanic component ≥50%). The average compressive strength and corresponding SAI val-
ues calculated per Eq. (4) are presented in Table 4. It is observed
XRD Analysis that the SA mortars exhibited a SAI value of 63.2%, whereas BA
XRD patterns were obtained by using a PANalytical Empyrean mortars exhibited a SAI value of 72.1%. The higher SAI value for
X-ray Diffractometer (Malvern Panalytical, Malvern, UK) equipped BA specimens was indicative of filler action (due to its small par-
with PreFIX (prealigned, fast interchangeable X-ray) modules and ticle size) and/or reactiveness of raw SCBA in contrast to nonreac-
Cuk α radiation at 40 mA and 45 kV. The data for the raw SCBA tive silica sand (crystalline silica). The possibility of the pozzolanic
were collected by scanning 2θ in the range from 5° to 90° with a step activity of SCBA was credited to the presence of amorphous silica
size of 0.026°. Fig. 5(a) presents the XRD pattern of the raw SCBA. (reactive silica) in raw SCBA. It is important to note that the differ-
The presence of the characteristic peaks of quartz in Fig. 5 is an ence in compressive strength between SA and BA specimens was

© ASCE 04021231-6 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 10 10
2 Crystalline Phase:

Intensity (Counts) x 10000

1: Sanidine (2.7%)
8 2: Quartz (38.9%)
8

Frequency (%)
3: Albite (5.5%)
6 4: Calcite (0.1%) 6
Amorphous Phase: 52.9%

4 4

2 2 2
1 3 2
4
0
0 0 1 10 100 1000 10000
0 20 40 60 80
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

Particle Size (μm)

2 (Degrees)
Raw SCBA Silica Sand
(a) (b)

100
Cummulative Passing (%)
80

60

40

20

0
0 1 10 100 1000 10000
Particle Size (μm)
Raw SCBA Silica Sand
(c)

Fig. 5. SCBA and sand characterization: (a) XRD pattern of raw SCBA; (b) raw SCBA and silica sand particle-size distribution frequency; and
(c) cumulative percent passing particle-size distributions.

Table 4. SAI test results for 28-day compressive strength

ID Description Average (MPa) Standard deviation (MPa) Coefficient of variation (%) SAI (%)
CO Control 42.73 4.90 11.48 N/A
BA 20% cement replacement with raw SCBA 30.82 1.03 3.35 72.12
SA 20% cement replacement with sand 27.23 3.84 14.12 63.72

statistically significant (i.e., p-value ¼ 0.03). However, the SAI ex- such, this can be used as a reliable indirect test method to assess the
hibited by the BA mortar (i.e., 72.1%) did not meet the minimum rheological properties (i.e., plastic viscosity and yield stress) of the
SAI requirement of 75% prescribed by ASTM C618 for a material ECC cementitious matrices used in this study.
to be classified as a pozzolan (ASTM 2017a). Even though SCBA In this test, the modified Marsh funnel is initially filled with
exhibited higher reactivity than inert sand and met ASTM C618 ECC mortar before the incorporation of PVA fibers. Then, the bot-
pozzolanic component requirement, it is concluded that raw SCBA tom outlet of the cone is opened to allow the ECC mortar to flow.
cannot be used as a SCM in concrete because its SAI is too low and The Marsh cone flow time, measured in seconds, is the time passed
its carbon content (reflected in the high loss on ignition of 11.4%) is from the opening of the bottom outlet of the cone to the time until
too high, thereby not meeting the corresponding minimum require- the light is visible when observed from the top (Li and Li 2013). A
ments set by ASTM C618. Marsh cone flow time between 24 and 33 s is optimal to ensure
uniform fiber dispersion on ECC utilizing PVA fibers (12 mm
in length) at 2% volume fraction (Li and Li 2013). However,
Fresh ECC Properties the PVA fiber length and content utilized in this study were 8 mm
The rheology of the ECC cementitious matrix in its fresh state is an and 1.5% volume fraction, respectively, both of which help fiber
important property controlling fiber dispersion, which affects the dispersion. Therefore, the aforementioned flow time range is not
capability of forming multiple cracks of small width (Li and Li directly applicable to this study. For this reason, all ECC mixtures
2013). This study reports the results of the modified Marsh funnel produced were inspected by touch to evaluate whether fiber clump
test proposed by Li and Li (2013), which was conducted to gain formation existed in any of the ECC mixtures. No fiber clump for-
insight into the rheology of the mixtures before the addition of mation was detected in any of the ECC materials produced.
PVA fibers. The flow time measured using the modified Marsh fun- Table 5 reports the HRWR dosage, Marsh cone flow time re-
nel test has been shown to strongly correlate (R ¼ 0.95) with the sults, and air content measured for each produced ECC mixture.
plastic viscosity of ECC cementitious matrices (Li and Li 2013). As As shown in Fig. 4(b), raw SCBA particles are very irregular, fine,

© ASCE 04021231-7 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 Table 5. Fresh ECC properties ANOVA was conducted to evaluate if the differences between
ID a
HRWR (%) Flow time (s) Air content (%) the mixtures were statistically significant. Per the ANOVA, statisti-
cally significant differences in mean compressive strength among
M-0 0 7.00 1.40
the five ECC mixtures were encountered at a significance level of
M-25 0.13 15.00 2.90
M-50 0.45 24.00 3.50
0.05 (p-value ¼ 0.0006). In addition, according to the Tukey pair-
M-75 0.80 23.00 5.30 wise comparison, statistically significant differences were encoun-
M-100 1.5 28.00 5.80 tered between control and SCBA-ECC mixtures incorporating
a SCBA contents greater than 50%. However, for the SCBA-ECC
By weight of cement (%).
materials with 50% or less sand replacement with SCBA, the differ-
ences between the SCBA-ECC mixtures and the control mixture
and porous; as such, when replacing silica sand with SCBA, a sub- were not statistically significant. The observed decrease in com-
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

