Mechanical characterization of a FRCM system with aramid ...

Journal of Asian Concrete Federation

Vol. 6, No. 2, pp. 1-13, December 2020

ISSN 2465-7964 / eISSN 2465-7972

https://doi.org/10.18702/acf.2020.12.6.2.1

1

Technical Paper

Mechanical characterization of a FRCM system with aramid

fiber fabric embedded in green high-strength cementitious

matrix

Munkhtuvshin Ochirbud, Donguk Choi*, Undram Naidanjav, S.-S. Ha, C.-Y. Lee

(Received August 13, 2020; Revised October 26, 2020; Accepted October 28, 2020; Published December 31, 2020)

Abstract: Fabric reinforced cementitious matrix (FRCM) system can be applied to strengthen existing

RC structures. In this experimental study, aramid fiber (AF) mesh was used along with green high-

strength mortar (f’c = 75.6 MPa) incorporating fine waste glass powder as partial replacement of cement

and 100% recycled fine aggregate. Test objective was to provide basic design parameters through me-

chanical characterization of the AF-FRCM system. Three different types of tests were conducted: Uni-

axial test of tensile specimens; flexural test of composite short beams; and pull-off test of thin FRCM

placed on top of normal strength concrete. Thickness of FRCM was about 10 mm while the volumetric

ratio of the fiber fabric to gross volume was 1.3% (0.65% in each direction). Tensile test results showed

that the load-displacement relationship of the FRCM was relatively ductile, while tensile behavior of the

AF governed at the peak load. Nominal tensile strength of the FRCM cross-section was 6.4 MPa at 4.6%

strain of the composite material. Short beam strength of 1.3 MPa was determined from flexural test of

composite short beams (or interlaminar shear test) performed following ASTM D2344M. In the pull-off

test, two different failure modes were identified: Interface failure or substrate failure in tension. Average

pull-off strength was 2.84 MPa. Design values were suggested based on current test results of the AF-

FRCM.

Keywords: FRCM; aramid fiber; green high-strength mortar; tensile test; composite short beam; pull-

off test.

1 Introduction

Fabric-reinforced cementitious matrix (FRCM)

system is a relatively new technology in the area of

strengthening and repair of RC and masonry struc-

tures. Externally-bonded FRCM system typically

consists of one or more layers of 2D or 3D fiber fab-

ric and the cementitious matrix in which the fiber

fabric is embedded. The performance of the FRCM

system at elevated temperatures is significantly en-

hanced compared to that of the external fiber rein-

forced polymer (FRP) strengthening as the fiber fab-

ric is protected in the inorganic cementitious matrix

[1]. The FRCM technology is applicable to wet sur-

face. As the fibers are not directly exposed to outdoor

environment, the fiber fabric is prevented from out-

door weathering such as ultra violet exposure. The

FRCM technology has originally evolved from the

ferrocement, where the metallic reinforcement is re-

placed by fabrics of dry fibers [2].

Research performed to define mechanical prop-

erties of the FRCM system and develop more effi-

cient technology to strengthen RC members have

been active during the last decade. Existing studies

on the mechanical characterization of the FRCM sys-

tem include tensile behavior of the FRCM, bond

and/or pull-off behavior of FRCM bonded to con-

crete, and interlaminar shear behavior [3-10]. Many

researchers concentrated on the behavior of the

FRCM-strengthened RC members such as flexural

Munkhtuvshin Ochirbud is a M.S. student of the School of Ar-

chitecture, Hankyong National University, South Korea.

Corresponding author Donguk Choi is a Professor in the

School of Architecture & Design Convergence, Hankyong Na-

tional University, South Korea.

Undram Naidanjav is a M.S. student of the School of Architec-

ture, Hankyong National University, South Korea.

S.-S. Ha is a Professor of the Divison of Real Estate and Con-

strucion Engineering, Kangnam University, South Korea.

C.-Y. Lee is a Principal Researcher and CEO of CareCon, Ltd,

South Korea.

Journal of Asian Concrete Federation, Vol. 6, No. 2, December 2020

2

and shear strengthening of RC beams as well as con-

finement of concrete columns [11-19]. Ebead and El-

Sherif studied the behavior of RC beams strength-

ened in flexure using near surface embedded-FRCM

[20]. It is noted that the FRCM can be an efficient

technology to repair historic structures such as

arches and vault as it is applicable on the curved sur-

face [21].

Although the performance of many different fi-

bers has been investigated such as carbon fiber, glass

fiber, basalt fiber, or polyparaphenylene benzobisox-

asole (PBO) fiber, there are still few published re-

sults of research on the mechanical characterization

of aramid fiber fabric embedded in cementitious ma-

trix. Caggegi et al. investigated the tensile properties

of uniaxial aramid textile and quadriaxial aramid tex-

tile fabric [22]. Due to dense textile of aramid fibers,

cementitious matrix delaminated during tensile test-

ing and hence low fiber stress of 1,089 MPa at ulti-

mate was reported for the uniaxial aramid textile.

Higher ultimate strength of the aramid fiber of 1,354

MPa was shown by the quadriaxial textile where the

aramid fabric slippage was observed from the ce-

mentitious matrix. In another investigation by

Ascione et al., coated bidirectional glass-aramid fi-

ber textile was tested [23]. The glass-aramid textile

showed 1,784 MPa strength of the fiber in tension at

ultimate. In this test, due to use of hydraulic grip of

the UTM which restrained the free slippage of the

fabric from the cementitious matrix during tensile

test, the final failure mode was rupture of the glass-

aramid fibers at 2.02% strain.

In this study, aramid fiber (AF) was used in a

form of 2D fabric (AF mesh) along with a green

high-strength mortar utilizing recycled materials

such as finely ground waste glass powder and recy-

cled fine aggregate. Three different experimental

programs were carried out: Uniaxial test of FRCM

tensile specimens, flexural test of composite short

beams made of double FRCM layers following

ASTM D2344M, and pull-off test using FRCM-con-

crete blocks [24]. The purpose was to provide basic

design parameters and the rational mechanical char-

acterization of the AF-FRCM system. It is noted that

the cementitious matrix (mortar) used in this study is

of significantly higher strength than low-to-medium

strength mortars often used for FRCM [8]. Use of a

high-strength mortar was deemed necessary to in-

crease the tensile capacity of the AF-FRCM system.

As the binder content increases with use of the high-

strength mortar, fine waste glass powder was used to

partially replace cement. Recycled fine aggregate

was also used to replace natural sand, considering

economical aspect of the AF-FRCM system as well

as to promote sustainability.

2 Experimental program

2.1 Materials

2.1.1 Aramid fiber mesh

Aramid fiber roving (1100 Dtex) was used to

fabricate the AF mesh in the laboratory where the

warp spacing and the weft spacing was 11.25 mm

and 22.5 mm on center, respectively, as shown in Fig.

1. Longitudinal and transverse fibers were bonded

together using an adhesive at each junction. Table 1

shows the physical and mechanical properties of the

AF roving determined in this study following ISO

10406-2 [25]. Thickness of a fiber roving was about

1.86 mm and AF covered about 24% of the surface

(i.e. surface area of AF to gross area ≅ 24%). The

AF roving has 2,331 MPa tensile strength at 3.74%

strain as shown in Table 1. The AF demonstrated a

linearly elastic stress-strain relationship until failure

in tension.

Fig. 1 – AF mesh


3

Table 1 – Physical and mechanical properties of AF roving

Cross-sectional area (mm2) 0.737

Density (g/mm3) 0.00144

Strength, ffu (MPa) 2,331

Ultimate strain, εfu 0.0374

Elastic modulus, Ef (GPa) 61.4

Coefficient of thermal expansion (m/m/°C) -2.4 × 10-6

NOTE: Results of 12 tensile tests of AF roving, where density and coefficient of thermal expansion were supplied by

manufacturer.

2.1.2 Green high-strength cementitious matrix

A high-strength cementitious matrix for the

FRCM system was necessary because the FRCM

technology was to be used for the purpose of

strengthening RC members after the current phase of

study. As the material cost of the high-strength ce-

mentitious matrix increases due to use of silica fume

(SF) and siliceous sand in large quantities, it was de-

termined to introduce recycled materials as constitu-

ents such as fine waste glass powder (WGP) and re-

cycled fine aggregate (RFA) [27]. While there are

several types of the waste glass powders used in Ko-

rea such as those produced from LCD waste glass,

green or brown glass bottles, plate glass, the WGP

used in this study was manufactured by crushing and

grinding waste green bottles. The finely ground

WGP (< 50 μm) is an amorphous material with more

than 78% silica content. Table 2 shows the mix de-

sign of the green high-strength mortar used in this

study, where WGP partly replaces SF in Table 2.

Sand used was 100% RFA produced from demol-

ished concrete. The wet-processed RFA was sup-

plied by a commercial producer. The maximum par-

ticle size was 2.5 mm. Figure 2 shows gradation of

cement, SF, WGP, and RFA. Polycarboxylic acid

base superplasticizer (SP) by 1 wt.% of total binder

was used to control flow. Flow of the fresh mortar

measured following KS L 5105 was 24.2 cm as

shown in Table 2 [28]. The compressive strength of

the mortar was determined by testing three 50 mm

× 50 mm × 50 mm cubes at 7, 28, and 56 days,

respectively. Flexural strength was determined by

testing two 40 mm × 40 mm × 160 mm beams

under three points loading in flexure 28 days after

casting. Density and voids of the hardened mortar

were determined following ASTM D642 after 28

days [29].

RFA produced from multi-stage crushing and

sieving process of demolished concrete and satisfy-

ing requirements of KS F 2527 was used [30]. When

the RFA replaces the natural sand, due to adhered

mortar of the recycled aggregate particles, which ab-

sorbs and releases water easily, the mortar with RFA

can experience higher shrinkage than the mortar with

natural sand [31]. Shrinkage behavior of the green

mortar was monitored for 90 days. Three 40 mm x

40 mm x 160 mm mortar bars were cast. After

demolding on the next day, the mortar bars were im-

mediately brought to an environmental chamber

where the relative humidity and temperature were

maintained at 60±5% and 20±2°C, respectively.

Length change of each mortar bar was monitored us-

ing a dial gauge with 0.05 mm accuracy.

Fig. 2 – Gradation of cement, silica fume, waste glass powder and recycled sand


4

Table 2 – Mix design for green high-strength mortar and flow of fresh mortar

C

(kg/m3)

SF

(kg/m3)

WGP

(kg/m3)

RFA

(kg/m3)

W

(kg/m3) w/b

SP

(kg/m3)

Flow

(cm)

550 27.5 110 1,487 206 0.3 6.88 24.2

NOTE: Density of C, SF, WGP, RFA is 3.15, 2.2, 2.5, 2.47, respectively; water absorption of RFA is 2.34% in oven-

dry state.

2.2 Test method

2.2.1 Tensile test of FRCM

A thin FRCM plate (width = 405 mm, height =

450 mm, thickness ≈ 10 mm) was fabricated. AF

mesh that consisted of one fiber roving at 11.25-mm

center-to-center spacing (warp) and two fiber

rovings at 22.5-mm spacing on center (weft) was

first installed at half-depth of the 10-mm-thick plate,

and then the green high-strength mortar was cast to

full depth. The thin plate was wet cured until test date.

Nine 45-mm-wide and 450-mm-long FRCM flat

bars were cut from the plate using waterjet 28 days

after casting. Two specimens cut from both sides of

the plate were used for preliminary tests, and there-

fore a total of seven tensile test specimens was re-

tained for the tensile tests. The tensile test started 56

days after casting. One week prior to testing, a set of

two rectangular 6-mm-thick steel plates was bonded

to each end of the tensile test specimen on both sides

using two-part epoxy. The behavior of 120-mm

length in the middle was monitored during the tensile

test. The tensile test was performed under displace-

ment control at a rate of 1 mm per minute using In-

stron 4495 universal testing machine (UTM). Figure

3 schematically shows the specimen fabrication

method and the testing configuration. A set of exten-

someter with 100-mm gauge length was used to

measure longitudinal strains developed in the mid-

part of a specimen as shown in Fig. 3(b). Test data

were electronically monitored and saved using a

TDS 530 data logger and a notebook PC.

2.2.2 Flexural test of composite short beams

The flexural test of composite short beams (or

interlaminar shear test) specimens were fabricated

following ASTM D2344M, which recommends that

[24]:

(a) Specimen length = thickness × 6

(b) Specimen width = thickness × 2

To fabricate flexural test specimens, a mold for

20-mm-thick FRCM plate (width = 405 mm, height

= 225 mm) was prepared. Two layers of AF mesh

was installed in the mold, where the first and the sec-

ond AF mesh plane was at 5-mm and 15-mm depth,

respectively. Green high-strength mortar was cast in

two shifts of equal thickness of about 10 mm each.

The bottom layer was cast first followed by the top

layer three days after casting the bottom layer with-

out any surface treatment. Nine 45-mm-wide and

225-mm-long double-layered beams were cut out

from the plate by waterjet 28 days after casting the

first layer. 52.5-mm length from each end of the

beam was cut off using a masonry saw. Two beams

cut from both sides of the plate were used for the pre-

liminary tests. Finally, seven composite short beams

(a) Specimen fabrication and setup for tensile test (unit: mm) (b) Test in progress

Fig. 3 – Tensile test


5

(a) Schematics of specimen fabrication (unit: mm) (b) Test under progress

Fig. 4 – Flexural test of composite short beams

were used toward the flexural tests: Each beam was

about 120 mm long, 45 mm wide, and consisted of

double layers of FRCM with equal depth of 10 mm

each: i.e., total thickness was about 20 mm (See Fig.

4). The composite beams were tested under three

points loading.

The short-beam strength was determined using

Eq. (1) as suggested by ASTM D2344M [24]:

𝐹𝑠𝑏𝑠 = 0.75 × 𝑃𝑚𝑎𝑥

𝑏 x ℎ (1)

where Fsbs is short-beam strength (MPa), Pmax is

maximum load observed during test (N), b = meas-

ured specimen width (mm), and h is measured spec-

imen thickness (mm).

Figure 4 shows schematics of the specimen fab-

rication and the flexural test setup for composite

short beams. The flexural test was performed under

displacement control at a ramp rate of 1 mm per mi-

nute using a 50-kN UTM. Test data were electroni-

cally monitored and saved. Load, displacement, fail-

ure modes, and crack patterns were carefully ob-

served during and after test.

2.2.3 Pull-off test

For the pull-off test, a 500 mm × 500 mm ×

100 mm normal strength concrete block (fc’ = 30

MPa) was first cast. After 28 days, top surface of the

concrete block was roughened using two different

methods: sand blasting and shot blasting. The sand

patch method following ASTM STP763 was used to

measure the roughness, which revealed that the av-

erage depth of the roughened surface was 0.69 mm

and 0.76 mm, respectively, for the sand blasting and

(a) AF mesh installed on top of concrete block prior

to mortar casting

(b) Pull-off test in progress

Fig. 5 – Pull-off test


6

shot blasting [32]. After the sand/shot blasting was

completed, 3.5-mm thick, 50-mm wide, and 500-mm

long hardwood sections (wood plates) were bonded

along perimeters on top of the concrete block using

adhesive. AF mesh was manually installed using the

AF rovings at 11.25-mm center-to-center spacing in

one direction and at 22.5-mm spacing on center in

the other direction. After the AF mesh was installed,

the second layer of 3.5-mm thick hardwood sections

were bonded on top of the existing hardwood sec-

tions using adhesive so that the total height of the

two-layers of hardwood sections plus the AF mesh

(at mid height in between the two hardwood layers)

altogether was about 8 mm (See Fig. 5(a)). Then

green high-strength mortar was cast in one shift. The

concrete block with FRCM overlay was consolidated

using a vibrating table. The FRCM system was wet

cured until test which commenced after 28 days. One

week before testing, using a hand grinder, the FRCM

overlaid on the concrete block was cut in 45 mm x

45 mm grid pattern as shown in Fig. 5(b). Depth of

the cutting was about 25 mm. A steel end plate (40

mm x 40 mm) was bonded to the top surface of the

FRCM using two-part epoxy. A pull-off testing ap-

paratus was connected to the steel end plate after al-

lowing the epoxy to develop full strength for one

week. To avoid any possible interference between

adjacent pull-off tests, a checker board testing pat-

tern was adopted as shown in Fig. 5(b). The maxi-

mum load was recorded and the failure mode was

observed and recorded.

3 Test results

3.1 Properties of green cementitious matrix

When tests for the mechanical characterization

started at 56 days, average compressive strength of

three cubes was 75.6 MPa. Average flexural strength

determined from flexural test of two mortar bars was

5.50 MPa at 28 days. Density and voids of the hard-

ened mortar were determined following ASTM

D642 after 28 days as shown in Table 3 [29]. It is

noted that the water-to-binder (w/b) ratio of 0.3

shown in Table 3 does not include free water availa-

ble from superplasticizer (the sp consists of 30%

solid and 70% water by wt.). Including the content

of free water available by addition of the sp, the ef-

fective water-to-bonder ratio is 0.365, which ex-

plains a relatively large voids of 17.6% in Table 3.