stantial loss in workability is expected. For this reason, the control pressive strength can be associated with the decrease in density
mixture (M-0) was designed to be a highly workable mixture (low- of the ECC materials with SCBA for increasing SBCA contents.
est flow time) to provide sufficient workability at high replacement Fig. 6(b) reports the hardened densities of all the ECC materials.
levels of sand with SCBA and allow for ECC manufacturing to be As expected, due to the lower specific gravity of the SCBA when
possible. Furthermore, to mitigate the drastic loss in workability, compared with the silica sand, the density of the ECCs showed a
HRWR dosage was incremented as the SCBA content increased decreasing trend with the increase in sand replacement with SCBA.
(Table 5). From the Marsh funnel test results presented in Table 5, Furthermore, the increase in air content reported in Table 5 is
it is observed that M-100 exhibited a high flow time even after the also a factor contributing to the measured density reduction. In
addition of HRWR at a dosage of 1.5% by weight of cement. In this general, for each 1% increase in air content, a 5% decrease in com-
study, the HRWR dosage was limited to 1.5% due to its influence pressive strength is expected in regular concrete [ACI 212.3 R-16
on other material properties such as air content. (ACI 2016)]. As indicated in Table 5, the air content in M-25,
Because air content can have an important effect on the hard- M-50, M-75, and M-100 increased by 1.5%, 2.1%, 3.9%, and
ened properties of concrete materials, the air content of all ECC 4.4% when compared with the air content of the control mixture
mixtures was evaluated according to ASTM C231/C231M M-0, respectively. Consequently, the expected reductions in the
(ASTM 2017c). The air content reported in Table 5 shows a strong compressive strength due to the air content increase are approxi-
correlation with HRWR content, i.e., the air content increased with mately 7.5%, 10.5%, 19.5%, and 22% for regular concrete mixtures
increasing HRWR content. For instance, the air content for M-0 with the same air content as M-25, M-50, M-75, and M-100, re-
was 1.4%; whereas the air content for M-100 was 5.8%. In fact, the spectively. However, the compressive strength decrease reported
influence of HRWR on air content increment is well documented in in this study for M-25, M-50, M-75, and M-100 compared with
the existing literature (Huang et al. 2018; Łaźniewska-Piekarczyk the control mixture (33.58 MPa) were only 0.4%, 5.8%, 11.0%, and
2014). It is also observed that the rate of increase in air content is 10.3%, respectively. Thus, in all cases, the compressive strength
slower than the rate of increase in HRWR content. It is hypoth- decrease was much lower than the expected decrease based on the
esized that the organic carbon contained in fine particles of partially air content increase. This phenomenon could be attributed to the
burned fibers may have reduced the formation of air bubbles, par- pozzolanic activity and/or filler action of the SCBA, which likely
tially contrasting the effects of the HRWR. However, further studies contributed to mitigating the compressive strength reduction.
are needed to test this hypothesis.
Tensile Properties
Compressive Strength
Figs. 7(a–e) show the tensile stress-tensile strain curves at 28 days
The compressive strength test results for all ECC mixtures after of curing for all ECC materials considered in this study. Figs. 8(a
28 days of hydration are presented in Fig. 6(a). The use of raw and b) show the average tensile strength and tensile strain capacity,
SCBA produced minimal effects on the compressive strength of respectively, for all ECC mixtures. In this study, all ECC mixtures
the ECC materials. In general, as the replacement of sand with exhibited PSH behavior. It is observed that the SCBA content
SCBA increased, the compressive strength decreased slightly from significantly influenced both tensile strength and tensile ductility
M-0 to M-75 and remained almost constant from M-75 to M-100. of ECC mixtures.
Relative to the control mixture (with compressive strength of From Fig. 8(b), a progressive enhancement in tensile ductility
33.58 MPa), the greatest decrease of 11% in compressive strength of the composites is observed for increasing contents of SCBA
was observed for M-75 (i.e., 29.87 MPa). (with the exception of M-50 compared with M-25). For instance,

40 1950
Compressive Strength (MPa)

35
1900
30
Density (kg/m3)

1850
25
20 1800
15
1750
10
1700
5
0 1650
0 25 50 75 100 0 25 50 75 100
(a) % Replacement of Sand with Raw SCBA (b) % Replacement of Sand with Raw SCBA

Fig. 6. Hardened ECC properties: (a) ECC compressive strength; and (b) ECC hardened density.

© ASCE 04021231-8 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 6 6

5 5

Tensile Stress (MPa)

Tensile Stress (MPa)
4 4

3 3

2 2

1 1

0 0
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
(a) Tensile Strain (%) (b) Tensile Strain (%)

6 6

5 5
Tensile Stress (MPa)

Tensile Stress (MPa)

4 4

3 3

2 2

1 1

0 0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
(c) Tensile Strain (%) (d) Tensile Strain (%)

5
Tensile Stress (MPa)

0
0 1 2 3 4 5 6 7
(e) Tensile Strain (%)

Fig. 7. Tensile stress-strain curves for (a) M-0; (b) M-25; (c) M-50; (d) M-75; and (e) M-100.

the control mixture M-0 exhibited a tensile strain capacity of the matrix (K m ) and the initial flaw size (c). The air voids disrupt
1.04%, which increased to 4.28% for 100% sand replacement with the continuity of the matrix and can act as a critical flaw in the
SCBA (M-100), with an improvement of 311%. Previous studies cementitious matrix. As such, depending on the size of the air bub-
have concluded that the increase in aggregate particle size increases bles, these can result in the initiation of a new crack at a relatively
the tortuosity of the fracture path, which in turn increases the re- low external load/energy (referred to as cracking strength), leading
sistance to the crack propagation, i.e., fracture toughness K m and to enhanced PSH behavior. Previous studies revealed that the en-
crack-tip matrix toughness J tip (Moseley et al. 1987; Nallathambi trapped air voids (with size larger than the critical flaw size) reduce
et al. 1984; Perdikaris and Romeo 1995). According to ECC design the ECC cracking strength, thus leading to an increase in the tensile
principles, the decrease in J tip of the composite promotes the PSH ductility (Chen et al. 2014; Li 2012; Tosun-Felekoğlu et al. 2014,
behavior of ECC materials by facilitating the multiple microcracks’ 2017; Zhang et al. 2019). Furthermore, a positive correlation is ob-
formation. This phenomenon is quantified by the PSH energy in- served between the increase in the air content and the tensile duc-
dex, J b0 =J tip . From the particle-size analysis presented in Fig. 5(b), tility of the ECC mixtures produced in this study. This observation
it is observed that the raw SCBA is much finer than the river sand is consistent with the results from a recent study that reported a
utilized in this study; therefore, the fracture toughness of the decrease in J tip due to the addition of air bubbles in the ECC matrix
SCBA-ECC cementitious matrices was expected to decrease with (Zhang et al. 2019), which also promotes a robust PSH behavior.
the increase in the SCBA content, resulting in the observed robust In addition to the effects due to aggregate particle size and air
PSH behavior. content, the carbon content of the SCBA may also be a contributing
The air content increase reported in Table 5 can also play a role factor in the ductility enhancement observed. A study by Wang and
in enhancing the ductility of ECC mixtures. The cracking strength Li (2007) showed that inert particles such as carbon particles tend
of the cementitious matrix is defined by the fracture toughness of to concentrate on the surface of PVA fiber during mixing and

© ASCE 04021231-9 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 7 7

Tensile Strain Capacity (%)

6 6

Tensile Stress (MPa)

+22.2% +37.2% +40.1% +55.2% +44.5%
5 5
4 4
3 3
2 2
1 1
0 0
0 25 50 75 100 0 25 50 75 100
% Replacement of Sand with Raw SCBA % Replacement of Sand with Raw SCBA
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

First-Cracking Strength Tensile Strength

(a) (b)

Fig. 8. Uniaxial tensile test results: (a) first-cracking and tensile strength; and (b) strain at peak strength.