Figure 6 shows the measured total shrinkage of the

three shrinkage test specimens. Figure 6 shows some

scatter in the measured total shrinkage versus time

curves. Most of the measured shrinkage for 90 days

occurred during the first three weeks after casting

which is 80.3%, 83.7%, and 85.0% of the 90-day

shrinkage, respectively, for each specimen after

three weeks. The average total shrinkage strain is

717 μm/m after 90 days.

ACI 209R (1992) technical report suggests an

ultimate shrinkage value of 780 μm/m with a correc-

tion coefficient of 0.72 for the shrinkage estimation

after 3 months, which results in a shrinkage of 562

μm/m after three months [33]. Current data sug-

gested that the shrinkage increase by using 100% re-

cycled sand can be as much as 27.5% compared to a

theoretical estimation.

3.2 Tensile test results

The tensile test of seven FRCM tensile speci-

mens was performed using an Instron UTM under

displacement control. While the load was monitored

from load cell of the UTM, the displacement was

measured using a set of 50-mm linear variable dis-

placement transducers (LVDTs), which monitored

movement of the UTM cross head. In addition, a set

of extensometers with 100-mm gauge length was at-

tached to the mid-part of the specimen to measure

tensile strains of the specimen. The tensile test re-

sults are summarized in Table 4. Figure 7 shows the

fiber stress-versus-strain plots of all tensile tests.

Fig. 6 – Total shrinkage of green high-strength mortar


7

Table 3 – Properties of hardened mortar

Compressive strength (MPa) Flexural

strength

(MPa)

Density

(kg/m3)

Water ab-

sorption

(%)

Voids

(%) 7d 28d 56d

55.8 71.7 75.6 5.50 2,280 8.37 17.6

Table 4 – Summary of tensile test results

Index w

(mm)

t

(mm)

Pcr

(kN)

Pmax

(kN)

ff

@ Pmax

(MPa)

Displ.

@ Pcr

(mm)

Displ.

@ Pmax

(mm)

Average

strain

@ Pmax

No. of cracks/

average crack

spacing (mm)

AR-T-1 44.05 9.65 0.49 3.08 1,393 0.18 4.85 0.0498 2/45

AR-T-2 44.05 10.55 0.64 3.01 1,361 0.41 5.85 0.0541 3/45

AR-T-3 43.85 11.05 0.73 3.02 1,366 0.17 4.67 0.0400 4/30

AR-T-4 44.15 11.35 0.89 2.71 1,226 0.36 3.36 0.0309 4/30

AR-T-5 43.85 11.05 0.94 2.99 1,352 0.10 4.71 0.0453 3/45

AR-T-6 44.05 10.65 0.52 3.20 1,447 0.17 5.28 0.0509 3/45

AR-T-7 43.95 10.15 0.70 2.90 1,312 0.37 5.13 0.0544 4/30

Average 0.70 2.99 1,351 0.25 4.84 0.0465

Standard deviation 0.17 0.15 69.1 0.12 0.77 0.0085

Coefficient of variation 0.24 0.05 0.05 0.49 0.16 0.18

NOTE: w – width of specimen; t – thickness of specimen; Pcr – cracking load; Pmax – maximum load; average strain =

displacement of LVDT divided by 120 mm; no. of cracks – number of tensile cracks developed in the middle 120-mm

length of a specimen

Fig. 7 – Tensile test results: load versus displacement

As the load increased, the first tensile crack ap-

peared within the mid 120-mm length of a test spec-

imen. At the first cracking load Pcr, the correspond-

ing tensile stress of the FRCM composite cross-sec-

tion (i.e. nominal stress) was 1.5 MPa at 0.21% av-

erage tensile strain (average of seven tests). As the

load further increased, the number of tensile cracks

increased. After the number of cracks reached the

maximum (2-4 cracks), no new cracks appeared but

the existing cracks widened. The tensile cracks typi-

cally appeared at position of the lateral fiber roving

(weft). Close to the peak load, significant slip of the

AF from cementitious matrix was observed adjacent

to wide tensile cracks. Test ended when the displace-

ment was 6 mm or greater. In general, the tensile test

results indicated a relatively ductile behavior with re-

sistance equal to 6.4 MPa in tension (nominal stress)

and 4.6% tensile strain at the peak (average of seven

tests). Due to a set of stiff steel plates adhered to both

ends of the tensile specimen, the displacement (or

strain) of a tensile specimen was dominated by the

tensile behavior in the mid part of the specimen (i.e.

120-mm length, See Fig. 3). In Table 4 and Fig. 7,

the AF tensile stress at the peak load is 1,351 MPa

(average of seven tests) which is only 58% of the fi-

ber tensile strength ffu (See Table 1). The full AF

strength in tension was not realized in the current


8

AF-FRCM system. The average strain at the peak of

4.6% (extensometer) was significantly greater than

the AF rupture strain of 3.74%, which indicated that

the AF mesh slipped from the cementitious matrix

close to the peak load. Figure 9(a) shows load-ver-

sus-displacement plots determined from LVDT read-

ings while Fig. 9(b) shows load-versus-strain read-

ings determined from extensometer. The strain val-

ues in Fig. 9(b) are smaller than the strain values de-

termined from measured displacements shown in

Fig. 9(a) because tensile cracks often developed out-

side the 100-mm gauge length of the extensometer.

As a result, the displacement readings from LVDTs

were used to determine average strains shown in Ta-

ble 4.

3.3 Results of flexural test of composite short

beams

Flexural test was performed on seven composite

short beams under three point loading. Beam length

(measured from support-to-support) was 100 mm for

all beams. The flexural test results are summarized

in Table 5. Figure 10 shows the cracks developed in

a composite beam after test (AR-IS-5). As shown in

Table 5 and Fig. 10, all beams show flexural failure

rather than shear failure mode.

(a) Before test

(b) After cracking/before peak

load

(c) At peak load

(d) Face A after test (e) Face B after test

Fig. 8 – A tensile test and test results: AR-T-1

Load-displacement (LVDT) Load-strain (extensometer)

Fig. 9 – Tensile test results of AR-T-5


9

In all short beams, one or two flexural crack(s)

occurred close to the mid-span. Crack(s) gradually

progressed beyond the interface and inclined toward

the loading point. At ultimate, the concrete crushing

at the top compression fiber was observed resulting

in a flexural failure. Figure 11 shows the load-ver-

sus-mid-span deflection plots of all flexural test

specimens. As shown in Fig. 11, the load resisting

capacity of the composite beams does not signifi-

cantly drop after the peak is reached. The flexural

test was terminated when the displacement at center

was 5 mm or greater.

In Table 5 and Figure 11, the peak loads range

between 1.30 kN and 1.51 kN. The peak loads are

reached at mid-span displacement between 1.34 mm

and 4.46 mm. The short beam strength in Table 5 was

determined using Eq. (1). The average short beam

strength is 1.30 MPa and the strength ranges between

1.24 MPa and 1.38 MPa.

Fig. 10 – Cracks developed after flexural test: AR-IS-5

Fig. 11 – Load vs. displacement: flexural test of composite short beams

Table 5 – Summary of flexural test results

Index

Short beam dimensions Shear span

Ratio (a/h)

Pmax

(kN)

Displ.

at Pmax

(mm)

Short beam

strength

(MPa)

Failure mode L

(mm)

h

(mm)

w

(mm)

AR1-IS-1 115.1 18.55 43.85 2.70 1.48 2.56 1.37 flexural failure







Average 1.40 2.55 1.30

Standard deviation 0.07 1.05 0.05

Coefficient of variation 0.05 0.41 0.04

NOTE: L – length of beam; h – height of beam; w – width of beam; a – shear span; Pmax – maximum load; Displ. at Pmax

– beam mid-span displacement at Pmax.


10

3.4 Pull-off test results

The pull-off test was performed using a pull-off

testing apparatus as shown in Fig. 5(b). Test results

are summarized in Table 6 in terms of peak load,

pull-off strength, and failure mode. A total of twelve

tests was performed on the shot-blasted and the sand-

blasted interfaces, respectively. From the shot-

blasted interface, four failures occurred at the inter-

face between the FRCM and concrete (average pull-

off strength = 3.25 MPa) while six failures occurred

in existing concrete below the interface or substrate

(average pull-off strength = 2.99 MPa). From the

sand-blasted interface, four failures occurred at the

interface (average pull-off strength = 1.62 MPa)

while eight failures occurred in the substrate (aver-

age pull-off strength = 3.15 MPa). Overall, the pull-

off strength of the shot-blasted interface (3.09 MPa)

is greater than that of the sand-blasted interface (2.64

MPa). From all pull-off tests, the average pull-off

strength is 2.84 MPa.

4 Discussions

The volumetric ratio of the AF mesh to gross

volume was 1.3% in this study (or 0.65% in the axial

direction). The fiber amount was sufficient so that

the fiber tensile behavior governed the tensile behav-

ior of the FRCM system. At Pmax of 2.99 kN (average

of seven tests), the nominal tensile resistance (i.e.

Pmax divided by the gross section) of the FRCM com-

posite section was 6.4 MPa. A characteristic strength

(or a design value) can be defined as the average

value minus one standard deviation (ACI 549.4R).

The peak load was reached at an average strain of

4.65% with standard deviation of 0.85%, resulting in

characteristic strain at the peak of 3.8%. Overall the

tensile behavior was ductile with relatively good ca-

pacity of deformation accompanied by multiple

number of clearly visible cracks (two to four cracks

over 120-mm length with average crack spacing be-

tween 30 mm and 45 mm, See Table 4).

(a) Interface failure (b) Substrate failure

Fig. 12 – Failure modes determined from pull-off test

Table 6 – Summary of pull-off test results

Shot-blasted interface Sand-blasted interface

No. Pmax (kN) Stress (MPa) Failed at No. Pmax (kN) Stress (MPa) Failed at

1 4.80 2.37 substrate 1 2.10 1.04 interface

2 6.35 3.14 substrate 2 6.33 3.13 substrate

3 n/a n/a n/a 3 2.12 1.05 interface


5 4.92 2.43 interface 5 6.67 3.29 interface

6 4.51 2.23 substrate 6 2.25 1.11 interface


8 8.65 4.27 interface 8 3.27 1.62 substrate


10 n/a n/a n/a 10 5.86 2.90 substrate



average 3.09 average 2.64

standard deviation 0.62 standard deviation 1.14

cov 0.20 cov 0.43

NOTE: n/a - test results are not available due to adhesive failure.


11

Table 7 – Design value (characteristic value = average minus one standard deviation, ACI 549.4R-13)

εfk

(%)

ffk

(MPa)

Efk

(MPa)

fisk

(MPa)

fbk

(MPa)

3.8 1,282 302 1.25 1.89

In Table 4, the AF tensile stress at the peak load

is 1,351 MPa (average of seven tests) with a standard

deviation of 69.1 MPa, which results in a character-

istic strength of 1,282 MPa. Only 58% of the full AF

strength in tension could be realized in the current

AF-FRCM system due to slip of AF from the ce-

mentitious matrix. Average strain at the peak is

46,480 μm/m with a standard deviation of 8,530

μm/m, which results in a characteristic value of

37,950 μm/m. From tensile test results, tensile mod-

ulus of elasticity of the cracked specimen can be

evaluated from slope of the line that connects two

points at stress levels equal to 0.90 ffu and 0.60 ffu as

shown in Eq. (2) as suggested by ACI549.4R-13 [2]:

𝐸𝑓 =∆𝑓

∆𝜀=

0.9𝑓𝑓𝑢 − 0.6𝑓𝑓𝑢

𝜀𝑓,09𝑓𝑓𝑢− 𝜀𝑓,06𝑓𝑓𝑢

(2)

From test results in Table 4 and Fig. 7, Ef is 302

MPa (Average tensile modulus of elasticity of seven

tests). It is noted that, unlike other design parameters,

an average value was used for the tensile modulus of

elasticity.

The interlaminar shear strength of a FRCM sys-

tem should be considered in case of shear strength-

ening and the flexural strengthening using the FRCM

technology. In this study, the average short beam

strength determined using Eq. (2) was 1.30 MPa with

a standard deviation of 0.05 MPa, which resulted in

1.25 MPa characteristic strength.

The pull-off test results showed two different

failure modes: interface failure in tension and tensile

failure in existing concrete (i.e. substrate tensile fail-

ure). Overall, the interface strength of the shot-

blasted interface was greater than the interface

strength of the sand-blasted interface. From all pull-

off tests, the average pull-off strength was 2.84 MPa

with standard deviation of 0.95 MPa, which results

in a design strength of 1.89 MPa. The design values

(characteristic strengths) are summarized in Table 7.

5 Conclusions

The research objective was to provide basic de-

sign parameters through mechanical characterization

of the aramid fiber mesh-green cementitious matrix

FRCM system used in this study. Three different

types of tests were conducted: Uniaxial test of tensile

specimens; flexural test of composite short beams;

and pull-off test of FRCM placed on top of normal

strength concrete.

(1) The fiber amount (volumetric ratio of the axial

AF to gross volume = 0.65%) was adequate

such that the fiber tensile behavior governed the

tensile behavior of the FRCM system at the

peak load.

(2) Relatively ductile tensile behavior with re-

sistance equal to 6.4 MPa in tension (nominal

stress) and 4.6% tensile strain at the peak was

observed (average of seven tests).

(3) The aramid fiber mesh slipped from the ce-

mentitious matrix close to the peak load. This

was evidenced by the fact that, although the av-

erage strain measured at the peak load was 4.6%,

the aramid fibers developed only 58% of the

tensile strength at the peak.

(4) The fiber slip could have occurred in two differ-

ent ways: Slip of internal fiber filaments from

the outer fiber filaments in an aramid fiber rov-

ing; and slip of the aramid fiber roving from ce-

mentitious matrix.

(5) In the flexural test of composite short beams, all

failure mode was flexural failure rather than

shear failure (such as interface delamination or

debonding in the fiber mesh plane). The average

short beam strength of seven tests was 1.3 MPa.

It is suggested that the interlaminar shear

strength is 1.3 MPa or greater.

(6) In the pull-off test, two different failure modes

were identified: Interface tensile failure (inter-

face failure) and tensile failure of the existing

concrete (substrate failure). Overall, an average

pull-off strength of 2.84 MPa was determined

(average of 22 tests).

(7) Design parameters for the AF-FRCM system

used in this study were suggested. The elastic

modulus is low in the AF-FRCM system indi-

cating significant slip of the fibers in the ce-

mentitious matrix.

(8) Green high-strength cementitious matrix used

in this study developed 75.6 MPa strength after

56 days and developed total shrinkage of 717

micro strain after 90 days. The mortar properties

are considered proper for the use in the AF-

FRCM system while it is a more economical

mix than the conventional mix


12

Acknowledgement

This research was supported by a grant

(20CTAP-C152175-02) from Technology Advance-

ment Research Program (TARP) funded by the Min-

istry of Land, Infrastructure, and Transport of the

Korean government

References

1. Firmo, J.P.; Correia, J.R.; and Bisby, L.A. (2015)

“Fire behavior of FRP-strengthened reinforced

concrete structural elements: A state-of-the-art

review,” Composites Part B: Engineering, vol.

80, pp. 198-216.

2. ACI 549.4R (2013) Guide to Design and Con-

struction of Externally-Bonded Fabric-Rein-

forced Cementitious Matrix (FRCM) Systems

for Repair and Strengthening Concrete and Ma-

sonry Structures, American Concrete Institute,

Farmington Hills, MI.

3. Larbi, A.S.; Agbossou, A.; and Hamelin, P.

(2013) “Experimental and numerical investiga-

tion about textile-reinforced concrete and hybrid

solutions for repairing and/or strengthening re-

inforced concrete beams,” Composite Structures,

vol. 99, pp. 152-162.

4. Peled, A.; and Mobasher B. (2007) “Tensile be-

havior of Fabric Cement-Based Composites:

Pultruded and Cast,” ASCE Journal of Materials,

vol. 19, no. 4, pp. 340-348.

5. Caggegi, C.; Lanoye, E.; Djama, K.; Bassil, A.;

and Gabor A. (2017) “Tensile behaviour of a

basalt TRM strengthening system: Influence of

mortar and reinforcing textile ratios,” Compo-

sites Part B: Engineering, vol. 130, pp. 90-102.

6. Domini J.; and Corinaldesi V. (2017) “Mechan-

ical characterization of different FRCM systems

for structural reinforcement,” Construction and

Building Materials, vol. 145, pp. 565-575.

7. Ebead, U.; and Younis, A. (2019) “Pull-off char-

acterization of FRCM/Concrete interface,”

Composites Part B: Engineering, vol. 165, pp.

545-553.

8. Elsanadedy, H.M.; Abbas, H.; Almusallam,

T.H.; and Al-Salloum, Y.A. (2019) “Organic

versus inorganic composites for bond-critical

strengthening applications of RC structures –

State-of-the-art review,” Composites Part B:

Engineering, vol. 174, pp. 106947.

9. Mazzuca, S.; Hadad, H.A.; Ombres, L.; and

Nanni, A. (2019) “Mechanical Characterization

of Steel-Reinforced Grout for Strengthening of

Existing Masonry and Concrete Structures,”

ASCE Journal of Materials, vol. 31, no. 5, pp.