provide a lubrication mechanism that eases fiber pullout. The chemical bond Gd is mainly due to the formation of a calcium
cited study observed that the frictional stress (τ 0 ) between the ce- hydroxide layer at the fiber/matrix interface (Wang and Li 2007;
mentitious matrix and fibers decreases due to the carbon coating on Wu 2001; Yang et al. 2007). Therefore, pozzolanic materials with
the PVA fiber surface. The same study also suggested that the car- low calcium, such as Class F fly ash, are effective in reducing Gd
bon coating could also lead to the weakening of the chemical bond (Wang and Li 2007; Yang et al. 2007). From the results of the EDS
(Gd ) between the cementitious matrix and the PVA fibers. analysis in Table 3, it is observed that the SCBA utilized in this
From the EDS analysis results in Table 3, the SCBA exhibits study has a low content of CaO and a high content of SiO2 content,
significant amounts of carbon content; thus, the presence of a car- similar to the composition of a highly pozzolanic fly ash. Thus, the
bon coating of the PVA fibers in the SCBA-ECC mixtures is highly SCBA may further reduce the chemical bond at the fiber/matrix
likely. As a result, a decrease in τ 0 due to this carbon coating can interface in ECC materials by altering the chemical composition
lead to either a reduction or an increase in J b0 depending on the of the cementitious matrix.
initial τ 0 conditions. If the initial τ 0 is large and significant fiber Another significant factor that affects the ductility of the com-
rupture is occurring during fiber pullout from the cementitious posites is the fiber distribution, which is influenced by different
matrix, lowering τ 0 can cause a reduction in fiber rupture and, thus, factors, including the rheology properties and the aggregate particle
an increase in J b0 (which enhances the tensile ductility). By contrast, size of the mixture. With respect to the rheology properties, a low
if the initial τ 0 is low, a decrease in τ 0 can lead to a reduction in plastic viscosity of the cementitious matrix can produce a clumping
σ0 and, thus, to a decrease in J b0 (which negatively affects the ten- and a nonuniform distribution of the fibers, causing a reduction in
sile ductility) (Li 2003). Wu (2001) recommended an optimal the effective volume of fibers (V f ) at the weakest locations within a
range for τ 0 between 1.0 and 2.1 MPa to facilitate fiber pullout specimen (Li and Li 2013). In turn, this local reduction of V f
without failure. causes a reduction in σ0 and in J b0 , which can result in a substantial
Furthermore, other effects produced by the SCBA, such as mi- decrease in the ductility and tensile strength of the material. In this
crostructure enhancement effects (due to pozzolanic and/or filler study, the flowability of the mixtures was controlled using the
behavior), can mitigate the potential decrease in τ 0 due to the effect HRWR in order to obtain similar rheological properties for the
of carbon coating. In the case of the decrease in the chemical bond different ECC mixtures.
(Gd ), J b0 is increased by reducing premature fiber rupture during the It is also likely that the plastic viscosity of the mixtures in-
debonding of the fibers from the cementitious matrix, thus posi- creased with the increase of the SCBA amount, which corre-
tively affecting the tensile ductility of the ECC materials (Redon sponded to an observed thickening of the cementitious matrix.
et al. 2001). In this study, it is hypothesized that the fiber carbon Therefore, it is reasonable to assume that the main factor contrib-
coating effect produced by the SCBA positively affects the tensile uting to the fiber distribution in the ECC mixtures considered in this
ductility of the composites mainly due to the reduction of Gd . study was the aggregate particle size. It is known that an aggregate
However, to determine whether the carbon coating effect has a pos- with finer particle size enhances the fiber dispersion (De Koker and
itive or negative impact on the tensile ductility of the composites Van Zijl 2004). It is hypothesized that a more uniform fiber
due to its effect on τ 0 , it is recommended that future studies focus dispersion state was likely obtained for an increasing content of
on an in-depth examination of the fiber/matrix interfacial parame- SCBA, thus decreasing the possibility of forming weak zones with
ters through a single-fiber pullout test. lower effective fiber volumes. This hypothesis is consistent with
Figs. 9(a–d) present SEM images from the failure cross sections the enhancement in ductility observed in Fig. 8(b), which can be
of different SCTT specimens of the M-0, M-25, and M-50 ECC partially due to the uniform fiber distribution produced by the finer
mixtures. A coating of microparticles is observed on the surface particle size of the SCBA when compared with the particle size of
of the fibers in the M-25 and M-50 specimens [Figs. 9(c and d), the silica sand.
respectively], whereas this coating is not observed on the fibers in In summary, the enhancement in ductility observed in the ECC
the M-0 specimens [Figs. 9(a and b)]. This result supports the materials for increasing contents of SCBA may be associated with
hypothesis that the carbon particles in the SCBA tend to coat the combined effect of a reduction in J tip and an increase in J b0 . The
the PVA fibers, thus likely lowering the value of Gd . J tip reduction is most likely due to the decrease of the aggregate
In addition to this carbon coating effect, it is suggested that the particle size and the increase in the air content. The J b0 the incre-
pozzolanic activity of SCBA may also contribute to the reduction ment is most likely due to a decrease in Gd (most likely caused by
in Gd . Previous studies have suggested that the strength of the the potential formation of a carbon coating layer on the PVA fibers

© ASCE 04021231-10 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 No particles on
fiber surface
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

(a) (b)

Particles on
Particles on fiber surface
fiber surface

Particles on
fiber surface

(c) (d)

Fig. 9. BSE SEM images of ECC materials: (a) M-0 at 100× magnification; (b) M-0 at 250× magnification; (c) M-25 at 150× magnification; and
(d) M-50 at 200× magnification.

and by the pozzolanic properties of the SCBA), and to an enhanced content for increasing amounts of SCBA would suggest a corre-
fiber dispersion (most likely caused by the thickening of the cemen- sponding decrease in tensile strength due to a weaker matrix/fiber
titious matrix and by the finer aggregate size). interfacial bond. In this study, the combination of the filler effect
Regarding the tensile strength of the ECC mixtures, a tensile of the pozzolanic reactivity of the SCBA appear to outweigh the
strength higher than that of the control mixture was obtained for decrease in the value of τ 0 due to the increase in the air content
all ECC mixtures containing the SCBA. From Fig. 8(a), it is and to the carbon coating on the fiber surface.
observed that the largest tensile strength increase reported was Whereas the tensile strength of all SCBA-ECC materials con-
for M-25, where the relative strength increases when compared sidered in this study was higher than that of the control mixture,
with the control mixture (i.e., M-0 with 3.73 MPa) was 22.3% the opposite was observed for the first-cracking strength of the
(i.e., 4.56 MPa). The relative increments in tensile strength for the SCBA-ECC materials, with the exception of M-25, as shown in
M-50, M-75, and M-100 mixtures were 5.1% (i.e., 3.92 MPa), Fig. 8(a). The first-cracking strength of the cementitious matrix
8.3% (i.e., 4.04 MPa), and 8.6% (i.e., 4.05 MPa), respectively. This is primarily influenced by the characteristics of the cementitious
behavior is credited to the filler effect and/or pozzolanic reactivity matrix (i.e., the matrix fracture toughness and the pre-existing ini-
of the SCBA, which likely enhanced the microstructure, thus in- tial flaw size) and not on the attributes of the fiber/matrix interface,
creasing the value of τ 0 . This effect has been previously observed which become dominant in the postcracking behavior (Chen et al.
in ECC materials when high levels of cement were replaced with 2014; Li 2003, 2019). Therefore, the observed decrease in the first-
fly ash are (Yang et al. 2007). The observed increase in the air cracking strength is consistent with a decrease in the matrix fracture