0019037.

10. Zamir, M.; Sripada, R.; and Peled, A. (2019)

“Hybrid fillers in carbon-fabric-reinforced ce-

ment-based composites,” Cement and Con-

crete Composites, vol. 98, pp. 113-124.

11. Dong, J.F.; Wang, Q.Y.; and Guan, Z.W. (2012)

“Structural behavior of RC beams externally

strengthened with FRP sheets under fatigue and

monotonic loading,” Engineering Structures,

vol. 41, pp. 24-33.

12. Baggio, D.; Soudki, K.; and Noel, M. (2014)

“Strengthening of shear critical RC beams with

various FRP systems,” Construction and Build-

ing Materials, vol. 66, pp. 634-644.

13. Bournas, D.A.; Pavese, A.; and Tizani W. (2015)

“Tensile capacity of FRP anchors in connecting

FRP and TRM sheets to concrete,” Engineering

Structures, vol. 82, pp. 72-81.

14. Gonzales-Libreros, J.H.; Sabau, C.; Sneed, L.H.;

Pellegrino, C.; and Sas G. (2017) “State of re-

search on shear strengthening of RC beams with

FRCM composites,” Construction and Building

Materials, vol. 149, pp. 444-458.

15. Al-Gemeel, A.N.; and Zhuge, Y. (2019) “Using

textile reinforced engineered cementitious com-

posite for concrete columns confinement,”

Composite Structures, vol. 210, pp. 695-706.

16. Marcinczak, D.; Trapko, T.; and Musial, M.

(2019) “Shear strengthening of reinforced con-

crete beams with PBO-FRCM composites with

anchorage,” Composites Part B: Engineering,

vol. 158, pp. 149-161.

17. Hadad, H.A.; Erickson, B.; and Nanni, A. (2020)

“Flexural analysis and design of FRCM-

strengthened RC beams,” Construction and

Building Materials, vol. 244, pp. 118371.

18. Raoof, S.M.; and Bournas, D.A. (2017) “TRM

versus FRP in flexural strengthening of RC

beams: behavior at high temperatures,” Con-

struction and Building Materials, vol. 154, pp.

424-437.

19. Messori, M.; Nobili, A.; Signorini, C.; and Sola

A. (2019) “Effect of high temperature exposure

on epoxy-coated glass textile reinforced mortar

(GTRM) composites,” Construction and Build-


20. Ebead, U.; and El-Sherif, H. (2019) “Near sur-

face embedded-FRCM for flexural strengthen-

ing of reinforced concrete beams,” Construction

and Building Materials, vol. 204, pp. 166-176.

21. Carozzi, F.G.; Poggi, C.; Bertolesi, E.; Milani,

G. (2018) “Ancient masonry arches and vaults

strengthened with TRM, SRG and FRP compo-

sites: Experimental evaluation,” Composite


22. Caggegi, C.; Carrozi, F.G.; Santis, S.D.; Fab-

brocino, F.; Focacci, F.; Hojdys, L.; Lanoye, E.;

and Zuccarino, L. (2017) “Experimental analy-

sis on tensile and bond properties of PBO and


13

aramid fabric reinforced cementitious matrix for

strengthening masonry structures,” Composites

Part B: Engineering, vol. 127, pp. 175-195.

23. Ascione, L.; De Felice, G.; and De Santis, S.

(2015) “A qualification method for externally

bonded Fibre Reinforced Cementitious Matrix

(FRCM) strengthening systems,” Composite

Part B: Engineering, vol. 78: pp. 497-506.

24. ASTM D2344M (2016) Standard Test Method

for Short-Beam Strength of Polymer Matrix

Composite Materials and Their Laminates,

ASTM International, West Conshohocken, PA.

25. ISO 10406-2 (2015) Fibre-reinforced polymer

(FRP) reinforcement of concrete – test methods

– Part 2: FRP sheets, Geneva, Switzerland.

26. Choi, D.-U.; Fowler, D.W.; and Wheat, D.L.

(1996) “Thermal Stresses in Polymer Concrete

Overlays,” American Concrete Institute, Special

Publication SP-166, Properties and Uses of Pol-

ymers in Concrete, pp. 93-122.

27. Soliman, N.A.; and Tagnit-Hamou, A. (2017)

“Partial substitution of silica fume with fine

glass powder in UHPC: Filling the micro gap,”

Construction and Building Materials, vol. 139,

pp. 374-383.

28. KS L 5105 (2007) Testing method for compres-

sive strength of hydraulic mortar, Korean

Agency for Technology and Standards (KATS).

29. ASTM D642 (2013) Standard Test Method for

Density, Absorption, and Voids in Hardened

concrete, ASTM International, West Con-

shohocken, PA.

30. KS F 2527 (2018) Concrete aggregates, Korean

Agency for Technology and Standards (KATS).

31. Chinzorigt, G.; Lim, M.K.; Yu, M.; Lee, H.;

Enkbold, O.; and Choi, D. (2020) “Strength,

shrinkage and creep and durability aspects of

concrete including CO2 treated recycled fine ag-

gregate,” Cement and Concrete research, vol.

136, pp. 106062.

32. ASTM STP763, Measuring Surface Texture by

the Sand-Patch Method, ASTM International,

West Conshohocken, PA.

33. ACI 209R (1992) Prediction of Creep, Shrink-

age, and Temperature Effects in Concrete Struc-

tures, American Concrete Institute, Farmington

Hills, MI.



ISSN 2465-7964 / eISSN 2465-7972

https://doi.org/10.18702/acf.2020.12.6.2.14

14

Technical Paper

Influences of nylon fiber geometries and contents on me-

chanical behavior of reinforced mortar

Teeranai Srimahachota*, Haruka Matsuura, Shun Yamaguchi, Hiroshi Yokota

(Received Jul 29, 2020; Revised November 12, 2020; Accepted December 5, 2020; Published December 31, 2020)

Abstract: Used nylon fishing nets were utilized as recycled nylon (RN) short fibers for reinforcing ce-

ment mortar. Fishing nets were cut into specified shapes and lengths, then mixed into mortar. In this

study, the influences of fiber geometries such as diameter, aspect ratio and the fiber content on the me-

chanical properties of mortar were emphasized. Changes in flowability of fresh mortar, compressive

strength, flexural strength, failure behavior, flexural toughness, residual strength factors were experimen-

tally investigated and compared among various mixes. Experimental results indicated that fiber geome-

tries as well as fiber content directly affect the mechanical properties of mortar. Adding fibers was found

to reduce flowability and compressive strength of the mortar. For instance, using sharp-shapes reduced

compressive strength by 41% while using cross-shapes improved flexural strength by 44.5%. Improve-

ment in flexural strength and flexural toughness were found in association with the fiber content. RN

fiber contributes to the post-peak loading capacity and prevents abrupt failure of concrete structures.

Keywords: Recycled nylon fiber; used fishing nets; fiber reinforced mortar; mechanical behavior.

1 Introduction

Abandoned, lost or discarded fishing gears

(ALDFG), particularly fishing nets, in the ocean is

becoming environmental issues. It was estimated

that more than 705,000 tons of ALDFG were lost in

the ocean and more than 100,000 marine lives were

killed by ALDFG annually [1]. ALDFG accounts

more than 46% of the plastics in the Great Pacific

Garbage Patch, and the number of ALDFG is grow-

ing rapidly [2]. Recent studies found that ALDFG

damages many coral reefs by scraping their tissues

[3]. ALDFG can be navigational threats by causing

entanglement of ship’s propeller causing economic

losses [4]. Therefore, there is a demanding issue in

finding suitable recycling solutions for ALDFG to

mitigate environmental impacts.

Modern fishing nets are usually made of very

strong, durable materials such as nylon and high-

density polyethylene (HDPE), which make fishing

nets basically non-biodegradable. Fishing nets can

be utilized into many textile products, such as clothes,

carpets, sunglasses, and accessories [1, 5, 6]. How-

ever, there are still challenges in recycling used fish-

ing nets because the considerable amounts of energy

and resources are required in the recycling process,

and the huge amounts of CO2 are emitted [7].

Synthetic fibers have been widely used as rein-

forcement in cementitious materials as they improve

mechanical properties and durability of concrete [8].

Polypropylene and nylon fibers were found to im-

prove freeze-thaw resistance, splitting tensile

strength, flexural strength of the mortar as well as

prevent spalling of concrete under high temperature.

However, the decrease in workability and compres-

sive strength was reported [9-11]. Polyvinyl alcohol

(PVA) fiber helps improving compressive strength,

tensile strength, and fatigue and freeze-thaw re-

sistance of the structure [12-14]. Nylon fiber also

helped mitigating micro-cracks propagation by the

crack bridging effect [15]. In addition, nylon fiber

reinforced mortar showed outstanding mechanical

properties over the polypropylene due to the better

distribution of fiber in the cement mix [16].

Recently, recycled fibers have drawn the inter-

est of engineers due to the relatively low material

cost and for environmental preservation. Recycled

fibers, such as polyethylene terephthalate (PET) fi-

bers from plastic bottles and recycled nylon (RN) fi-

bers from waste carpets were found to improve both

Corresponding author Teeranai Srimahachota is a PhD Can-

didate in the Graduate School of Engineering, Hokkaido Uni-

versity, Japan.

Haruka Matsuura is a M.S. student in the Graduate School of

Engineering, Hokkaido University, Japan.

Shun Yamaguchi is a M.S. student in the Graduate School of

Engineering, Hokkaido University, Japan.

Hiroshi Yokota is a Professor in the Faculty of Engineering,

Hokkaido University, Japan.


15

mechanical properties and durability of cementitious

materials [17-22]. Recycled fibers showed accepta-

ble alkaline resistivity, which ensures safe applica-

tion for concrete [23-25]. In addition, recycled

HDPE fibers express comparable mechanical perfor-

mance and durability as the new polypropylene fi-

bers in reinforced cement mortar [24]. RN fiber from

used fishing nets was found safely applicable for ce-

mentitious materials without harmful effects [25].

Orasutthikul et al. reported that RN fiber from used

fishing nets improves flexural strength and flexural

toughness as well as contributes to post-peak capac-

ity of the mortar under bending loads [26]. The RN

fiber from used fishing nets showed the comparable

efficiency in reinforcing mortar as of other recycled

fibers.

This research utilized used fishing nets as RN

short fibers for reinforcing cement mortar. The aim

of this study is to investigate the influences of fiber

geometries such as diameter, length and shape as

well as fiber content on the mechanical behavior of

reinforced mortar. Flowability of fresh mortar, com-

pressive strength, flexural strength, failure behavior,

flexural toughness, and residual capacity factors

were experimentally investigated to evaluate the ef-

fectiveness of the reinforcement.

2 Experimental program 2.1 Test specimens

Nylon used fishing nets used in this study were

obtained from local fishermen in Hokkaido. Fishing

nets were washed by soaking in water for 72 hours

and dried indoor under room temperature. RN fibers

were prepared by manually cutting the fishing nets

by hand to control their length and shape. Diameter

of fiber was measured using microscope, and it is

confirmed that no sign of serious deterioration found

on the surface of the fiber (Fig. 1). Three different

nylon waste fishing nets were used in this study.

Type A, type B and type C fibers are the RN fibers

cut from each of waste fishing nets. Only the straight

parts of the net (i.e. the nodes are not included) were

used for RN type A, B and C. For RN type C, other

two configurations of cutting were introduced to

study the effect of the shapes of fibers which are

cross-shapes with a node at the middle (type CS) and

sharp-shapes with 4 nodes at the end of each section

(type CR). Configuration of fibers are shown in Fig.

2.

Fig. 1 – Microscope images of the fibers

Fig. 2 – Types of fibers

(a) Type A (b) Type B (c) Type C

(d) Type CR (e) Type CS


20

Table 1 – Mechanical properties of the fibers

Fiber Diameter

(mm)

Fiber configu-

ration

Tensile strength

(MPa)

Failure strain

(%)

Young’s modulus

(GPa)

Type A 0.24 Straight 344 36.5 0.94

Type B 0.52 Straight 246 19.8 1.24

Type C 0.23 Straight 143 34.2 0.42

Type CR 0.23 Cross shapes 143 34.2 0.42

Type CS 0.23 Sharp shapes 143 34.2 0.42

Table 2 – Properties of cement

Density

(g/cm3)

Specific surface area

(cm2/g)

28 days compressive

strength (MPa)

Ignition

loss %

MgO

(%)

SO3

(%) Cl- (%)

3.16 3340 61.6 2.26 1.41 2.10 0.015

Table 3 – Mix proportion of mortar (kg/m3)

Cement Sand Water Fiber

742 1087 334 11.3

Uniaxial tensile tests following ASTM C1557

[27] were conducted on each type fibers using uni-

versal testing machine (UTM) with the constant

cross-head displacement at 2 mm/min. The proper-

ties of the fibers are given in Table 1. Assuming lin-

ear relationship between stress and strain during the

test, Young’s modulus of fibers were calculated from

the ratio of tensile strength and the failure strain.

2.2 Mix design and casting procedures

Mortar prisms with the dimension of 40 mm ×

40 mm × 160 mm and the mortar cylinders measur-

ing 50 mm in diameter and 100 mm in length were

prepared for the tests. Ordinary Portland cement

(OPC) having the density of 3.16 g/cm3 and the river

sand having the fineness modulus of 2.99 were used

for the mixing. Properties of the OPC are given in

Table 2. The mix proportion of the mortar is pre-

sented in Table 3; the water-to-cement ratio was 0.45.

The density of RN fiber was set at 1.13 g/cm3 accord-

ing to the general value of nylon. The fiber content

by volume was set at 1.0% and 2.0% to avoid the

formation of fiber cluster during the mixing. Details

of the test specimens are given in Table 4. The con-

trol specimen, plain mortar without fiber added, is

named as NF (non-fiber). In addition, the mix with

RN type A and type B fibers at the fiber fraction of

1.0% each was introduced to investigate the com-

bined effect of fiber diameters. The mixes with of

types A and B are named as M-20-1.0 and M-40-1.0

for the length of 20 mm and 40 mm fiber, respec-

tively.

Preparation of mortar specimens was conducted

according to our previous study [26]. Cement and

sand were mixed together for 1 minutes at first, then

RN fibers were slowly added during the mixing. Wa-

ter was subsequently added and mixed for further 2-

3 minutes to avoid fiber cluster. Two prism speci-

mens and three cylinder specimens were casted and

cured in water for 28 days before the tests.

2.3 Testing methods

Compressive tests were performed as per JIS A

1108 [28], and three-point flexural tests were con-

ducted in accordance with JIS R 5201 [29]. Com-

pressive tests and three-point flexural tests were con-

ducted on the cylinder specimens and the prism spec-

imens respectively. Two linear variable differential

transformers (LVDTs) attached on the front and back

sides of the specimen were used to measure vertical

displacement at the midspan of the specimen during

the flexural loading as shown in Fig. 3. The flexural

load was applied with the vertical displacement rate

of 0.05 mm/min until the vertical displacement

reached 2.00 mm.

Fig. 3 – Three-point flexural test setup


17

Table 4 – Details of test specimens

Specimen

name

Diameter of fiber, D

(mm)

Length of fiber, L

(mm)

Aspect ratio of fi-

ber (L/D)

Fiber content by

volume (%)

NF - - - -

A-20-1.0 0.24 20 83 1.0

A-20-2.0 0.24 20 83 2.0

A-40-1.0 0.24 40 167 1.0

A-40-2.0 0.24 40 167 2.0

B-20-1.0 0.52 20 38 1.0

B-20-2.0 0.52 20 38 2.0

B-30-1.0 0.52 30 58 1.0

B-30-2.0 0.52 30 58 2.0

B-40-1.0 0.52 40 77 1.0

B-40-2.0 0.52 40 77 2.0

C-10-1.0 0.23 10 43 1.0

C-10-2.0 0.23 10 43 2.0

M-20-1.0 0.24 + 0.52 20 - 1.0 each

M-40-1.0 0.24 + 0.52 40 - 1.0 each

CR-20-1.0 0.23 20 - 1.0

CR-20-2.0 0.23 20 - 2.0

CS-20-1.0 0.23 20 - 1.0

CS-20-2.0 0.23 20 - 2.0

3 Results and discussions

3.1 Mortar flow

The flow diameter of fresh mortar was meas-

ured in accordance with JIS R 5201 [29], and the re-

sults are given in Table 5. Adding fibers resulted in

the reduction of flow diameter ranging from 2 – 22%.

The reduction was remarkable for type C, which was

11% to 19%. Using fiber with the higher aspect ratio

reduced flowability of fresh mortar in type A and

type B mixes as seen in A-40-1.0 and B-40-1.0 that

have the aspect ratio of 167 and 77, respectively.

However, for type C, regardless of the smaller aspect

ratio, fiber cluster was formed during mixing be-

cause of relatively small diameter. Increase in the fi-

ber fraction to 2.0% caused further reduction in flow

diameter by approximately 1.5 – 2.0 times as ob-

served from the cases of A-20-2.0, C-10-2.0, CS-20-

2.0 and CR-20-2.0. Using fibers with the smaller di-

ameter tended to reduce flowability. Moreover, it

was observed during the mixing that type A and type

C tend to form fiber cluster during the mixing.