© ASCE 04021231-11 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 toughness due to the presence of aggregate with smaller particle a robust PSH behavior) likely resulted in the formation of multiple
size (i.e., SCBA), and with an increase in the pre-existing initial microcracks in the thin notched area.
flaw size due to added artificial defects (i.e., additional air voids). Based on these results, it is recommended that future studies
evaluate the fiber-bridging relation of SCBA-ECC mixtures mate-
rials through the SCTT with a reduced fiber content (i.e., 1% vol-
Fiber-Bridging Properties ume fraction or lower) so that a single crack can be consistently
Figs. 10(a–e) plot the fiber-bridging stress (σ) versus crack opening obtained and a more accurate representation of the fiber-bridging
(δ) curves (i.e., the fiber-bridging relation) obtained through the relation can be achieved. It is important to consider that, whereas a
SCTT for all ECC mixtures produced in this study. The dashed lower fiber content can help produce a single crack in the notched
lines in Figs. 10(a–e) correspond to the fiber-bridging relations testing section of the SCTT, the fiber/matrix micromechanical in-
of each specimen, whereas the solid lines correspond to the average terfacial parameters, such as τ 0 , Gd , and the slip-hardening coef-
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

fiber-bridging relation for each mixture. The experimental fiber- ficient β, tend to increase when the fiber content decreases due to a
bridging relations do not always exhibit only a single peak (as lower air content in the cementitious matrix (Yang et al. 2008).
it would be expected in the case of a single crack). On the contrary, Thus, it is important to consider that using fiber-bridging relations
fiber-bridging relations with more than one peak were observed for obtained at lower fiber contents (i.e., 1% volume fraction) to predict
all mixtures. For the ECC mixtures with high SCBA content the fiber-bridging relation of a composite with higher fiber content
(i.e., M-75 and M-100), the curves with multiple peaks (i.e., three (i.e., 1.5%) can also result in inaccurate results.
to four peaks) were predominant, as shown in Figs. 10(d and e). Fig. 10(f) compares the average fiber-bridging curves for all
This behavior is consistent with the uniaxial tensile test results ECC mixtures. It is observed that for all SCBA-ECC mixtures with
because the high ductility shown by M-75 and M-100 (through the exception of M-25, the fiber-bridging curves shifted downward

8 8
(MPa)

6 (MPa) 6
Tensile Stress

Tensile Stress

4 4

2 2

0 0
0.0 0.5 1.0 0.0 0.5 1.0
(a) Crack opening (mm) (b) Crack opening (mm)

8 8
(MPa)

(MPa)

6 6
Tensile Stress

Tensile Stress

4 4

2 2

0 0
0.0 0.5 1.0 0.0 0.5 1.0
(c) Crack opening (mm) (d) Crack opening (mm)

8 8
M-0
M-25
(MPa)

(MPa)

M-50
6 6 M-75
M-100
Tensile Stress

Tensile Stress

4 4

2 2

0 0
0.0 0.5 1.0 0.0 0.5 1.0
(e) Crack opening (mm) (f) Crack opening (mm)

Fig. 10. SCTT bridging stress versus crack-opening curves: (a) M-0; (b) M-25; (c) M-50; (d) M-75; (e) M-100; and (f) average curves for all
mixtures.

© ASCE 04021231-12 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 (i.e., with a decrease in σ0 ) and to the right (i.e., with an increase in control mixture). In fact, specimens with high contents of SCBA
the crack opening δ0 corresponding to σ0 ) when compared with the experienced multiple cracking within the notched area during the
fiber-bridging curve of the control mixture M-0. By contrast, the SCTT, thus explaining the difference observed in tensile strength
fiber-bridging curve of mixture M-25 shifted upward (i.e., with results (from the uniaxial tensile test) and fiber-bridging capacity
an increase in σ0 ) and had a similar δ 0 as the corresponding curve results (from the SCTT).
for M-0. Based on the ECC design criteria, an increase in σ0 and the From Fig. 11(b), it is observed that δ 0 increases for increasing
shift of the fiber-bridging relation to the right (corresponding to an SCBA content up to 50% replacement of silica sand. For instance,
increase in δ0 ) promotes the PSH behavior of the ECC materials by M-0 exhibited a crack opening of 0.26 mm at σ0 , whereas both
corresponding to an increase of J b0 . M-50 and M-100 presented a crack opening equal to 0.36 mm.
Figs. 11(a–c) report the influence of the SCBA content on the The increase in δ 0 may be due to the combined influence of the
main parameters describing the fiber-bridging relation of ECC multiple microcrack formation observed in the SCTT and the
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

materials, i.e., σ0 , δ 0 , and J b0 , respectively. The fiber-bridging possible decrease in Gd . As discussed previously in the “Uniaxial
capacity σ0 shown in Fig. 11(a) presents a similar trend as that of Tensile Test” section, the increment in the air content of the cemen-
the tensile strength of the ECC materials observed in Fig. 8(a). In titious matrix, the carbon coating formed on the surface of the fi-
the SCTT, M-0 exhibited a fiber-bridging capacity of 5.58 MPa, bers, and the pozzolanic effect of SCBA can all contribute to the
which increased to 6.26 MPa for M-25 (i.e., corresponding to a decrease in the chemical bond between the fibers and the matrix.
12.2% increase). For ECC mixtures incorporating SCBA contents Fig. 11(c) presents the complementary energy, J b0 , for all ECC
higher than 25% (i.e., M-50, M-75, and M-100), the fiber-bridging mixtures. In terms of fiber/matrix interfacial properties, the re-
capacity was lower than that of the control mixture, with M-100 sults show that the ECC mixture most effective in maximizing
exhibiting the lowest fiber-bridging capacity of 4.64 MPa (i.e., cor- the tensile ductility is M-50 because it exhibited the highest J b0
responding to a 25% decrease when compared with that of the (i.e., 582.32 J=m2 ). However, M-50 was also the ECC material
control mixture). with the lowest tensile strain capacity among all SCBA-ECC mix-
In general, it is observed that σ0 is only slightly influenced by tures. This phenomenon may be because the energy PSH index de-
the replacement amount of sand with raw SCBA. The σ0 values pends on both J b0 and J tip . Thus, although J b0 was highest for M-50,
were higher than corresponding tensile strength values. This result J tip was likely not. In turn, this result suggests that the effect of the
is because the failure of an ECC specimen in a regular uniaxial SCBA in reducing J tip may be the predominant effect in the ob-
tensile test occurs in the weakest of several cracks developing in served enhancements of the tensile ductility. It is recommended that
the material, whereas the failure in the SCTT occurs in a single fracture toughness tests should be conducted in future research to
random crack developing in the notched area of the specimen. confirm this hypothesis.
Therefore, on average, the tensile strength is always lower than
σ0 . This phenomenon, in turn, is likely responsible for the lower
Surface Resistivity
fiber-bridging capacities observed for M-50, M-75, and M-100
when compared with the control mixture (which is in contrast with The same cylindrical specimens utilized for the compressive
the tensile strength results from the uniaxial tensile tests, i.e., all strength assessment were first used for the surface resistivity test
SCBA-ECC mixtures exhibited higher tensile strength than the to gain insight into the effect of raw SCBA on the permeability