Longer fibers (i.e. 40 mm) tend to further reduce the

flow diameter as observed in A-40-2.0 and M-40-1.0.

On the contrary, the length and fiber content of

type B did not show clear influences on the flowabil-

ity. For M-20-1.0 and M-40-1.0, the reduction in

flow diameter seems to have the same tendency as

type A. Fresh mortar with CS fibers showed greater

reduction in flow diameter than that with CR fibers

because the CS fiber has two knots at its ends (see

Fig. 2(e)). Fresh mortar with CR and CS expressed

similar trend to that with type C; therefore, the diam-

eter of fiber shows a greater influence on the flowa-

bility than the shape of fiber.

The fiber geometry had a great influence on the

flowability of fresh mortar. For the same type of fi-

bers, fiber with higher aspect ratio seems to reduce

the flowability of fresh mortar. Using fibers with the

smaller diameter also reduces the flowability of fresh

mortar because more fibers are presenting in the mix

at the same fiber content. In addition, thinner fibers

tend to be tangle together and form fiber cluster dur-

ing the mixing. However, this behavior depends on

the surface characteristics and the stiffness of the fi-

ber, which needs more confirmations in the future.

3.2 Compressive strength

The results from the compressive strength tests

and the three-point flexural tests are summarized in

Table 5. These results were averaged from 3 cylinder

specimens and 2 prism specimens for compressive

and flexural strengths, respectively.

Test results showed that adding fibers reduces

the compressive strength of mortar, especially for

CR and CS mixes. Increasing fiber content from 1%

to 2% causes further reduction of the compressive

strength. It was suggested by Lee et al. (2012) and

Karahan et al. (2011) that adding fibers reduces the

modulus of elasticity and increases air content in ce-

ment matrix [11, 30-31]. Moreover, the reduction in

compressive strength of CR and CS was probably

caused by the knots of fibers, which increases void

in the cement matrix. Reduction in compressive

strength was observed when applying RN fibers


18

Table 5 – Flow and compressive and flexural strengths of tested mortar

Aspect ratio Flow Compressive strength Flexural strength

Specimen d

(mm) d

(%)

f'c

(MPa)

SD

(MPa) f’c

(%)

fb

(MPa)

SD

(MPa) fb

(%)

NF - 254 - 45.1 7.7 - 5.3 0.2

A-20-1.0 83 244 -3.9 48.1 10.8 6.7 5.5 0.8 4.5

A-20-2.0 83 239 -5.9 38.1 4.4 -15.4 5.7 0.5 8.9

A-40-1.0 167 234 -7.9 43.8 14.1 -2.9 6.7 1.2 26.7

A-40-2.0 167 205 -19.3 36.4 4.0 -19.2 5.2 0.0 -2.2

B-20-1.0 38 235 -7.5 42.1 14.1 -6.6 4.9 0.3 -6.7

B-20-2.0 38 240 -5.5 33.0 4.0 -26.7 5.7 0.5 8.9

B-30-1.0 58 238 -6.3 36.0 0.4 -20.2 5.3 0.8 0.0

B-30-2.0 58 244 -3.9 33.1 4.0 -26.7 5.6 0.0 6.7

B-40-1.0 77 249 -2.0 36.9 2.2 -18.2 6.2 0.8 17.8

B-40-2.0 77 238 -6.3 43.3 10.4 -4.0 5.4 0.3 2.2

C-10-1.0 43 226 -11.0 36.3 1.2 -19.4 5.6 0.3 6.7

C-10-2.0 43 206 -18.9 35.3 0.8 -21.8 6.7 0.5 26.7

M-20-1.0 - 247 -2.8 35.9 2.9 -20.3 5.4 0.0 2.2

M-40-1.0 - 210 -17.3 38.0 6.4 -15.7 6.3 0.0 20.0

CR-20-1.0 - 231 -9.1 33.5 0.8 -25.7 6.9 0.5 31.1

CR-20-2.0 - 208 -18.1 30.8 0.7 -31.8 7.6 0.2 44.5

CS-20-1.0 - 224 -11.8 30.6 2.5 -32.0 5.9 0.0 11.1

CS-20-2.0 - 198 -22.0 26.6 1.7 -41.0 5.7 0.8 8.9

Note: d – flow diameter, %d – percent difference in flow diameter compared with NF, f'c – compressive strength, SD –

standard deviation, %f’c – percent difference in compressive strength compared with NF, fb – flexural strength, and %fb

– percent difference in flexural strength compared with NF.

from used fishing nets [25, 26].

Mortar mix with the fibers of lower aspect ratio

showed considerable reduction in compressive

strength as seen from B-20-2.0 and B-30-2.0 that has

the lowest compressive strengths among type A, type

B and type C. However, A-20-1.0, A-40-1.0 and B-

40-2.0 showed relatively less reduction or even in-

crease in compressive strength. This behavior was

found by Ozger et al. that short fiber helps improving

lateral tensile strength of mortar [32].

3.3 Flexural strength

Adding fibers seems to improve flexural

strength of the mortar. However, the tendency is still

unclear. Mortar reinforced with CR type showed

highest flexural strength with the increase of 44% for

CR-20-2.0 compared to the plain mortar (NF). C-10-

2.0 and A-40-1.0 showed the same level of incre-

ment at 27% followed by M-40-1.0 and B-40-1.0.

Flexural strength of the CS mix was lower than that

of CR, but was still higher than most of the type A

and type B that use straight fibers without knots. Ora-

sutthikul et al. [26] explained that the knots at the

ends of fiber can form fiber clusters during mixing;

therefore, fibers were not uniformly distributed.

Some of the mortar mixes, particularly in type

B mixes, showed a comparatively low or even

slightly decreased flexural strength compared to NF.

It is possible that the voids created by the fiber lower

the strength of mortar rather than improve it.

The fiber content and the aspect ratio of fibers

did not show clear trend in the increment of flexural

strength of mortar, and the effects of those parame-

ters cannot be concluded. The contribution of fibers

to flexural strength was found to depend on the sur-

face friction and the bond behavior between fibers

and cement substrate [33]. No breakage of fibers was

observed during the loading tests; however, the

smooth surface of RN fiber may lead to poor bonding

between fibers and the cement substrate. The overall

results showed that adding fibers gives a positive ef-

fect to the flexural strength of mortar.

3.4 Failure behavior

The load-midspan deflection curves from the

three-point bending tests are shown in Fig. 4. All fi-

ber-reinforced specimens expressed ductile failure

whereas plain mortar (NF) showed brittle failure.

The load dropped after the peak load and maintained

post-peak loads in the range of 0 – 0.5 kN and 0.5 –

1.0 kN for the fiber fraction of 1.0% and 2.0% re-

spectively. In addition, a hardening stage was ob-

served in which the load increased slightly after the

peak as observed in type B and type CR (Fig. 4 (d-f,

i)). This behavior indicated that fibers would be able

to transfer the loads through cracks and to prevent

sudden collapse of concrete. This hardening stage

was also found for recycled PET fibers [23, 26] and

HDPE fibers [24].


20

(a) Type A with 20 mm long fiber

(b)Type A with 40 mm long fiber

(c) Type B with 20 mm long fiber

(d) Type B with 30 mm long fiber

(e) Type B with 40 mm long fiber

(f) Mix of type A and type B with

20 and 40 mm long fiber

(g) Type C with 10 mm long fiber (h) Type CR with 20 mm long fiber

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)

Midspan deflection (mm)

A20-1.0 (1)

A20-1.0 (2)

A20-2.0 (1)

A20-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


A40-1.0 (1)

A40-1.0 (2)

A40-2.0 (1)

A40-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


B20-1.0 (1)

B20-1.0 (2)

B20-2.0 (1)

B20-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


B30-1.0 (1)

B30-1.0 (2)

B30-2.0 (1)

B30-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


B40-1.0 (1)

B40-1.0 (2)

B40-2.0 (1)

B40-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


Mix20-1.0 (1)

Mix20-1.0 (2)

Mix40-1.0 (1)

Mix40-1.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


C-10-1.0 (1)

C-10-1.0 (2)

C-10-2.0 (1)

C-10-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


CR-20-1.0 (1)

CR-20-1.0 (2)

CR-20-2.0 (1)

CR-20-2.0 (2)


20

(i) Type CS with 20 mm long fiber

(j) Plain mortar (NF)

Fig. 4 – Load-midspan deflection curves

Increasing fiber content (i.e. fiber fraction to

2.0%) as well as increasing the length of fiber im-

proved the post-peak behavior and reduced the drop

in post-peak loads. This characteristic indicated that

fiber delays the failure as well as prevents sudden

collapse of the structure. The diameter of fiber does

not show a noticeable effect on the first-rack strength;

however, the larger diameter of fiber showed the

higher post-peak loads as indicated in Fig 4 (e). The

increase in the post-peak capacity of type CS con-

firmed that the stresses were transferred by the fibers.

The knots at the ends of fiber was found to improve

bond behavior between fiber and the matrix; thus, the

fiber was elongated rather than being pulled out.

3.5 Flexural toughness and residual strength

factors

Flexural toughness (I5, I10, I20) are defined as

given in Fig. 5 as per ASTM C1018 [34]. They were

calculated from the area under the load-deflection

curve where 𝛿 stands for the first-crack deflection.

The residual strength factors are defined by the fol-

lowing equations:

𝑅5,10 =100

10 − 5 (𝐼10 − 𝐼5) (1)

𝑅10,20 =100

20 − 10 (𝐼20 − 𝐼10) (2)

Table 6 lists flexural toughness and residual

strength factors. The results confirmed that adding

fibers affords the improvement of flexural toughness.

The load application was terminated when the verti-

cal mid-span displacement reaches 2.0 mm in some

mixes. Therefore, 𝐼20 and 𝑅10,20 cannot be calcu-

lated for them. Residual strength factor, 𝑅5,10 ,

seems to be higher for the mix containing fiber with

higher aspect ratio. Increasing in fiber content and

using longer fiber improves flexural toughness of

mortar as seen from A-40-2.0, B-40-2.0, M-40-1.0

and CS-20-2.0 in Fig. 6. Moreover, the residual

strength factor of the 40-mm long fibers is higher

than those of 20-mm long fibers. Similar behavior of

flexural toughness was also found when using recy-

cled PET and PVA fibers [26].

Fig. 5 – Load-deflection curves as defined by

ASTM C 1018

Fig. 6 – Toughness indices of fiber reinforced mortar

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


CS-20-1.0 (1)

CS-20-1.0 (2)

CS-20-2.0 (1)

CS-20-2.0 (2)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0 0.5 1.0 1.5 2.0

Load

(kN

)


PL (2)

PL (1)

𝛿

5.5𝛿

𝛿

10 5𝛿

A

B CD

EFGHO

Deflection

Load

𝐼5 = ( )

( )

𝐼10 = ( )

( )

𝐼20 = ( )

( )


21

Table 6 – Toughness indices and residual strength factors

Specimen Aspect ratio I5 I10 I20 R5,10 R10,20

NF - - - - - -

A-20-1.0 83 1.92 2.23 - 6.2 -

A-20-2.0 83 2.52 3.64 - 22.4 -

A-40-1.0 167 2.36 3.47 6.39 22.2 29.1

A-40-2.0 167 3.73 6.7 - 59.4 -

B-20-1.0 38 1.93 2.35 - 8.4 -

B-20-2.0 38 1.83 2.28 - 9.2 -

B-30-1.0 58 1.88 2.42 - 10.8 -

B-30-2.0 58 2.33 3.89 7.32 31.2 -

B-40-1.0 77 1.56 1.92 - 7.2 -

B-40-2.0 77 3.19 5.83 11.7 52.8 58.7

C-10-1.0 43 1.58 1.64 1.90 1.18 2.61

C-10-2.0 43 2.04 2.58 3.67 10.8 10.8

M-20-1.0 - 2.46 3.26 4.98 16.0 17.2

M-40-1.0 - 3.04 5.16 9.72 42.4 45.6

CR-20-1.0 - 2.04 2.79 - 15.0 -

CR-20-2.0 - 1.89 2.85 - 19.2 -

CS-20-1.0 - 1.95 2.58 4.10 12.6 15.2

CS-20-2.0 - 3.32 5.86 12.71 50.8 68.5

4 Conclusions

Recycled nylon fibers from used fishing nets

were mixed in cement mortar with various fiber con-

figurations and contents. The experimental results

confirmed that fiber geometries as well as the fiber

contents have significant influences on the mechani-

cal properties of mortar, such as flowability, com-

pressive strength, flexural strength, failure behavior,

flexural toughness and residual flexural strength.

From this study, the following conclusions were

drawn:

(1) Adding fibers considerably reduces the flow di-

ameter of fresh mortar in the range of 2 – 22%.

Fibers with higher aspect ratio as well as high

content of fibers greatly reduce the flowability

of mortar. In addition, fiber cluster tends to form

during the mixing when using the small diame-

ter of fibers or fibers with knots (CR and CS).

(2) Significant reduction of compressive strength is

expected with the addition of fibers. Adding fi-

bers that have lower aspect ratio or the fiber

with knots (CR and CS) reduces the compres-

sive strength of mortar up to 41%. The reduc-

tion in compressive strength becomes severe as

the fiber content is increased.

(3) Adding fibers tends to improve flexural strength

of the mortar. However, its influence is still un-

clear. Cross-shapes fiber shows highest perfor-

mance at 45% increment in flexural strength

among all fiber types.

(4) Adding fibers contributes to the post-peak be-

havior in which the beam can retain some loads

after the peak. Increasing in diameter, length

and volume fraction of fiber improves post-

peak capacities. The post-peak load is increased

with the addition of the sharp-shape fiber. Fiber

helps preventing abrupt failure of mortar.

(5) Flexural toughness of the mortar is improved

with the addition of fiber. Increase in fiber con-

tent as well as the length of fiber yields higher

flexural toughness. Using fibers with higher as-

pect ratio also improves the residual strength

factor.

Recycled nylon fibers from waste fishing nets

have potential to be used in cementitious materials.

The addition of fiber causes both positive and nega-

tive effects simultaneously to the mechanical prop-

erties of mortar. Therefore, careful consideration

should be taken before applying recycled nylon fi-

bers. Further studies are still needed to understand

the behavior of recycled fiber reinforced mortar.

Acknowledgments

This research was supported by JSPS Grant-in-

Aid for Scientific Research (B) #17H03293.

References

1. National Geographic. https://www.nationalgeo-

graphic.com (accessed Mar. 23, 2020)


22

2. Lebreton, L.; Slat, B,; Ferrari, F.; Sainte, B. R.;

Aitken, J.; Marthouse, R.; Hajbane, S.; Cunsolo,

S.; Schwarz, A.; Levivier, A.; Noble, K.;

Debeljak, P.; Maral, H.; Schoeneich, R. A.;

Brambini, R.; and Reisser, J. (2018) “Evidence

that the Great Pacific Garbage Patch is rapidly

accumulating plastic,” Scientific Reports, vol. 8,

pp. 1-15.

3. Valderrama, L. B.; Matthews, L.; and Hoeksema,

W. (2018) “Pollution and coral damage caused

by derelict fishing gear on coral reefs around

Koh Tao, Gulf of Thailand,” Marine Pollution

Bulletin, vol. 135, pp. 1107-1116.

4. Hong, S.; Lee, J.; and Lim, S. (2017) “Naviga-

tional threats by derelict fishing gear to navy

ships in the Korean seas,” Marine Pollution Bul-

letin, vol. 119, pp. 100-105.

5. Wang, Y. (2010) “Fiber and textile waste utili-

zation,” Waste and Biomass Valorization, vol. 1,

pp. 135-143.

6. Charter, M.; Carruthers, R.; and Femmer, J. S.

(2018) “Products from waste fishing nets acces-

sories, clothing, footwear, home ware, recrea-

tion,” Circular Ocean, vol. 04, pp. 1-31.

7. National Geographic. https://www.nationalgeo-

graphic.com (accessed Mar. 23, 2020)

8. Zheng, Z.; and Feldman, D. (1995) “Synthetic

fibre-reinforced Concrete,” Progress in Polymer

Science, vol. 20, pp. 185-210.

9. Zhang, P.; and Li, Q. F. (2013) “Effect of poly-

propylene fiber on durability of concrete com-

posite containing fly ash and silica fume,” Com-

posites Part B: Engineering, vol. 45, pp. 1587-

1594.

10. Ramezanianpour, A. A.; Esmaeili, M.; Ghahari,

S. A.; and Najafi, M. H. (2013) “Laboratory

study on the effect of polypropylene fiber on du-

rability, and physical and mechanical character-

istic of concrete for application in sleepers,”

Construction and Building Materials, vol. 44, pp.

411-418.

11. Lee, G.; Han, D.; Han, M. C.; Han, C. G.; and

Son, H. J. (2012) “Combining polypropylene

and nylon fibers to optimize fiber addition for

spalling protection of high-strength concrete,”


313-320.

12. Khan, S. U.; Nuruddin, M. F.; and Shafiq, N.

(2014) “Strength development of concrete in-

corporating metakaolin and PVA fibres,” Ap-

plied Mechanics and Materials, vol. 567, pp.

505-510.

13. Jang, J. G.; Kim, H. K.; Kim, T. S.; Min, B. J.,

and Lee, H. K. (2014) “Improved flexural fa-

tigue resistance of PVA fiber-reinforced con-

crete subjected to freezing and thawing cycles,”


129-135.

14. Ozkan, S.; and Demir, F. (2020) “The hybrid ef-

fects of PVA fiber and basalt fiber on mechani-

cal performance of cost effective hybrid ce-

mentitious composites,” Construction and

Building Materials, vol. 263, 120564.

15. Lee, S. (2019) “Effect of nylon fiber addition on

the performance of recycled aggregate concrete,”

Applied Sciences, vol. 9, pp. 767.

16. Song, P. S.; Hwang, S.; and Sheu, B. C. (2005)

“Strength properties of nylon- and polypropyl-

ene-fiber-reinforced concretes,” Cement and

Concrete Research, vol. 35, pp. 1546-1550.

17. Siddique, R.; Khatib, J.; and Kaur, I. (2008)

“Use of recycled plastic in concrete: A review,”

Waste Management, vol. 28, pp. 1835-1852.

18. Ozger, O. B.; Girardi, F.; Giannuzzi, G. M.; Sa-

lomoni, V. A.; Majorana, C. E.; Fambri, L.; Bal-

dassino, N.; and Di Maggio, R. (2013) “Effect

of nylon fibres on mechanical and thermal prop-

erties of hardened concrete for energy storage

systems,” Materials and Design, vol. 51, pp.

989-997.

19. Qin, Y.; Zhang, X.; and Chai, J. (2019) “Dam-

age performance and compressive behavior of

early-age green concrete with recycled nylon fi-

ber fabric under an axial load,” Construction and


20. Silva, D. A.; Betioli, A. M.; Gleize, P. J. P.; Ro-

man, H. R.; Gómez, L. A.; and Ribeiro, J. L.D.

(2005) “Degradation of recycled PET fibers in

Portland cement-based materials,” Cement and

Concrete Research, vol. 35, pp. 1741-1746.

21. Foti, D. (2013) “Use of recycled waste pet bot-

tles fibers for the reinforcement of concrete,”

Composite Structures, vol. 96, pp. 396-404.

22. Pereira De Oliveira, L. A.; and Castro-Gomes, J.

P. (2011) “Physical and mechanical behaviour

of recycled PET fibre reinforced mortar,” Con-


1712-1717

23. Fraternali, F.; Farina, I.; Polzone, C.; Pagliuca,

E.; and Feo, L. (2013) “On the use of R-PET

strips for the reinforcement of cement mortars,”

Composites Part B: Engineering, vol. 46, pp.

207-210.

24. Pešić, N.; Živanović, S.; Garcia, R.; and Papas-

tergiou, P. (2016) “Mechanical properties of

concrete reinforced with recycled HDPE plastic

fibres,” Construction and Building Materials,

vol. 114, pp. 362-370.

25. Spadea, S.; Farina, I.; Carrafiello, A.; and Fra-

ternali, F. (2015) “Recycled nylon fibers as ce-

ment mortar reinforcement,” Construction and



23

26. Orasutthikul, S.; Unno, D.; and Yokota, H.

(2017) “Effectiveness of recycled nylon fiber

from waste fishing net with respect to fiber rein-

forced mortar,” Construction and Building Ma-

terials, vol. 146, pp. 594-602.

27. ASTM C1557 (2003). “Standard test method for

tensile strength and Young’s modulus of fibers,”

ASTM International, PA.

28. JIS A 1108 (2018). “Method of test for compres-

sive strength of concrete,” Japan Standards As-

sociation, Tokyo. (In Japanese)

29. JIS R 5201 (2015). “Physical testing methods

for cement,” Japan Standards Association, To-

kyo. (In Japanese)

30. Karahan, O.; and Atiş, C. D. (2011) “The dura-

bility properties of polypropylene fiber rein-

forced fly ash concrete,” Materials and Design,

vol. 32, pp. 1044-1049.

31. Palmquist, S. M.; Kintzel, E.; and Andrew, K.

(2011) “Scanning electron microscopy to exam-

ine concrete with carbon nanofibers,” 5th Pan

American Conference for NDT, Cancun, Mex-

ico.

32. Ozger, O.; Girardi, F.; Giannuzzi, G.; Salomoni,

V.; Majorana, C.; Fambri, L.; Baldassino, N.;

and Di Maggio, R. (2013) “Effect of nylon fibres

on mechanical and thermal properties of hard-

ened concrete for energy storage systems,” Con-


989-997.

33. Redon, C.; Li, V. C.; Wu, C.; Hoshiro H.; Saito

T.; and Ogawa, A. (2001) “Measuring and mod-

ifying interface properties of PVA fibers in ECC

matrix,” Journal of Materials in Civil Engineer-

ing, vol. 13, pp. 399-406.

34. ASTM C 1018. “Standard test method for flex-

ural toughness and first-crack strength of fiber

reinforced concrete,” ASTM International, PA.



ISSN 2465-7964 / eISSN 2465-7972

https://doi.org/10.18702/acf.2020.12.6.2.24

24

Technical Paper

Comparison of creep models and experimental verification of

creep coefficients for normal and high strength concrete

P. N. Ojha*, Brijesh Singh, Abhishek Singh, Vikas Patel

(Received May 4, 2020; Revised December 10, 2020; Accepted December 13, 2020; Published December 31, 2020)

Abstract: A concrete structure when subjected to sustained load presents progressive strain over time,

which is associated with the creep phenomenon. The creep characteristic of high strength concrete as-

sumes importance in the back drop of increase in prestressed concrete constructions. The paper covers

the comparison of creep coefficients with different creep models like Bazant’s B-3, ACI, AASHTO, GL-

2000 and FIB model code 2010 for concrete mixes having water to cementitious ratio of 0.47, 0.36, 0.27

and 0.20. The comparison of different models are done for a relative humidity of 60 percent and design

life of 100 years. For comparison of creep coefficient using different models the age at loading are kept

as 7, 28 and 365 days. Thereafter, values are compared with experimentally obtained results of concrete

mixes having water to cementitious ratio of 0.47 and 0.20 for age at loading of 28 days and up to 180

days loading period. Time induced creep strain of high strength concrete is determined using creep rig of

capacity 2000 kN. Creep strains are measured at regular time intervals on concrete designed with water

to cementitious ratio of 0.47 and 0.20 wherein fly ash and silica fume were also used.

Keywords: Creep coefficient; normal strength concrete; high strength concrete; creep model.

1 Introduction

Creep performance is an important index in the

long-term properties of concrete, and the linear com-

pressive creep deformation can reach 1-4 times of the

short-term elasticity compressive deformation.

Therefore, the creep behaviour must be considered

in the design of concrete structures in order to pro-

vide necessary safety and serviceability. For the im-

portant engineering structures, creep experiment of

the specimen, which is made from the same concrete

used in the structures, is the most reliable method.

However, due to the complexity and diversity, there

are not always sufficient condition to carry out creep

experiment, so the empirical formula fitted from the

obtained experimental data is essential [1]. There are

many creep models available internationally, such as

CEB-FIP series models, ACI 209 series models, GL-

2000 model, AASHTO, B3 model, China Academy

of Building Research model, Zhu Bofang model and

Li Chengmu model et al. [2-7]. However, there are

many differences in the influence factors, formula

forms, applicable scope and prediction accuracy of

these models due to limitation of specific experi-

mental condition and the emphasis of different re-

searchers. The correction factor of mixture ratio of

concrete was given in CEB-FIP series models. The

correction factor of collapsibility, sand ratio and air

content were considered in ACI 209 series models.

The correction factor of water cement ratio, cement

content, sand ratio and concrete density was consid-

ered in B3 model. Recent research relates the creep

response to the packing density distributions of cal-

cium silicate-hydrates. At high stress levels, addi-

tional deformation occurs due to the breakdown of

the bond between the cement paste and aggregate

particles [8-15]. Therefore, designers and engineers

need to know the creep properties of concrete and

must be able to take them into account in the struc-

ture analysis. As per IS: 456-2000 [16], creep of con-

crete depends on the constituents of concrete, size of

the member, environmental conditions (humidity

and temperature), stress in the concrete, age at load-

ing and the duration of loading. As long as the

stress in concrete does not exceed one-third of its

characteristic compressive strength, creep may be as-

sumed to be proportional to the stress. High strength

concrete is significantly in use now a days in number

Corresponding author P. N. Ojha is a Joint Director in the Na-

tional Council for Cement & Building Materials, India.

Brijesh Singh is a Group Manager in National Council for Ce-

ment & Building Materials, India.

Abhishek Singh is a Project Engineer in the National Council

for Cement & Building Materials, India.

Vikas Patel is a Project Engineer in National Council for Ce-



25

of concrete structures, the most common applica-

tions being the columns of high rise buildings, long

span bridges, longer spans for beams or fewer beams

for a given span length, offshore structures, etc.

High-strength concrete is a more sensitive material

than normal strength concrete and it must be treated

with care both in design and in construction. The aim

of the paper is to compare the creep coefficients with

different creep models like Bazant’s B-3, ACI,

AASHTO, GL-2000 and FIB model code 2010 for

concrete mixes having water to cementitious ratio of

0.47, 0.36, 0.27 and 0.20. The comparison of differ-

ent models is done for a relative humidity of 60 per-

cent and design life of 100 years. For comparison of

creep coefficient using different models the age at

loading are kept as 7, 28 and 365 days. Thereafter,

values are compared with experimentally obtained

results of concrete mixes having water to cementi-

tious ratio of 0.47 and 0.20 for age at loading of 28

days.

2 Experimental program

2.1 Concrete ingredients:

Crushed aggregate with a maximum nominal

size of 20 mm was used as coarse aggregate and nat-

ural riverbed sand confirming to Zone II as per IS:

383 was used as fine aggregate. Their physical prop-

erties are given in Table 1. The petrographic studies

conducted on coarse aggregate indicated that the ag-

gregate sample is medium grained with a crystalline

texture and partially weathered sample of granite.

The major mineral constituents were quartz, biotite,

plagioclase-feldspar and orthoclase-feldspar. Acces-

sory minerals are calcite, muscovite, tourmaline and

iron oxide. The petrographic studies of fine aggre-

gate indicated that the minerals present in order of

abundance are quartz, orthoclase-feldspar, horn-

blende, biotite, muscovite, microcline-feldspar, gar-

net, plagioclase-feldspar, tourmaline, calcite and

iron oxide. For both the coarse aggregate and fine

aggregate sample the strained quartz percentage and

their Undulatory Extinction Angle (UEA) are within

permissible limits as per IS: 383-2016 (Strain Quartz

percentage less than 20% and Undulatory Extinction

Angle less than 15o). The silt content in fine aggre-

gate as per wet sieving method is 0.70 percent.

Ordinary Portland cement (OPC 53 Grade) with

fly ash and silica fume are used in this study. The

chemical and physical compositions of cement OPC

53 Grade, Properties of fly ash and silica fume are

given in Table 2. Polycarboxylic group-based super-

plasticizer for w/c ratio 0.36, 0.27 and 0.20 and

Naphthalene based for w/c ratio 0.47 complying with

requirements of Indian Standard: 9103 is used

throughout the investigation. Water complying with

requirements of IS: 456-2000 for construction pur-

pose was used. The 3 days, 7 days and 28 days’ com-

pressive strength of cement OPC 53 Grade were

36.00 MPa, 45.50 MPa and 57.50 MPa respectively.

The 28 days’ compressive strength of controlled

sample and sample cast with fly ash was 38.53 MPa

and 31.64 MPa respectively, when testing was done

in accordance with IS: 1727. The 7 days’ compres-

sive strength of controlled sample and sample cast

with silica fume was 12.76 MPa and 14.46 MPa re-

spectively, when testing was done in accordance

with IS: 1727.

2.2 Mix design details

In this study, the four different mixes with w/c

ratio 0.47, 0.36, 0.27 and 0.20 using granite aggre-

gate were selected for studying creep coefficient.

The slump of the fresh concrete was kept in the range

of 75-100 mm. A pre-study was carried out to deter-

mine the optimum superplasticizer dosage for

achieving the desired workability based on the slump

Table 1 – Properties of aggregates

Property Coarse Aggregate

Fine Aggregate 20 mm 10 mm

Specific gravity 2.83 2.83 2.64

Water absorption (%) 0.3 0.3 0.8

Sieve

Analysis

Cumulative Per-

centage

Passing (%)

20mm 98 100 100

10 mm 1 68 100

4.75 mm 0 2 95

2.36 mm 0 0 87

1.18 mm 0 0 68

600 µ 0 0 38

300 µ 0 0 10

150 µ 0 0 2

Pan 0 0 0

Abrasion, Impact & Crushing Value 19, 13, 19 - -

Flakiness % & Elongation % 29, 25 - -


26

Table 2 – Physical, chemical and strength characteristics of cement

Characteristics OPC -53 Grade Silica Fume Fly Ash

Physical Tests:

Fineness (m2/kg) 320.00 22000 403

Soundness Autoclave (%) 00.05 - -

Soundness Le Chatelier (mm) 1.00 - -

Setting Time Initial (min.) & (max.) 170.00 & 220.00 - -

Specific gravity 3.16 2.24 2.2

Chemical Tests:

Loss of Ignition (LOI) (%) 1.50 1.16 -

Silica (SiO2) (%) 20.38 95.02 -

Iron Oxide (Fe2O3) (%) 3.96 0.80 -

Aluminum Oxide (Al2O3) 4.95 - -

Calcium Oxide (CaO) (%) 60.73 - -

Magnesium Oxide (MgO) (%) 4.78 - -

Sulphate (SO3) (%) 2.07 - -

Alkalis (%) Na2O & K2O 0.57 & 0.59 -

Chloride (Cl) (%) 0.04 - -

IR (%) 1.20 - -

Moisture (%) - 0.43 -

Table 3 – Concrete mix design details for study done

W/Cem

Total Ce-

mentitious

Content

[Cement + Fly

ash + Silica

Fume]

(kg/m3)

Water

Con-

tent

(kg/m3)

Admix-

ture %

by

weight of

Cement

Fine

Aggre-

gate

(kg/m3)

Coarse Aggre-

gate

28-Days Com-

pressive

strength

10 mm

(kg/m3)

20 mm

(kg/m3) Cube

(MPa)

Cylin-

drical

(MPa)

0.47 (Mix-A) 362

(290+72+0) 170 0.40 650 777 518 45.72 36.57

0.36 (Mix-B) 417

(334+83+0) 150 0.35 726 730 487 68.57 57.14

0.27 (Mix-C) 525

(400+75+50) 140 1.00 692 754 406 88.60 76.37

0.20 (Mix-D) 750

(563+112+75) 150 1.16 536 640 427 103.55 90.83

cone test as per Indian Standard. The mix design de-

tails are given in Table 3. Adjustment was made in

mixing water as a correction for aggregate water ab-

sorption. For conducting studies, the concrete mixes

were prepared in pan type concrete mixer. Before use,

the moulds were properly painted with mineral oil,

casting was done in three different layers and each

layer was compacted on vibration table to minimize

air bubbles and voids. After 24 hours, the specimens

were demoulded from their respective moulds. The

laboratory conditions of temperature and relative hu-

midity were monitored during the different ages at

27±2oC and relative humidity 65% or more. The

specimens were taken out from the tank and allowed

for surface drying and then tested in saturated surface

dried condition.

3 Creep models

3.1 Creep as per B-3 model

This model (B3) was developed by Bazant and

Baweja [5] and described by ACI in 1997. The B3

Model has been found to be useful for both simple

and complex structures and it clearly separates basic

and drying creep. As per B3 model, for constant

stress applied at age at loading t’, Total strain at time

t,

ϵ(t) = 𝐽(𝑡, 𝑡′)σ + ϵ𝑠ℎ(𝑡) + α∆T(t) (1)

Where, J(t, t’) is the compliance function =

strain (creep plus elastic) at time t caused by a unit

uniaxial constant stress applied at age t’ in days’, σ

= uniaxial stress, ϵ = strain, ϵsh =shrinkage strain

(negative if volume decreases), ∆T(t) = temperature


27

change from reference temperature at time t, and α =

thermal expansion coefficient.

The compliance function may further be de-

composed as

𝐽(𝑡, 𝑡′) = 𝑞1 + C0(𝑡, 𝑡′) + 𝐶𝑑(𝑡, 𝑡′, 𝑡0) (2)

where, q1 = instantaneous strain due to unit stress,

q1 = 0.6 x 106 /E28 and E28 (MPa) = 4734(fc)0.5, C0(t,

t’) = compliance function for basic creep (creep at

constant moisture content and no moisture move-

ment through the material), and Cd(t, t’, t0) = addi-

tional compliance function due to simultaneous dry-

ing.