8 0.5
(MPa)

0 (mm)

0.4
6
Fiber Bridging Capacity,

Crack opening at σ0 ,

0.3
4
0.2

2
0.1

0 0.0
0 25 50 75 100 0 25 50 75 100
(a) % Replacement of Sand with Raw SCBA (b) % Replacement of Sand with Raw SCBA
800
Complementary Energy (J/m2)

700
600
500
400
300
200
100
0
0 25 50 75 100
(c) % Replacement of Sand with Raw SCBA

Fig. 11. SCTT experimental results: (a) fiber-bridging capacity, σ0 ; (b) crack opening at σ0 , δ 0 ; and (c) complementary energy, J b0 .

© ASCE 04021231-13 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 30

Surface Resistivity (kΩ-cm)

Low CIP
25 SCBA Particle

20

15 Moderate CIP

10
High CIP
5
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

0
0 25 50 75 100
% Replacement of Sand with Raw SCBA
(a) (b)

Fig. 12. Surface resistivity analysis: (a) surface resistivity test results; and (b) BSE SEM image of M-50.

of ECC materials. Fig. 12(a) presents the surface resistivity test 250
results as a function of the percent replacement of silica sand with
raw SCBA. The surface resistivity of all SCBA-ECC mixtures 200

Length Change (µ )
was lower than that of the control ECC mixture, with a clear
negative correlation between the SCBA content and the surface re- 150
sistivity. The control ECC mixture exhibited a surface resistivity of
23.9 kΩ · cm, which qualified as a low chloride ion penetrability 100
(CIP) as per AASHTO T 358 (with low CIP corresponding to the
range 21–37 kΩ · cm) (AASHTO 2017). The surface resistivity for 50
all SCBA-ECC mixtures was lower and corresponded to a
moderate CIP (with moderate CIP corresponding to the range 0
12–21 kΩ · cm), except for M-25, which correspond to low CIP. 0 25 50 75 100
% Replacement of Sand with Raw SCBA
The observed decrease in surface resistivity may be attributed to
the increasing air content in the ECC mixtures as the SCBA content 7 Days 14 Days 28 Days

was increased (Table 5). In fact, the air voids provide empty spaces Fig. 13. ECC length change during curing.
inside the concrete material, which increase its permeability and
decrease its surface resistivity. Another possible reason for this de-
crease in the surface resistivity may be the porous nature of the
SCBA, as shown in Fig. 12(b). The presence of porous components observed in regular concrete, this result indicates that it is crucial to
in the microstructure of the ECC cementitious matrices increases assess the dimensional stability of SCBA-ECC materials during the
the total porosity of the materials, which again can negatively in- curing period.
fluence the surface resistivity of the ECC. Finally, the carbon The decrease in the length change for the SCBA-ECC materials
present in the SCBA may also affect the electrical conductivity in comparison with the control mixture may be partially attributed
of the ECC materials, producing a decrease in the electrical resis- to the restraining action of coarser aggregate in the mortar, i.e., an
tivity of the SCBA-ECC mixtures. In order to more effectively increase in aggregate size increases the restraining action (Liu et al.
assess the permeability of the SCBA-ECC materials and separate 2016; Şahmaran et al. 2009). Because the particle size of the raw
these hypothesized effects, it is recommended to use in future SCBA utilized in this study is smaller than that of the silica sand,
research more direct permeability assessment methods, such as the restraining action of the larger aggregate is likely limiting the
the rapid chloride ion penetrability test. length change of the control mixture. Another potential contributor
to this length change may be the porous nature of the SCBA, which
Dimensional Stability can two conflicting effects. On one hand, the higher porosity of
SCBA-ECC mixtures could increase the length change due to the
The results of the length change at 7, 14, and 28 days of curing are presence of more capillary pores than in the control mixture, which
shown in Fig. 13 for all ECC mixtures considered in this study. could increase the loss of free water (Chatveera and Lertwattanaruk
These length change values were obtained by averaging the mea- 2011). On the other hand, porous aggregates can absorb part of the
surements from four specimens. It is observed that with the ex- mixing water and then release it during drying, thus acting as an
ception of M-25 at 28 days of curing, the length change of all internal curing agent and reducing the length change. It is recom-
SCBA-ECC materials is higher than that of the control mixture mended that future studies investigate these different mechanisms
at all ages of curing. At 28 days of curing, the length change in- to identify their relative contributions to the length changes ob-
creases with respect to that of the control mixture were 34%, 2%, served in the present study.
and 24% for M-50, M-75, and M-100, respectively. The length
change of the M-25 mixture at 28 days of curing was 8% lower
than the corresponding length change of the control mixture. No Summary and Conclusion
distinctive correlation was observed between length change and
the contents of SCBA. Even though the magnitude of the length This study investigated the effects of SCBA on the properties of
changes is relatively low in comparison with the shrinkage values ECC materials when utilized as a partial and complete replacement

© ASCE 04021231-14 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 for silica sand. Furthermore, this study explored the effects of single crack and a more accurate representation of the fiber-
SCBA on the fiber-bridging relation of the composites. Based bridging relations.
on the results of the experimental program, the following conclu- • As the sand replacement with SCBA increased, the surface
sions are drawn: resistivity decreased accordingly. The decrease in surface resis-
• SEM-EDS analysis revealed that the raw SCBA consisted tivity was credited to the increment in the air content, the pres-
mainly of small (i.e., 256 μm average particle size), porous, and ence of porous SCBA particles, and the possible electrical
irregularly shaped particles with carbon and silica as its main conductivity enhancement effect of carbon particles present
constituents. In addition, the pozzolanic components of the in the SCBA.
SCBA were determined to be 52.68%, thus meeting the ASTM • With the exclusion of the ECC mixture with 25% sand replace-
C618 requirement to be classified as a Class C pozzolan. More- ment with SCBA at 28 days of curing, the length changes of all
over, the SCBA exhibited a SAI of 72.1%, which was greater SCBA-ECC materials at all ages of hydration were higher than
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