The creep coefficient, φ(t, t′ ) should be calcu-

lated from the compliance function,

φ(𝑡, 𝑡′) = E(𝑡′)𝐽(𝑡, 𝑡′) − 1 (3)

where, E(t′) = (static) modulus of elasticity at loading

age t ′

Calculations of Creep and Time Dependent Strain

Components

The total basic creep compliance is obtained by

equation as follows:

C0(𝑡, 𝑡

′) = 𝑞2 ∙ 𝑄(𝑡, 𝑡′) + 𝑞3 ∙ ln[1 +(𝑡 − 𝑡′)𝑛] + 𝑞4 ∙ ln(𝑡/𝑡′)

(4)

where, q2, q3 and q4 represent the aging viscoelas-

tic compliance, non-aging viscoelastic compliance,

and flow compliance respectively, as deduced from

the solidification theory, q2 = 185.4 c0.5 fc-0.9 , q3 =

0.29(w/c)4.q2, q4 = 20.3(a/c)-0.7.

The values of Q(t, t′) can be obtained from the

following approximate formula (derived by Bazant

and Prasannan, 1989 [17]) which has an error of less

than 1% for n = 0.1 and m = 0.5;

𝑄(𝑡, 𝑡′) = 𝑄𝑓(𝑡′) [1 + (

𝑄𝑓(𝑡′)

𝑍(𝑡, 𝑡′))

𝑟(𝑡′)

]

−1/𝑟(𝑡′)

(5)

where, r(t’) = 1.7(t’)0.12+8, Z(t, t’) = (t’)-m ln[1+(t-t’)n]

(m=0.5, n=0.1), Qf(t’) = [0.086(t’)2/9+1.21(t’)4/9]−1

Additional Creep Due to Drying (Drying Creep)

𝐶𝑑(𝑡, 𝑡′, 𝑡0) = 𝑞5 ∙ [𝑒−8𝐻(𝑡) − 𝑒−8𝐻(𝑡0′)]

0.5 (6)

If t ≥t/0, t/

0= max(t/ , t0). Otherwise, Cd(t, t/, t0) = 0,

t/0 is the time at which drying and loading first act

simultaneously, and

𝐻(𝑡) = 1 − (1 − ℎ)𝑆(𝑡) (7)

where, q5 = 7.57 × 105 .fC-1. |ɛsh∞|-0.60.

ϵ𝑠ℎ∞ = ϵ𝑠∞ (𝐸(607)

𝐸(𝑡0 + 𝜏𝑠ℎ)) (8)

where,

𝐸(𝑡) = 𝐸(28) (𝑡

4 + 0.85𝑡)0.5

(9)

ϵ𝑠∞ = −𝛼1𝛼2(1.9 × 10−2𝑤2.1𝑓𝑐

−0.28

+ 270) (𝑖𝑛 10−6) (10)

This means that ɛs∞= ɛsh∞ for t0 = 7 days and τsh

= 600 days.

Time dependence: S(t) = tanh((t-t0)/ τsh)0.5, size

dependence: τsh= kt(ks.D)2, effective cross-section

thickness (D = 2v/s) which coincides with the actual

thickness in the case of a slab, v/s = volume to sur-

face ratio of the concrete member. kt = 295740.59 ×

t0-0.08.fc

-0.25 days/cm2, ks is the cross-section shape

factor (Table 5).

High accuracy in this respect is not needed ks ≈

1 can be assumed for analysis.

Following parameters and coefficients were

considered while making calculations for experi-

mental mixes using creep and shrinkage prediction

model B3 by Zdenek P. Bazant and Sandeep Baweja,

Type I cement was used in this study. Hence,

α1 was taken as 1.

Since all the samples were sealed by wrap-

ping in Butyl Rubber Sheet up to 28 days, α2

was taken as 1.2

Age at which drying of specimen began was

taken as 28 days.

Relative humidity of environment during

curing and loading was maintained at 60%

and same was used for calculations.

Type of specimen was considered as infinite

cylinder. Hence, kS was taken as 1.15.

All the other factors were calculated using

above mentioned formulas by using differ-

ent values of fcm, t, t0 and other parameters

associated to individual mixes.

Table 4 – Coefficients based on cement type and

curing conditions

α1

1.0 for type I cement

0.85 for type II cement

1.1 for type III cement

α2

0.75 for steam-curing

1.2 for sealed or normal curing in air

with initial protection against drying

1.0 for curing in water or at 100% rela-

tive humidity.

Table 5 – Cross-section shape factor (ks)

ks

1 Infinite slab

1.15 Infinite cylinder

1.25 Infinite square prism

1.30 Sphere

1.55 Cube


28

3.2 Creep as per FIB model code 2010

The fib 2010 Model [3] was introduced by the

International Federation for Structural Concrete in

2013. As per FIB model code 2010, within the range

of service stresses |σc| ≤ 0 4.fcm (to), creep is assumed

to be linearly related to stress.

For a constant stress applied at time to this leads

to creep strain at age of concrete t,

ϵ𝑐𝑐(𝑡, 𝑡0) = (σ𝑐(𝑡0)/𝐸𝑐𝑖)φ(𝑡, 𝑡0) (11)

where, φ(t, to) is creep coefficient, Eci is the modulus

of elasticity in MPa at the age of 28 days.

The stress dependent strain ɛcσ(t, to),

ϵ𝑐𝑐(𝑡, 𝑡0) = σ𝑐(𝑡0) (1

𝐸𝑐𝑖(𝑡0)+

φ(𝑡, 𝑡0)

𝐸𝑐𝑖)

= σ𝑐(𝑡0)J(𝑡, 𝑡0)

(12)

where, J(t, to) is the creep function or creep compli-

ance, representing the total stress dependent strain

per unit stress and Eci(to) is the modulus of elasticity

at the time of loading to.

Creep coefficient

The creep coefficient may be calculated from

φ(𝑡, 𝑡0) = φ0𝛽𝑐(𝑡, 𝑡0) (13)

where, φo is the notional creep coefficient and βc(t, to)

is the coefficient to describe the development of

creep with time after loading, t is the age of concrete

in days at the moment considered and to is the age of

concrete at loading in days.

φ0 = φ𝑅𝐻𝛽(𝑓𝑐𝑚)𝛽(𝑡0) (14)

where, β(fcm) = 16.8/(fcm)0.5, β(to) = 1/ (0.1 + t00.2),

and

φ𝑅𝐻 = 𝛼2 [1 + 𝛼1 (1 −

𝑅𝐻100

0.1ℎ13

)] (15)

fcm is the mean compressive strength at the age

of 28 days in MPa, RH is the relative humidity of the

ambient environment in %. h = 2Ac/u = notional size

of member in [mm], where Ac is the cross-section in

mm² and u is the perimeter of the member in contact

with the atmosphere in mm. α1 = (35/fcm)0.7 and α2 =

(35/fcm)0.2.

The development of creep with time, βc(t, to), is

described by:

𝛽𝑐(𝑡, 𝑡0) = [𝑡 − 𝑡0

𝛽𝐻 + 𝑡 − 𝑡0]0.3

(16)

where

𝛽𝐻 = 1.5ℎ[1 + (1.2𝑅𝐻/100)18] + 250𝛼3

≤ 1500𝛼3 (17)

and α3 = (35/fcm)0.5



mental mixes using FIB model code 2010,



and same was used for calculations

All the samples were concrete cylinders hav-

ing diameter 150 mm and height 300 mm



ent values of fcm t, t0 and other parameters


3.3 Creep as per AASHTO 2014 model

The AASHTO Model [18] is described by

AASHTO LRFD Bridge Design Specifications 7th

Edition (Section 5.4.2.3) in 2014. The creep compli-

ance J(t, to) is given by,

J(𝑡, 𝑡0) =1

𝐸𝑐𝑚28+

φ(𝑡, 𝑡0)

𝐸𝑐𝑚28 (18)

where

𝐸𝑐𝑚28(MPa) = 0.043𝐾1𝛾1.5(𝑓𝑐𝑚28)

0.5 (19)

K1 = correction factor for source of aggregate to

be taken as 1.0 unless determined by physical test. γ

= concrete unit weight (kg/m3). Creep coefficient φ(t,

to) = 1.9.ks.khc.kf.ktd.to-0.118. Where, kf = factor for the

effect of concrete strength, kf = 35/(7+fcmto). ks = fac-

tor for the effect of volume-surface ratio of the com-

ponent, ks= 1.45-0.0051(V/S), khc= 1.56 – 0.008H,

where H is the relative humidity (%), ktd = [t / (61 –

0.58fcmto + t)].



mental mixes using AASHTO 2014 model

Unit weight of concrete was considered as

2400 kg/m3.

Since all the samples were cylindrical con-

crete specimen having diameter 150 mm

and height 300 mm, V/S was taken as 0.03.




K1 was taken as 1 for all the mixes.



ent values of fcm, t, t0, and other parameters


3.4 Creep as per ACI 209R-92 model

The American Concrete Institute recommends

the ACI 209 Model [19] as the current standard code

model. The creep compliance function J(t, to) that

represents the total stress-dependent strain by unit

stress is given by


29

J(𝑡, 𝑡0) =1 + φ(𝑡, 𝑡0)

𝐸𝑐𝑚(𝑡0) (20)

where, φ(t, to) is creep coefficient.

Ecm(to) = modulus of elasticity at the age of

loading (MPa) is given by Ecm(to) in MPa

=(0.043)γ3/2.(fcm(to))0.5

γ is concrete unit weight in kg/m3 and fcm(to)

mean concrete compressive strength at age of load-

ing. fcm(to) = fcm28.[to/(a+b.to)], where fcm28 is the av-

erage 28-day concrete compressive strength (MPa) a

and b are constants according to table 6 below.

Table 6 – a and b based on curing conditions

Type of

cement

Moist cured

concrete

Steam cured

concrete

I a = 4.0, b = 0.85 a = 1.0, b = 0.95

III a = 2.30, b = 0.92 a = 0.70, b = 0.98

φ(𝑡, 𝑡0) = φ𝑢 [(𝑡 − 𝑡0)

0.6

10 + (𝑡 − 𝑡0)0.6

] (21)

where, φu = 2.35 γH. γto. γs. γvs. γα. γψ, to = age of con-

crete at loading (days), t = age of concrete (days), H=

relative humidity (%), φu = ultimate creep coefficient.

Relative humidity correction factor, γH

𝛾𝐻 = 1.27 − 0.0067𝐻 (22)

Age of loading correction factor, γto

𝛾𝑡𝑜 = 1.25𝑡0−0.118 (23)

for moist curing, and

𝛾𝑡𝑜 = 1.13𝑡0−0.094 (24)

for steam curing

Slump correction factor, γs

𝛾𝑠 = 0.82 + 0.00264𝑠 (25)

where s is the slump of fresh concrete (mm). Vol-

ume-surface ratio correction factor, γvs

𝛾𝑣𝑠 =2

3(1 + 1.13𝑒

−0.0213(𝑉𝑆)) (26)

where, V/S is the volume-surface ratio (mm)

Air content correction factor, γα

𝛾𝑎 = 0.46 + 0.09α ≥ 1 (27)

where, α is the air content (%).

Fine aggregate correction factor, γψ

𝛾ψ = 0.88 + 0.0024ψ (28)

where, ψ is the fine aggregate to total aggregate by

weight (%).



mental mixes using ACI 209R-92 model

Type of curing was considered as moist

curing.

Unit weight of concrete was considered as

2400 kg/m3.

All the samples were concrete cylinders

having diameter 150 mm and height 300

mm.






ent values of fcm, t, t0, slump, ratio of fine

aggregate to total aggregate, air content and

other parameters associated to individual

mixes.

3.5 Creep as per GL2000 model

This original GL 2000 Model [20] was devel-

oped by Gardner and Lockman in 2001. The creep

compliance, J(t, to) contains two parts: elastic and

creep strain.

J(𝑡, 𝑡0) =1

𝐸𝑐𝑚𝑡𝑜+

φ(𝑡, 𝑡0)

𝐸𝑐𝑚28 (29)

𝐸𝑐𝑚𝑡(MPa) = 3500 + 4300𝑓𝑐𝑚𝑡0.5 (30)

𝑓cmt = 𝑓cm28𝛽𝑒2 (31)

𝛽e = 𝑒

(𝑠2)(1−(

28𝑡

))

0.5

(32)

where s is CEB style strength development parame-

ter related to cement type.

The correction term for effect of drying before

loading φ(tc), could be determined as:

if to = tc, φ(tc) = 1, if to > tc, φ(tc) = [1-((to-tc)/(to-

tc + 0.12(V/S)2))0.5]0.5

φ(𝑡, 𝑡0) = φ(𝑡𝑐)

[

2(𝑡 − 𝑡0)0.3

(𝑡 − 𝑡0)0.3 + 14

+ (7

𝑡0)

0.5

(𝑡 − 𝑡0

𝑡 − 𝑡0 + 7)

0.5

+ 2.5(1 − 1.086ℎ2)(𝑡 − 𝑡0

𝑡 − 𝑡0 + 0.12 (𝑉𝑆)

2

)

0.5

]

(33)


30

Table 7 – Strength development factor (s) based on

type of cement

Cement type s

I 0.335

II 0.4

III 0.13



mental mixes using GL 2000 model

Strength development parameter (s) related

to cement type was taken as 0.13.

Since all the samples were cylindrical con-

crete specimen having diameter 150 mm

and height 300 mm, V/S was taken as 0.03






ent values of fcm, t, t0 and other parameters


3.6 Comparison of creep models

There are several differences in the influence

factors, formula forms, applicable scope and predic-

tion accuracy of these models due to limitation of

specific experimental condition and the emphasis of

different researchers. Few common parameters are

used by all the five models (B3, FIB model code

2010, AASHTO 2014, ACI 209R-92 and GL 2000

model) discussed in the paper. However, B3 Model

consider additional parameters than FIB model code

2010 and same have been listed in Table 8 below.

The magnitude and the rate of development of

creep depends upon many factors such as composi-

tion of concrete mix, environmental conditions and

load level. In terms of applicability, the use of B3

and AASHTO 2014 model is restricted to concrete

having 28-day standard cylinder compression

strength of 15 to 70 MPa. Similarly, the use of GL

2000 model is restricted to concrete having compres-

sive strength in the range of 16 MPa to 82 MPa.

However, FIB model code 2010 is applicable to both

normal and high strength concrete up to 130 MPa.

Restrictions based on grade of concrete have not

been suggested for application of ACI 209R-92

model. Therefore, creep related calculations for high

strength concrete using B3, AASHTO 2014 and GL

2000 models may show deviations from the corre-

sponding experimental creep values. Factors and pa-

rameters associated with the use of mineral and

chemical admixtures in the concrete are not taken

into account by any of the above mentioned five

models. FIB model adopted new functions and cor-

rection factors which modifies long term behaviour

of concrete for prediction and for wider applicability.

Table 8 – Parameters required by analytical models for prediction of creep

Parameter

Creep models

B3 FIB

2010

AASHTO

2014

ACI

209R-92 GL 2000

Concrete Unit Weight √ √

Effective Thickness √

Volume-Surface Ratio √ √ √

Cross Section Shape of Member √ √ √

Cement Content √

Water Content √

Water-Cement Ratio √

Aggregate-Cement Ratio √

Fine Aggregate Percentage √

Cement Type √ √ √ √

Curing Method √ √

Slump √

Air Content √

Relative Humidity √ √ √ √ √

Age of Concrete at loading √ √ √ √ √

Age of Concrete at drying (end of curing) √ √

Compressive Strength at loading

Compressive Strength at 28 days √ √ √ √ √

Temperature of curing & environment

Factors associated with chemical admixture

Factors associated with mineral admixture

Aggregate dependent parameter scaling factor


31

4 Creep coefficient as per creep models

Creep coefficient of four different mixes (A, B,

C & D) as shown in Table- has been worked out us-

ing Bazant’s B-3, ACI 209-R 92, AASHTO 2014,

GL-2000 and FIB model code 2010. The creep coef-

ficients are determined for three different ages at

loading (7, 28 and 365 days) for design life of 100

years and relative humidity of 60% (Figure 1 to Fig-

ure 3). The comparison of creep coefficients as per

different models indicates that there is sharp increase

in creep coefficient for each model upto around 365

days age. The rate of increase of each model drasti-

cally slows down after 365 days irrespective of the

grade of concrete. Both B3 model and GL 2000

shows higher creep coefficients at early age except

in case of mix A having water to cementitious ratio

of 0.47 and age at loading of 7 days. The AASHTO

2014 Model in general gave the lowest values of

creep coefficient except in case of mix A having wa-

ter to cementitious ratio of 0.47 and age at loading of

7 days. The rate of increase in creep coefficient after

365 days age in case of B3 Model is relatively higher

than other models. Both ACI and FIB model code

2010 gave creep coefficients in between the B3 and

AASHTO models except in case of mix A having

water to cementitious ratio of 0.47 and age at loading

of 7 days and similar trend is observed in higher

grades of concrete. The magnitude of creep coeffi-

cient depends on a wide range of factors including

the stress range, element size, concrete mix, coarse

gravel content, cement content, type of cement, wa-

ter/cement ratio, relative humidity, temperature, time

of loading, type and duration of curing and maturity.

Including most of these factors in creep coefficient

calculations is tedious. B3 Model and ACI 209R-92

requires most numbers of parameters for creep pre-

diction. FIB Model code 2010, GL 2000 Model and

AASHTO 2014 Model require less number of pa-

rameters to predict the creep coefficient.

In order to check the performance of these mod-

els for high strength concrete; an experimental study

has been conducted with two mixes EM-1 and EM-

2 with water to cementitious ratio of 0.47 and 0.20

respectively and results are discussed in paragraph 5.

(a) Mix-A, w/c =0.47, age at loading = 7 days (b) Mix-B, w/c =0.36, age at loading = 7 days

(c) Mix-C, w/c =0.27, age at loading = 7 days (d) Mix-D, w/c =0.20, age at loading = 7 days

Fig. 1 – Comparison of creep coefficient of concrete mix (a) Mix-A, (b) Mix-B, (c) Mix-C, (d) Mix-D

with different creep models (age at loading of 7 days)


32










20

5 Experimental creep study in compression

Creep is the continuous increase of the strain in

concrete without any change in the applied stress.

Creep depends on several factors, including mixture

proportioning, environmental conditions, curing

conditions, geometry of concrete member, loading

history and stress conditions. Creep of concrete de-

pends on the stress in the concrete, age at loading and

the duration of loading. As long as the stress in con-

crete does not exceed about 40 percent of character-

istic compressive strength, creep may be assumed to

be proportional to the stress. The creep co-efficient

ɸ(t, to) is given by the equation:

ɸ(𝑡, 𝑡0) =𝜀𝑐𝑐(𝑡)

𝜀𝑐𝑖(𝑡0) (34)

where, εcc(t) = creep strain at time t > t0, (This does

not include the instantaneous strain in concrete at the

time of loading), εci(t0) = initial strain at loading, and

t0 = age of concrete at the time of loading

The creep test was carried out on a cylindrical

specimen of size 150 mm diameter and 300 mm

height as per ASTM C-512 for concrete with water

to cementitious ratio of 0.47 (EM-1) and 0.20 (EM-

2) with same mix proportions as shown in table 3 for

mix A and mix D respectively. The compressive

strength of each mix was used for calculation of the

load to be applied to the specimens, which was taken

as 40% of the average compressive strength. The cyl-

inders were sulphur capped before being stacked up

on top of one another in the creep rig. The vibrating

wire strain gauges were inserted in cylindrical spec-

imens at the time of casting. The specimens were

cured by wrapping in Butyl Rubber Sheet up to 28

days. Relative Humidity was maintained at 60% and

temperature was maintained at 270C. The tempera-

ture and relative humidity were maintained at same

level after 28 days as well. The creep as per ASTM

C-512 is being measured using manual data readout

units. In creep test, samples are kept in controlled

and loaded condition for the time period of 180 days

(Figure 4). Each strength and control specimen was

kept under the same curing and storage treatment as

the loaded specimen.

The steps for calculating creep strain at a given age

are as follows:

EM-1: Water Cementitious Ratio: 0.47and Aver-

age: fcy: 45.66 MPa

Stress applied: 18.26 MPa (40% of fcy)

Total load applied: 323 kN

Age at the time of loading: 28 days

Average strain immediately after loading at time t0 =

484.31 (µ-strain)

Average strain of unloaded specimens at the time of

loading at time t0 = 19.03 (µ-strain)

Load induced strain per unit stress immediately after

loading = (484.31-19.03)/18.26 = 25.48 (µ-strain/

(MPa))

Average strain of loaded specimens at 180 days of

loading = 1321.08 (µ-strain)

Average strain of unloaded specimens at 180 days of


Load induced strain per unit stress at 180 days of

loading = (1321.08-258.57)/18.26 = 58.19 (µ-strain/

(MPa)

Therefore, the Creep strain per unit stress = (58.19-

25.48) = 32.71 µ-strain/ (MPa)

EM-2: Water Cementitious Ratio: 0.20 and Aver-

age fcy: 100.21 MPa

Stress applied: 40.08 MPa (40% of fcy)

Total load applied: 708 kN

Age at the time of loading: 28 Days

Average strain immediately after loading at time t0:

1006.80 (µ-strain)

Average strain of unloaded specimens immediately

after loading: 0.00 (µ-strain)

Load induced strain per unit stress immediately after

loading = (1006.80 – 0)/40.08 = 25.11(µ-strain)

Average strain of loaded specimens at 180 days of


Average strain of unloaded specimens at 180 days of


Load induced strain per unit stress at 180 days of

loading = (1784.08-131.77)/40.08 = 41.22 (µ-strain/

(MPa)

Therefore, the Creep strain per unit stress = (41.22-

25.11) = 16.11 µ-strain/ (MPa)

Test results of creep up to 180 days are given in

Table 9.

Fig. 4 – Creep testing arrangement


34

(a) EM-1, w/c =0.47, age at loading = 28 days (b) EM-2, w/c =0.20, age at loading = 28 days

Fig. 5 – Comparison of creep coefficient of concrete mix EM-1 having (a) w/c = 0.47 (b) w/c = 0.20 and

age at loading of 28 days with different creep models

Table 9 – Test results of creep up to 180 days with water to cementitious ratio 0.20 and 0.47

Age of

con-

crete

(Days)

Dura-

tion of

loading

(Days)

Avg. To-

tal

Strain

Loaded

Samples

A

(µ-

strain)

Avg.

Strain

Un-

loaded

Samples

B

(µ-

strain)

Total

load in-

duced

strain

C =A-B

(µ-

strain)

Total Load

Induced

Strain per

unit stress

(µ-Strain/

MPa)

D

Load induced

Strain per unit

stress Immedi-

ately After Load-

ing

(µ-Strain/ MPa)

E

Creep

Strain

per unit

stress

(µ-

Strain/

MPa)

D-E

Water to Cementitious Ratio 0.20

28 0 1006.80 0 1006.80 25.11 25.11 0

56 28 1528.45 41.29 1487.16 37.10 25.11 12.49

88 60 1601.53 73.54 1527.99 38.12 25.11 13.01

118 90 1648.27 104.16 1544.11 38.52 25.11 13.42

148 120 1694.81 114.50 1580.31 39.42 25.11 14.31

178 150 1738.40 125.82 1612.58 40.21 25.11 15.10

208 180 1784.05 131.77 1652.28 41.22 25.11 16.11

Water to Cementitious Ratio 0.47

28 0 484.31 19.03 465.28 25.48 25.48 0

56 28 958.61 135.75 822.86 45.06 25.48 19.58

88 60 1143.93 189.75 954.18 52.25 25.48 26.77

118 90 1200.90 199.82 1001.08 54.82 25.48 29.34

148 120 1230.97 210.38 1020.59 55.89 25.48 30.41

178 150 1273.54 229.91 1043.63 57.15 25.48 31.67

208 180 1321.08 258.57 1062.51 58.19 25.48 32.71

6 Comparison of experimental strains with

models

The creep coefficients are determined experi-

mentally for EM-1 and EM-2 for age at loading of 28

days and upto 180 days loading duration and relative

humidity of 60% (Figure 5 and Figure 6). The test

results of the experimentally obtained creep coeffi-

cient values for experimental mixes EM-1 and EM-

2 has been compared with Bazant’s B3 model ACI

209-R 92, AASHTO 2014, GL-2000 and FIB model

code 2010. The results indicate that experimentally

obtained creep coefficients for water cementitious

ratio of 0.47 (normal strength concrete) are closer to

corresponding creep coefficients predicted using all

the models except GL2000. However, in case of high

strength concrete, B3 model, GL-2000 and ACI 209-

R 92 predicts higher values of creep coefficient when

compared with experimentally obtained creep coef-

ficients for water cementitious ratio of 0.20. The re-

sults indicate that experimentally obtained creep co-

efficients for high strength concrete are closer to cor-

responding creep coefficients obtained using FIB

model code 2010 and AASHTO 2014 model.


35

7 Conclusions

Based on the comparison of creep coefficients

using Bazant’s B-3, ACI 209-R 92, AASHTO 2014,

GL-2000 and FIB model code 2010 and the experi-

mentally obtained creep coefficients; following con-

clusions can be drawn:

(1) The comparison of creep coefficients as per dif-

ferent models indicates that there is sharp in-

crease in creep coefficient for each model upto

around 365 days age. The rate of increase of

each model drastically slows down after 365

days irrespective of the grade of concrete.

(2) Both B3 model and GL 2000 shows higher

creep coefficients at early age except in case of

mix A having water to cementitious ratio of

0.47 and age at loading of 7 days. The

AASHTO 2014 Model in general gave the low-

est values of creep coefficient except in case of

mix A having water to cementitious ratio of

0.47 and age at loading of 7 days. The rate of

increase in creep coefficient after 365 days age

in case of B3 Model is relatively higher than

other models. Both ACI and FIB model code

2010 gave creep coefficients in between the B3

and AASHTO models except in case of mix A

having water to cementitious ratio of 0.47 and

age at loading of 7 days and similar trend is ob-

served in higher grades of concrete.

(3) The results indicate that experimentally ob-

tained creep coefficients for water cementitious

ratio of 0.47 (normal strength concrete) are

closer to corresponding creep coefficients pre-

dicted using all the models except GL2000.

However, in case of high strength concrete, B3

model, GL-2000 and ACI 209-R 92 predicts

higher values of creep coefficient when com-

pared with experimentally obtained creep coef-

ficients for water cementitious ratio of 0.20. Use

of B3, GL 2000 and AASHTO 2014 models are

recommended for concrete mixes having com-

pressive strength up to 80 MPa. Therefore,

creep related calculations for high strength con-

crete using B3 and GL 2000 models showed de-

viations from the corresponding experimental

creep values. However, AASHTO 2014 model

remain exception in this regard and holds good

even in the case of high strength concrete. The

results indicate that experimentally obtained

creep coefficients for high strength concrete are

closer to corresponding creep coefficients ob-

tained using FIB model code 2010 and

AASHTO 2014 model.

(4) The comparison of experimental data of creep

coefficient with all the five models shows that

Bazant’s B3 model, GL-2000 and ACI 209-R

92 will not hold good for high strength concrete.

FIB model code 2010 and AASHTO 2014

model enables a more accurate analysis for both

high and normal strength concrete and better as-

sessment of the creep coefficient of concrete

structures at the design stage. In FIB model

code 2010 and AASHTO 2014, complexity is

significantly reduced and a range of influencing

parameters are excluded from the model for

simplicity and easy adaptation at the design

stage.

References

1. Guo, Z.; and Shi, X. (2011) “Reinforced Con-

crete Theory and Analyse,” Tsinghua University

Press, Beijing, pp. 58-64.

2. Huang, G.; Hui, R.; and Wang, X. (2012) “Creep

and Shrinkage of Concrete” China Electric

Power Press, Beijing, pp. 95-104.

3. FIB model code 2010, International Federation

for Structural Concrete (fib).

4. Bazant, Z. P. (2001) “Prediction of concrete

creep and shrinkage: past, present and future,”

Nuclear Engineering and Design, vol. 203, pp.

27-38.

5. Bazant, Z. P.; and Baweja S. (1995) “Creep and

shrinkage Prediction Model for Analysis and

Design of Concrete Structures-Model B3,” Ma-

terials and Structures, vol. 28, pp. 357-365.

6. Gong, L.; Hui, M.; and Yang, B. (1988). “Prac-

tical mathematical expressions for creep and

shrinkage of concrete,” Journal of Building


7. Hanson J. A. (1953) “A Ten-year Study of Creep

Properties of Concrete. Concrete Laboratory,”

US Department of the Interior, Bureau of Recla-

mation, Denver, Report no. SP-38.

8. Neville A.M. (1970) “Creep of Concrete: Plain,

Reinforced and Prestressed,” Amsterdam: North

Holland Publishing Co.

9. Neville, A. M.; Dilger W. H.; and Brooks J. J.

(1983) “Creep of Plain and Structural Concrete,”

London and New York: Construction Press.

10. Vandamme, M.; and Ulm, F.-J. (2009)

“Nanogranular origin of concrete creep,” Pro-

ceedings of the National Academy of Sciences,

vol. 106, no. 26, pp. 10552–10557.

11. Westman, G. (1999) “Concrete creep and ther-

mal stresses,” Doctoral thesis. Lulea, Sweden,

pp. 7.

12. Gilbert, R.I.; and Ranzi G. (2011) “Time-De-

pendent Behaviour of Concrete Structures,”

London and New York: Span Press, pp. 25-30.


36

13. Gilbert, R.I. (1988) “Time Effects in Concrete

Structures,” Amsterdam: Elsevier Science Pub-

lishers.

14. Janusz, H. (2015) “Conventional Models for

Creep in Normal) and High-Strength Concrete,”

Architecture Civil Engineering Environment

Journal, vol. 8, no. 4, pp.31-38.

15. ACI Committee 211.1 (1991) “Standard Prac-

tice for Selecting Proportions for Normal, Heav-

yweight and Mass Concrete”, American Con-

crete Institute.

16. IS 456:2000, “Plain and reinforced concrete”-

code of practice, BIS New Delhi.

17. Bazant, Z. P.; and Prasannan S. (1989) “Solidi-

fication theory for concrete creep: I. Formula-

tion, and II. Verification and application,”

ASCE Journal of Engineering Mechanics, vol.

115, no. 8, pp. 1691–1725.

18. American Association of State Highway and

Transportation Officials (AASHTO). (2014).

“AASHTO LRFD bridge design specifications,”

American Association of State Highway.

19. ACI Committee 209. (1992). “Prediction of

Creep, Shrinkage, and Temperature Effects in

Concrete Structures,” American Concrete Insti-

tute.

20. Gardner, N. J.; and Zhao, J. W. (1993) “Creep

and shrinkage revisited,” ACI Materials Journal,

vol. 90, no. 3, pp. 236-246.



ISSN 2465-7964 / eISSN 2465-7972

https://doi.org/10.18702/acf.2020.12.6.2.37

37

Technical Paper

Appraisal of geopolymer lightweight aggregates sintered by

microwave radiations

Nimra Saleem, Khuram Rashid*, Noor Fatima, Sadia Hanif, Ghinwa Naeem, Aamna Aslam,

Miral Fatima, Kiran Aslam

(Received May 15, 2020; Revised November 6, 2020; Accepted December 26, 2020; Published December 31, 2020)

Abstract: This work was designed for the production of geopolymer based lightweight aggregate (LWA)

using industrial by-products. Combination of fly ash (FA) and silica fume (SF) were used as precursors,

whereas, combination of sodium hydroxide and sodium silicate were used as activator. Small amount of

sodium bicarbonate was also used for surface hardening and early strength development. Pellets of dif-

ferent sizes were crafted manually and cured by microwave radiations just for 5 minutes. The physico-

mechanical properties of produced pellets (LWA) were discussed in light of: morphology, density, water

absorption, specific gravity, porosity, aggregate impact value, and particle crushing strength. The prop-

erties of LWA were also compared with literature reported synthetic LWAs cured with different tech-

niques. The water absorption and specific gravity of LWAs were within the specified range provided by

ACI standard. Mechanical strength properties briefed that the produced LWAs were strong enough to

resist compressive load comparable to natural LWAs and many other synthetic LWAs. Thus, proposed

curing method, microwave irradiation, has been found to be a sustainable and fast curing technique than

conventional energy-intensive curing regimes. The results also confirmed that produced LWAs have po-

tential to replace natural LWAs both in cast-in-place and precast concrete elements with possible eco-

nomic, environmental, and technical benefits.

Keywords: Geopolymer lightweight aggregates; geopolymerization; pellets; microwave radiations;

physical and mechanical properties.

1 Introduction

The construction industry is considered to be

one of the most important indicators of economic

state of a country and concrete is the major and most

widely used construction material in civil engineer-

ing field. Bulk concrete production, however, leads

to both environmental pollution and excessive re-

sources consumption [1]. Growing industrial wastes

such as fly ash (FA), ground granulated blast furnace

slag (GBFS), and silica fume (SF) can be utilized as

construction materials, which is considered a healthy

and sustainable practice to dispose the waste off and

conserve the available resources for future genera-

tions [2]. The incorporation of these industrial by-

products as a partial replacement of cement is done

in order to reduce huge CO2 emissions from cement

production [3]. On the other hand, it is well known

that self-weight of concrete structures considerably

influences the design load and economy of structures.

Since, aggregate phase occupies 60-80% of total vol-

ume of concrete [4]. Therefore, the utilization of ar-

tificial LWAs, manufactured from waste and by-

products, in concrete production has attracted signif-

icant research interest. The use of artificial LWAs as

an alternative to natural aggregates not only reduces

the dead load of structures but can also lead to many

positive environmental consequences including (1)

the preservation of natural resources; (2) conserva-

tion of energy required for quarrying processes [5];

Corresponding author Khuram Rashid is an Associate Profes-

sor in the Dept. of Architectural Engineering and Design, Uni-

versity of Engineering and Technology, Lahore, Pakistan.

Nimra Saleem is a student in the Dept. of Architectural Engi-

neering and Design, UET, Lahore, Pakistan.

Noor Fatima is a student in the Dept. of Architectural Engi-


Sadia Hanif is a student in the Dept. of Architectural Engineer-

ing and Design, UET, Lahore, Pakistan.