than that of inert silica sand (i.e., 63.72%), thus confirming the that of the control mix.
presence of pozzolanic and/or filler activity. However, it is con- Overall, the experimental results showed that replacing silica
cluded that the raw SCBA cannot be utilized as SCM because it sand with the raw SCBA in the ECC materials considered in this
exhibited a lower SAI and a higher loss on ignition than those study enhanced the tensile strength and ductility and had only mi-
indicated as minimum requirements for SCMs according to the nor effects on the compressive strength of these materials. There-
ASTM C618 standard. fore, raw SCBA presents a great potential to be utilized as a sand
• Increasing the replacement amounts of sand with SCBA in the replacement in ECC mixtures because the use of low-cost materials
ECCs resulted in a substantial loss in workability, which was such as the raw SCBA could allow the production of more practical
mainly attributed to the irregular shape and small size of the and cost-effective ECC materials in several places around the
SCBA particles. The loss in workability was successfully miti- world (including the US) where this agricultural waste is available.
gated by incrementing the dosage of HRWR. At high replace- However, this study did not investigate the progress with age of
ment levels, the dosage of HRWR necessary to achieve a the mechanical properties of ECC materials. It is recommended
workable mixture produced a considerable increase in the air to investigate the mechanical properties of SCBA-ECC materials
content (i.e., the air content increased from 1.4% for the control at ages later than 28-day curing in future research. In addition, a
mixture to 5.8% for 100% sand replacement with SCBA). life cycle assessment of the usage of SCBA to replace microsilica
• As the SCBA content increased, the compressive strength de- in ECC materials is beyond the scope of this study. However, the
creased slightly from the control mixture with 0% replacement life cycle assessment comparison of SCBA and microsilica used in
to the mixture with 75% replacement of sand with SCBA, for ECC materials and obtained from industrial processes optimized
which a maximum strength decrease of 11% was observed. for commercialization is also an important research topic for future
There was no further degradation of the compressive strength studies.
for the mixture with 100% replacement of sand with SCBA.
The decrease in compressive strength was attributed to a de-
crease in hardened density and an increase in the air content Data Availability Statement
of the SCBA-ECC materials for increasing replacements of sand
with SCBA. All data, models, and code generated or used during the study ap-
• An improvement in the tensile ductility of 311% compared with pear in the published article.
the control mixture was observed for the SCBA-ECC material at
100% replacement of sand with SCBA. The enhancement in the
tensile ductility was related to the reduction in J tip (credited to References
an increase in the air content and a decrease in the aggregate
particle size), reduction in ECC cracking strength (due to in- AASHTO. 2017. Standard method of test for surface resistivity indication
crease in air content), and increase in J b0 [associated with a po- of concrete’s ability to resist chloride ion penetration. AASHTO T 358.
tential decrease in the fiber/matrix chemical bond (Gd ) and an Washington, DC: AASHTO.
ACI (American Concrete Institute). 2016. Report on chemical admixtures
enhanced fiber dispersion]. With respect to the tensile strength,
for concrete. ACI 212.3 R-16. Farmington Hills, MI: ACI
all SCBA-ECC materials outperformed the control mixture,
American Sugar Cane League. 2019. Louisiana sugarcane industry
with the largest improvement in tensile strength of 22.3% occur- production data from 1975 to 2016. Thibodaux, LA: American Sugar
ring at 25% sand replacement with SCBA. The observed tensile Cane League.
strength improvements were attributed to a combination of the Arce, G., M. M. Hassan, M. T. Gutierrez-Wing, and M. Barbato. 2019. Use
filler effect and the pozzolanic reactivity of the SCBA material, of bagasse ash as a concrete additive for road pavement application.
which likely increased the fiber/matrix frictional bond (τ 0 ). Tran-Set Project No. 18CLSU03. Baton Rouge, LA: Transportation
• The fiber-bridging relations obtained from the SCTT show that Consortium of South-Central States.
in terms of fiber/matrix interfacial properties, the most effective Arce, G., H. Noorvand, M. Hassan, and T. Rupnow. 2018. Use of bagasse
ECC mixture to maximize tensile ductility was the one with ash as a concrete additive for road pavement application. Tran-SET
50% sand replacement with SCBA because it exhibited the Project No. 18CLSU03. Baton Rouge, LA: Transportation Consortium
highest J b0 (i.e., 582.32 J=m2 ). Interestingly, the same mixture of South-Central States.
was also the SCBA-ECC material with the lowest tensile strain ASTM. 2016a. Standard test method for compressive strength of hydraulic
cement mortars (using 2-in. or [ 50-mm ] cube specimens). ASTM
capacity. This finding suggests that the effect of the SCBA in
C109/109M. West Conshohocken, PA: ASTM.
reducing J tip may be the predominant effect in the observed ASTM. 2016b. Standard test methods for sampling and testing fly ash or
enhancements of the tensile ductility. Because multiple micro- natural pozzolans for use. ASTM C311/C311M. West Conshohocken,
cracks were observed in the SCTT, it is suggested that future PA: ASTM.
studies evaluate the fiber-bridging relation through the SCTT ASTM. 2017a. Standard specification for coal fly ash and raw or
by utilizing a smaller fiber content than the one used in this calcined natural Pozzolan for use in concrete. ASTM C618. West
study (e.g., with a 1% volume fraction or lower) to obtain a Conshohocken, PA: ASTM.

© ASCE 04021231-15 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 ASTM. 2017b. Standard specification for standard sand. ASTM C778. concrete.” Constr. Build. Mater. 53 (Feb): 488–502. https://doi.org/10
West Conshohocken, PA: ASTM. .1016/j.conbuildmat.2013.11.092.
ASTM. 2017c. Standard test method for air content of freshly mixed Li, M., and V. C. Li. 2013. “Rheology, fiber dispersion, and robust proper-
concrete by the pressure method. ASTM C231/C231M. West ties of engineered cementitious composites.” Mater. Struct. 46 (3):
Conshohocken, PA: ASTM. 405–420. https://doi.org/10.1617/s11527-012-9909-z.
ASTM. 2017d. Standard test method for length change of hardened Li, V. C. 1992. “Postcrack scaling relations for fiber-reinforced cementi-
hydraulic-cement mortar and concrete. ASTM C157/C157M. West tious composites.” J. Mater. Civ. Eng. 4 (1): 41–57. https://doi.org/10
Conshohocken, PA: ASTM. .1061/(ASCE)0899-1561(1992)4:1(41).
ASTM. 2018. Standard test method for compressive strength of cylindrical Li, V. C. 1993. “From micromechanics to structural engineering—the
concrete specimens. ASTM C39/C39M. West Conshohocken, PA: design of cementitious composites for civil engineering applications.”
ASTM. J. Struct. Mech. Earthquake Eng. 10 (2): 37–48.
ASTM. 2019. Standard specification for mixing rooms, moist cabinets, Li, V. C. 2003. “On engineered cementitious composites (ECC). A review
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

moist rooms, and water storage tanks used in the testing of hydraulic of the material and its applications.” J. Adv. Concr. Technol. 1 (3):
cements and concrete. ASTM C511. West Conshohocken, PA: ASTM. 215–230. https://doi.org/10.3151/jact.1.215.
Bahurudeen, A., and M. Santhanam. 2015. “Influence of different process- Li, V. C. 2008. “Engineered cementitious composites (ECC)—Material,
ing methods on the pozzolanic performance of sugarcane bagasse
structural, and durability performance.” In Concrete construction engi-
ash.” Cem. Concr. Compos. 56 (Feb): 32–45. https://doi.org/10.1016/j
neering handbook, edited by E. Nawy, 1–78. Boca Raton, FL: CRC
.cemconcomp.2014.11.002.
Press.
Chang, S. J. 2020. “To burn or not to burn: Bagasse that is.” Accessed July
Li, V. C. 2012. “Tailoring ECC for special attributes: A review.” Int. J.
15, 2020. https://www.swst.org/wp/meetings/AM15/pdfs/presentations
Concr. Struct. Mater. 6 (3): 135–144. https://doi.org/10.1007/s40069
/chang.pdf.
-012-0018-8.
Chatveera, B., and P. Lertwattanaruk. 2011. “Durability of conventional
concretes containing black rice husk ash.” J. Environ. Manage. 92 (1): Li, V. C. 2019. “Engineered cementitious composites (ECC).” In Engi-
59–66. https://doi.org/10.1016/j.jenvman.2010.08.007. neered cementitious composites (ECC): Bendable concrete for sustain-
Chen, Z., E. H. Yang, Y. Yang, and Y. Yao. 2014. “Latex-modified engi- able and resilient infrastructure, 286–290. Berlin: Springer.
neered cementitious composites (L-ECC)” J. Adv. Concr. Technol. Li, V. C., and C. K. Y. Leung. 1992. “Steady-state and multiple cracking of
12 (12): 510–519. https://doi.org/10.3151/jact.12.510. short random fiber composites.” J. Eng. Mech. 118 (11): 2246–2264.
Cordeiro, G. C., L. M. Tavares, and R. D. Toledo Filho. 2016. “Improved https://doi.org/10.1061/(ASCE)0733-9399(1992)118:11(2246).
pozzolanic activity of sugar cane bagasse ash by selective grinding and Li, V. C., C. Wu, S. Wang, A. Ogawa, and T. Saito. 2002. “Interface tailor-
classification.” Cem. Concr. Res. 89: 269–275. https://doi.org/10.1016/j ing for strain-hardening polyvinyl alcohol-engineered cementitious
.cemconres.2016.08.020. composite.” ACI Mater. J. 99 (5): 463–472.
Cordeiro, G. C., R. D. Toledo Filho, L. M. Tavares, and E. D. M. R. Liu, J., F. Han, G. Cui, Q. Zhang, J. Lv, L. Zhang, and Z. Yang. 2016.
Fairbairn. 2009. “Ultrafine grinding of sugar cane bagasse ash for ap- “Combined effect of coarse aggregate and fiber on tensile behavior of
plication as pozzolanic admixture in concrete.” Cem. Concr. Res. 39 (2): ultra-high performance concrete.” Constr. Build. Mater. 121 (Sep):
110–115. https://doi.org/10.1016/j.cemconres.2008.11.005. 310–318. https://doi.org/10.1016/j.conbuildmat.2016.05.039.
Cordeiro, G. C., R. D. Toledo Filho, L. M. Tavares, and E. M. R. Fairbairn. Ma, H., S. Qian, Z. Zhang, Z. Lin, and V. C. Li. 2015. “Tailoring engineered
2008. “Pozzolanic activity and filler effect of sugar cane bagasse ash cementitious composites with local ingredients.” Constr. Build. Mater.
in portland cement and lime mortars.” Cem. Concr. Compos. 30 (5): 101 (Dec): 584–595. https://doi.org/10.1016/j.conbuildmat.2015.10.146.
410–418. https://doi.org/10.1016/j.cemconcomp.2008.01.001. Marshall, D. B., and B. N. Cox. 1988. “A J-integral method for calculating
De Koker, D., and G. P. A. G. Van Zijl. 2004. “Extrusion of engineered steady-state matrix cracking stresses in composites.” Mech. Mater.
cement-based composite material.” In Proc., 6th RILEM Symp. on 7 (2): 127–133. https://doi.org/10.1016/0167-6636(88)90011-7.
Fiber-Reinforced Concretes (FRC)-BEFIB, 1301–1310. Paris: Interna- Modani, P. O., and M. R. Vyawahare. 2013. “Utilization of bagasse ash as a
tional Union of Laboratories. partial replacement of fine aggregate in concrete.” Procedia Eng. 51 (0):
Degen, T., M. Sadki, E. Bron, U. König, and G. Nénert. 2014. “The high 25–29. https://doi.org/10.1016/j.proeng.2013.01.007.
score suite.” Powder Diffr. 29 (May): S13–S18. https://doi.org/10.1017 Moseley, M. D., R. P. Ojdrovic, and H. J. Petroski. 1987. “Influence of
/S0885715614000840. aggregate size on fracture toughness of concrete.” Theor. Appl. Fract.
Dhengare, S., S. Amrodiya, M. Shelote, A. Asati, N. Bandwaf, K. Anand, Mech. 7 (3): 207–210. https://doi.org/10.1016/0167-8442(87)90038-3.
and R. Jichkar. 2015. “Utilization of sugarcane Bagasse ash as a sup- Şahmaran, M., M. Lachemi, K. M. A. Hossain, R. Ranade, and V. C. Li.
plementary cementitious material in concrete and mortar—A review.” 2009. “Influence of aggregate type and size on ductility and mechanical
Int. J. Civ. Eng. Technol. 6 (4): 94–106.
properties of engineered cementitious composites.” ACI Mater. J.
FAO (Food and Agriculture Organization of the United Nations). 2020.
106 (6): 308–316.
“Sugarcane crops production.” Accessed July 15, 2020. https://faostat3
Nallathambi, P., B. L. Karihaloo, and B. S. Heaton. 1984. “Effect of
.fao.org.
specimen and crack sizes, water/cement ratio, and coarse aggregate tex-
Huang, F., H. Li, Z. Yi, Z. Wang, and Y. Xie. 2018. “The rheological
ture upon fracture toughness of concrete.” Mag. Concr. Res. 36 (129):
properties of self-compacting concrete containing superplasticizer and
air-entraining agent.” Constr. Build. Mater. 166 (Mar): 833–838. 227–236. https://doi.org/10.1680/macr.1984.36.129.227.
https://doi.org/10.1016/j.conbuildmat.2018.01.169. Noorvand, H., G. Arce, M. Hassan, and T. Rupnow. 2019. “Investigation
JSCE (Japan Society of Civil Engineers). 2008. “Recommendations for of the mechanical properties of engineered cementitious compo-
design and construction of high performance fiber reinforced cement sites with low fiber content and with crumb rubber and high fly ash
composites with multiple fine cracks (HPFRCC).” Concr. Eng. Ser. content.” Transp. Res. Rec. 2673 (5): 418–428. https://doi.org/10.1177
82 (Mar): 6–10. /0361198119837510.
Kanda, T., and V. C. Li. 2006. “Practical design criteria for saturated pseudo O’Connor, B. H., and M. D. Raven. 1988. “Application of the Rietveld
strain hardening behavior in ECC.” J. Adv. Concr. Technol. 4 (1): refinement procedure in assaying powdered mixtures.” Powder Diffr.
59–72. https://doi.org/10.3151/jact.4.59. 3 (1): 2–6.
Kim, J. K., J. S. Kim, G. J. Ha, and Y. Y. Kim. 2007. “Tensile and fiber Parra-Montesinos, G. J. 2005. High-performance fiber reinforced cement
dispersion performance of ECC (engineered cementitious composites) composites, 6. Berlin: Springer.
produced with ground granulated blast furnace slag.” Cem. Concr. Res. Payá, J., J. Monzó, M. V. Borrachero, L. Díaz-Pinzón, and L. M. Ordónez.
37 (7): 1096–1105. https://doi.org/10.1016/j.cemconres.2007.04.006. 2002. “Sugar-cane bagasse ash (SCBA): Studies on its properties for
Łaźniewska-Piekarczyk, B. 2014. “The methodology for assessing the impact reusing in concrete production.” J. Chem. Technol. Biotechnol. 77 (3):
of new generation superplasticizers on air content in self-compacting 321–325. https://doi.org/10.1002/jctb.549.

© ASCE 04021231-16 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231

 Perdikaris, P. C., and A. Romeo. 1995. “Size effect on fracture energy of sand.” Can. J. Civ. Eng. 40 (2): 151–157. https://doi.org/10.1139/cjce
concrete and stability issues in three-point bending fracture toughness -2012-0281.
testing.” ACI Mater. J. 92 (5): 483–496. Wang, S., and V. C. Li. 2006. “High-early-strength engineered cementitious
Pereira, E. B., and G. Fischer. 2010. “Crack formation and tensile stress- composites.” ACI Mater. J. 103 (2): 97–105.
crack opening behavior of fiber reinforced cementitious composites Wang, S., and V. C. Li. 2007. “Engineered cementitious composites with
(FRCC).” In Proc., Institute for Sustainability and Innovation in Struc- high-volume fly ash.” ACI Mater. J. 104 (3): 233–241.
tural Engineering (ISISE)—Comunicações a Conferências Internacio- Wu, C. 2001. “Micromechanical tailoring of PVA-ECC for structural ap-
nais, 1–10. Guimarães, Portugal: Institute for Sustainability and plications.” Accessed January 12, 2020. https://deepblue.lib.umich.edu
Innovation in Structural Engineering. /handle/2027.42/124257.
Pereira, E. B., G. Fischer, and J. A. O. Barros. 2012. “Direct assessment of Wu, H.-L., J. Yu, D. Zhang, J. X. Zheng, and V. C. Li. 2019. “Effect of
tensile stress-crack opening behavior of strain hardening cementitious morphological parameters of natural sand on mechanical properties of
composites (SHCC).” Cem. Concr. Res. 42 (6): 834–846. https://doi.org engineered cementitious composites.” Cem. Concr. Compos. 100 (Jul):
Downloaded from ascelibrary.org by GMR INSTITUTE OF TECHNOLOGY on 07/19/22. Copyright ASCE. For personal use only; all rights reserved.

/10.1016/j.cemconres.2012.03.006. 108–119. https://doi.org/10.1016/j.cemconcomp.2019.04.007.

Redon, C., V. C. Li, C. Wu, H. Hoshiro, T. Saito, and A. Ogawa. 2001. Yang, E., S. Wang, Y. Yang, and V. C. Li. 2008. “Fiber-bridging constit-
“Measuring and modifying interface properties of PVA fibers in the utive law of engineered cementitious composites.” J. Adv. Concr. Tech-
ECC matrix.” J. Mater. Civ. Eng. 13 (6): 399–406. https://doi.org/10 nol. 6 (1): 181–193. https://doi.org/10.3151/jact.6.181.
.1061/(ASCE)0899-1561(2001)13:6(399). Yang, E. H., Y. Yang, and V. C. Li. 2007. “Use of high volumes of fly ash
Sales, A., and S. A. Lima. 2010. “Use of Brazilian sugarcane bagasse ash to improve ECC mechanical properties and material greenness.” ACI
in concrete as sand replacement.” Waste Manage. 30 (6): 1114–1122. Mater. J. 104 (6): 620–628.
https://doi.org/10.1016/j.wasman.2010.01.026. Yang, Y., M. D. Lepech, E. H. Yang, and V. C. Li. 2009. “Autogenous
Sua-Iam, G., and N. Makul. 2013. “Use of increasing amounts of bagasse healing of engineered cementitious composites under wet–dry cycles.”
ash waste to produce self-compacting concrete by adding limestone Cem. Concr. Res. 39 (5): 382–390. https://doi.org/10.1016/j.cemconres
powder waste.” J. Cleaner Prod. 57 (Oct): 308–319. https://doi.org/10 .2009.01.013.
.1016/j.jclepro.2013.06.009. Yu, K., Y. Wang, J. Yu, and S. Xu. 2017. “A strain-hardening cementitious
Subedi, S., G. Arce, M. Hassan, N. Kumar, M. Barbato, and M. T. composites with the tensile capacity up to 8%.” Constr. Build. Mater.
Gutierrez-Wing. 2019. “Influence of production methodology on the 137 (Apr): 410–419. https://doi.org/10.1016/j.conbuildmat.2017
pozzolanic activity of sugarcane bagasse ash.” In Proc., MATEC Web .01.060.
of Conf., 1–5. Baton Rouge, LA: Transportation Consortium of South- Yu, K. Q., Z. D. Lu, J. G. Dai, and S. P. Shah. 2020. “Direct tensile prop-
Central States. erties and stress-strain model of UHP-ECC.” J. Mater. Civ. Eng.
Tosun-Felekoǧlu, K., B. Felekoǧlu, R. Ranade, B. Y. Lee, and V. C. Li. 32 (1): 04019334. https://doi.org/10.1061/(ASCE)MT.1943-5533
2014. “The role of flaw size and fiber distribution on tensile ductility .0002975.
of PVA-ECC.” Composites, Part B 56 (Jan): 536–545. https://doi.org Zhang, Z., A. Yuvaraj, J. Di, and S. Qian. 2019. “Matrix design of light
/10.1016/j.compositesb.2013.08.089. weight, high strength, high ductility ECC.” Constr. Build. Mater.
Tosun-Felekoğlu, K., E. Gödek, M. Keskinateş, and B. Felekoğlu. 2017. 210 (Jun): 188–197. https://doi.org/10.1016/j.conbuildmat.2019
“Utilization and selection of proper fly ash in cost-effective green .03.159.
HTPP-ECC design.” J. Cleaner Prod. 149 (Apr): 557–568. https://doi Zhou, J., S. Qian, M. G. Sierra Beltran, G. Ye, K. van Breugel, and V. C. Li.
.org/10.1016/j.jclepro.2017.02.117. 2010. “Development of engineered cementitious composites with lime-
Turk, K. 2013. “The mechanical properties of engineered cementitious stone powder and blast furnace slag.” Mater. Struct. 43 (6): 803–814.
composites containing limestone powder replaced by microsilica https://doi.org/10.1617/s11527-009-9549-0.

© ASCE 04021231-17 J. Mater. Civ. Eng.

J. Mater. Civ. Eng., 2021, 33(9): 04021231