Ghinwa Naeem is a student in the Dept. of Architectural Engi-


Aamna Aslam is a student in the Dept. of Architectural Engi-


Miral Fatima is a student in the Dept. of Architectural Engi-


Kiran Aslam is a student in the Dept. of Architectural Engi-



38

(3) conversion of waste into value-added products

[6].

Generally, LWAs can be classified into two ma-

jor categories: natural LWAs (pumice, scoria, diato-

mite, volcanic cinders, sawdust and rice husk) and

artificial LWAs (foamed slag, bloated clay, ex-

panded shales and slate, sintered fly ash and ex-

panded perlite) [7]. LWAs have been produced arti-

ficially due to its significant demand and to avoid the

depletion of natural LWA resources [8-10]. Gener-

ally, artificial LWAs are produced through agglom-

eration which involves granulation or compaction of

powdered waste materials into fresh pellets with the

desired shape and size. The raw materials are

pelletized by mixing with liquid as a binding agent

to get the desired size and consistency. In agglomer-

ation process by granulation, different types of

pelletizer machines can be used such as disc or pan

type, drum type, cone type or mixer type [11]. The

fresh pellets are then cured, either by autoclaving,

sintering, cold bonding processes or by microwave

radiations [6]. Sintering process is based on the cre-

ation of a ceramic matrix. The matrix consists of alu-

mina silicates and the sintering temperature for alu-

mina silicate fly ashes is typically in the range of

1100-1200 °C [8]. Many researchers have developed

LWAs by sintering, as high engineering properties

can be obtained depending on agglomerated mate-

rial’s properties and process efficiency. Requirement

of high temperature in sintering process leads to CO2

emissions. Thus, it comes up with the drawback of

environmental harms along with high production

cost [12]. Autoclaving process involves the mixing

of chemical such as cement, lime or gypsum with

source material at agglomeration stage. After that,

the specimen is exposed to autoclaving or cured in

pressurized saturated steam at a temperature of

140oC for several hours [13]. Cold bonding method

is normal water curing at ambient temperature to

bind the mixing materials. In this method, the mate-

rials are stabilized at granulation stage using any

binder such as cement, lime or alkali activation

mechanisms like geo-polymerization at ambient

temperature evading high temperature requirements,

which is a significant advantage of cold bonding

over sintering process [14]. Autoclave or steam cur-

ing process is less efficient to enhance LWA proper-

ties in comparison to water curing. This process does

not show significant difference in strength and dura-

bility properties like ordinary water curing [15].

Since curing method plays a significant role in deter-

mining the LWA properties, economy, and sustaina-

bility, it must be selected wisely. A prospective and

competitive solution to conventional methods of cur-

ing is the usage of microwave radiations.

Microwave radiations are the electromagnetic

radiations covering both electric and magnetic fields

oscillating in the direction of propagation at right an-

gles [16]. A significant difference between micro-

wave cured and conventionally cured material is in-

ternal microstructure. Materials cured with micro-

wave possess more consistent external and internal

structure and present better strength than conven-

tionally-cured materials. The properties of fly ash

based LWAs synthesized using microwave radia-

tions have been studied experimentally. Compared

with sintering and autoclaving, microwave heating

does not introduce thermal cracking, thermal stresses,

and provides durable aggregates [8]. The microwave

irradiation has potential to reduce both time and en-

ergy required for processing materials due to brisk,

efficient, and quick energy transfer mechanism [16].

Therefore, microwave radiation can be used as a

cost-effective and fast curing method for LWAs.

With this background, the aim of this work is to

manufacture geopolymer based LWAs using micro-

wave radiations. FA and SF are used as precursors.

Physical properties (density, void’s ratio, specific

gravity, water absorption, percentage expansion, po-

rosity, and morphology) and mechanical properties

(particle crushing strength and aggregate impact

value) have been examined for produced LWAs and

are compared with natural LWAs and previously for-

mulated LWAs.

2 Experimental methodology

The methodology adopted to achieve the target

was divided into two sub-goals. The first section

covered the specifications of materials used in this

work and production of geopolymer LWAs (AGP). In

the second section, testing was done to examine the

properties of AGP like morphology, density, porosity,

water absorption, particle crushing strength, and ag-

gregate impact value. The summary of research

methodology adopted for production and experimen-

tation of AGP is shown in Figure 1.

2.1 Materials and specimen preparation

The materials used in this study for production

of AGP were coal fly ash (FA), silica fume (SF), alka-

line activators (NaOH and Na2SiO3), and sodium bi-

carbonate (NaHCO3) as shown in Figure 2. Precur-

sors used for the manufacture of geopolymer based

LWAs were FA and SF. FA was obtained from DG

Cement Pakistan and its chemical composition re-

sembled with Class-F FA according to ASTM C618

[17]. The amount of oxides and other chemical con-

stituents of FA and SF are presented in Table 1. Al-

kaline activators used were the solutions of NaOH

and Na2SiO3. White crystalline flakes of NaOH and

alkaline solution of Na2SiO3 were purchased from

Akbari Mandi (Lahore, Pakistan).


39

Fig. 1 – Curing condition II and III

Fig. 2 – Materials used in research work: (a) FA, (b) SF, (c) NaHCO3, (d) NaOH, (e) Na2SiO3

400g of NaOH pellets were used to prepare 1

liter of 10M NaOH solution. NaHCO3 (baking soda)

having molecular mass of 84g/mol and density of 2.1

g/cm3 was used for surface hardening of aggregates,

which acts as an accelerator to decrease the setting

time of geopolymer paste [18]. The selection of mix

proportion for the production of AGP was based on hit

and trial method and on previous research

knowledge. The proportion of the AGP for three types

of pellets have been mentioned in Table 2. FA and

SF were used in amounts of 90% and 10% of total

weight of solid materials, respectively. 1% NaHCO3

of solid materials was used as an accelerator. One

mixture was selected (FA20-80SF) and their corre-

sponding properties were investigated in detail. Mix-

ture of two alkaline solutions, 10M NaOH and

Na2SiO3, was used such that the ratio of two solu-

tions (NaOH/Na2SiO3) was kept 0.25. The alkaline

activator to solid ratio selected was equal to 0.53.

Calculated amounts of materials were dry

mixed first for about 1-2 minutes. Further mixing

was carried out for 1-2 minutes on adding 50% of the

alkaline solution. After that, the remaining half of the

alkaline solution was added, and the mixing was con-

tinued for the same duration in order to ensure fine

blending.

Table 1 – Composition of oxides present in FA and SF

Material Oxides (%) Cl LOI Moiture

Content

CaO SiO2 Al2O3 MgO Fe2O3 K2O Na2O (%) (%) (%)

FA 9.02 56.34 23.08 1.70 6.43 0.56 0.28 0.025 < 3 < 1

SF 0.27 93.65 0.28 0.25 0.58 0.49 0.02 3.62 <5 -

Manual Mixing/Homogenization

Hand Crafting (Pellet Shape)

Drying at Room Temperature

Microwave Curing (5 min)

Materials: FA, SF, NaHCO3, Alkaline Activators (NaOH and Na2SiO3)

Composition: NaOH/ Na2SiO3 = 0.25Activator/Solid = 0.53

LWA Sample

2. Testing

Mechanical Properties:• Particle Crushing Strength• Aggregate Impact Value

Physical Properties:• Water Absorption• Specific Gravity• Percentage Expansion

• Morphology • Bulk Density• Porosity

1. LWA Production

(a) (b) (c) (d) (e)


40

Table 2 – Mixing proportion for production of AGP.

Aggregate

Type

Pellets

Designation

Binding Material

(% by total solid)

Alkaline Activator

(% by total liquid) Liquid

/Solid FA SF NaHCO3 Na2SiO3 NaOH

AGP FA20-80SF 89 10 1 80 20 0.53

FA30-70SF 89 10 1 70 30 0.53

FA40-60SF 89 10 1 60 40 0.53

Fig. 3 – AGP pellet formation process

After the formation of homogenous mixture, ag-

gregates were shaped by hands in laboratory having

size range about 11-17mm in diameter. Then, pellets

were cured under single curing regime that was mi-

crowave radiation curing. Aggregates were placed in

mud pot and cured for about 4-5 minutes in micro-

wave. The process for the manufacturing of aggre-

gates is illustrated in Figure 3. After the process of

curing, the aggregates were wrapped in plastic bags

to avoid the penetration of moisture so that it may

not alter the test results. The pellets designation such

as FA20-80SF is explained as: first two alphabets tell

the primary precursor (Fly ash as FA), after alpha-

bets, first two numerals tell the percentage of NaOH;

next two numerals tell the percentage of Na2SiO3,

and last two alphabets are for secondary precursor

(Silica fume abbreviated as SF).

2.2 Testing

LWAs were tested for physical and mechanical

properties in accordance with the respective stand-

ards. Figure 4 shows the procedure of different tests

performed on AGP in this work.

2.2.1 Physical properties

To examine the physical properties of LWAs,

bulk density, water absorption, specific gravity, po-

rosity, and expansion tests were performed on devel-

oped pellets. The morphological features of LWAs

were also examined. The particle shape and color

were observed from naked eye. Surface texture was

examined by touching the surface of LWAs and size

of AGP was computed by passing aggregates through

sieves as well as Vernier calipers. The bulk density

and percentage void test were carried out in accord-

ance with ASTM C29 [19]. The loose and compacted

bulk densities were determined by weighing LWAs

in a cylinder of known volume and were obtained

from Eq. (1) and Eq. (2). From loose and compacted

bulk densities, percentage of voids (spaces between

LWAs) was determined by using Eq. (3).

LBD = 𝑤𝐿𝐴+𝐶 − 𝑤𝐶

𝑉𝐶 (1)

CBD = 𝑤𝐶𝐴+𝐶 − 𝑤𝐶

𝑉𝐶 (2)

Percentage Voids = CBD − LBD

CBD × 100 (3)

Relative density (specific gravity) is used in the

computation of voids in aggregate. Saturated surface

dry (SSD) specific gravity is used if the aggregate is

wet, that is, if its absorption has been satisfied. Con-

versely, the oven dried (OD) specific gravity is used

for computations when the aggregate is dry or as-

sumed to be dry. Apparent relative density pertains

to the solid material making up the constituent parti-

cles, not including the pore space within the particles


41

Fig. 4 – Pictorial views of different tests performed on produced LWAs (AGP)

which is accessible to water. Absorption values are

used to calculate the change in the mass of an aggre-

gate due to water absorbed in the pore spaces within

the constituent particles, compared to the dry condi-

tion. For water absorption measurement, aggregates

retained on 4.75mm opening (Sieve # 4) were im-

mersed in water for 24 hours according to ASTM

C128-15 [20]. The relative density (specific gravity)

and water absorption values of LWAs were deter-

mined by means of equations (Eq. (4); Eq. (5); Eq.

(6) and Eq. (7)) given as follows:

Relative Density (OD) = 𝑚𝑂𝐷

𝑚𝑆𝑆𝐷 − 𝑚𝐴𝑃 (4)

Relative Density (SSD) = 𝑚𝑆𝑆𝐷

𝑚𝑆𝑆𝐷 − 𝑚𝐴𝑃 (5)

Apparent Specific Gravity = 𝑚𝑂𝐷

𝑚𝑂𝐷 − 𝑚𝐴𝑃 (6)

Absorption = 𝑚𝑆𝑆𝐷

𝑚𝑆𝑆𝐷 − 𝑚𝑂𝐷 × 100 (7)

LWAs contain pores which contribute to vol-

ume of aggregates. Therefore, it is necessary to de-

termine the true porosity of LWAs, which is the per-

centage of total pore volume of bulk sample relative

to its own volume; it includes the volume of the

sealed pores also.

Aggregate volume density, true density and true

porosity were determined by using Eq. (8), Eq. (9)

and Eq. (10) respectively, where true density is the

weight of one cm3 of fine powder of aggregate with-

out any air in its open pores and its value was deter-

mined with the help of pycnometer as given in liter-

ature [21]. Expansion of LWAs was also determined

for five different sizes of LWAs through Vernier cal-

iper to evaluate the increase in diameter of aggre-

gates after microwave oven curing. Eq. (11) was

used to determine the percentage of expansion.

Aggregate Volume Density (𝜌𝑏) = m1

𝑣1⁄ (8)

True Density (ρd ) =

w2 − w1

w2 − w3

× density of water

(9)

True Porosity = (1 − ρ

b

ρd

)× 100 (10)

% Expansion = D2 − 𝐷1

𝐷1

× 100 (11)

2.2.2 Strength properties

Aggregate impact value (AIV) and particle

crushing strength tests were carried out to establish

the mechanical properties of AGP. AIV test, which

gives the strength of AGP under sudden or impact

loads, was carried out in accordance with BS 812-

112 [22] on oven dried sample. Similarly, particle

crushing strength test was performed on California

Bearing Ratio (CBR) apparatus to determine the

crushing value of LWAs in order to compute their

(a) (b)

(c) (d)

(e)

(a) (b)

(c) (d)

(e)

a) Grading, b) Oven dried sample, c) Crushed sample, d)

Fines passed through 2.3mm opening, e) AIV test apparatus

Aggregate Impact Value Test

Expansion Test LWAs (AGP)

(a) (b)

Determination of expansion

by Vernier caliper

a) Oven dried

sample, b) Water

immersion at room

temperature for 24

hours, c) Surface

saturation

Water Absorption Test

(c)

Particle crushing strength

test configuration


42

ability to take compressive load. The capacity of load

ring was 10 kN. The crushing strength of individual

pellet was determined in accordance with previous

researches [23, 24], where pellets of about 12-16mm

in size were placed between two parallel plates indi-

vidually and loaded diametrically until failure oc-

curred. For more reliable estimate, 5 pellets of dif-

ferent sizes were tested using strength index formula

[11]. Equations used to find aggregate impact value

(Eq. (12)) and particle crushing strength value (Eq.

(13)) are as follows:

𝐴𝐼𝑉 = 𝑀2

𝑀1× 100 (12)

Individual crushing strength of pallet =2.8 P

π d 2

(13)

where: M1 = weight of sample before compaction

(g), M2 = weight of sample passing through 2.36mm

opening or sieve # 8 (g), P = failure load (kN), d =

distance between two plates (m).

3 Results and discussion

The physical and mechanical properties of syn-

thesized and natural LWAs obtained from literature

are presented in Table 3.

The results obtained from the tests performed on pro-

duced LWAs are presented in Table 4. The results

are also compared with natural and synthetic LWAs

which justify our approach towards objective of this

research.

3.1 Physical properties

3.1.1 Morphological features

The LWAs produced in this study were round

in shape as shown in Fig. 5. Before microwave cur-

ing, AGP were shiny with smooth surface texture. Af-

ter microwave curing, AGP remained smooth textured

with small exposed pores, however, large number of

pores were generated inside the AGP. It was observed

that before microwave curing the color of AGP was

dark grey, while after microwave curing, a slight

change in color was observed with internal dark grey

core. Since, grading determines the activator require-

ment and binder content for geopolymer concrete;

various sizes of AGP were produced in this study as

shown in Fig. 6. It was observed that average particle

size for AGP was 13.2 mm, with the smallest and larg-

est size of 11 mm and 17 mm, respectively. Moreo-

ver, all the produced aggregates were coarse aggre-

gates as they retained on 4.75 mm opening (Sieve #

4).

3.1.2 Density and percentage void

Aggregate density is considered to be a conclu-

sive parameter for determining the unit weight of

concrete and consequently, the dead load of concrete

structures. The loose and compacted bulk densities

of AGP were found to be 699kg/m3 and 738 kg/m3 re-

spectively. The compacted bulk density greatly de-

pends upon the shape and size of aggregates, deter-

mining the degree of compaction and presence of

voids between aggregates; the percentage of voids

was found to be 5.58% for produced LWAs. It was

observed that loose bulk density of specimen was

less than benchmark (880 kg/m3) given by ACI

213R-03 [36], which verified their applicability as

LWA. It was also noted that: (1) LWAs manufac-

tured in this study were lighter than many previously

Fig. 5 – Particle shape, color, and surface texture of (AGP)

Fig. 6 – Grading of (AGP)


43

Table 3 – Raw materials, binders, physical and mechanical properties of natural and synthetic LWAs briefed

in literature

Ref. Binder Alkaline

Activators

Size LBD WA SG PCS AIV Curing

Method

(mm) (kg/m3) (%) OD SSD (MPa) (%)

[11] FA + Me-

takaolin

NaOH (8M) 12 794 29.75 1.16 1.50 2.07 (12mm),

2.03 (10mm)

---- Cold

bonding

[11] FA +

Bentonite

NaOH (8M) 14 867 30.90 1.49 1.96 3.26. (14mm),

2.96 (16mm)

---- Cold

bonding

[25]

FA +

Cement

---- 4-12.5 840

15.00 1.46 8.7 ---- Sintering at

(>900oC)

[26] FA +

Cement

---- ---- 830 16.80 1.40 ---- 27.78 Sintering at

(1000-

1200oC)

[27] FA +

Cement

NaOH,

Na2SiO3

8.125 878 20.25 ---- ---- 22.10 Oven

curing at

70oC

for 24 h

[27] FA +

GBFS

NaOH,

Na2SiO3

8.125 809 28.30 ---- ---- 27.90 Cold

bonding

[28] FA NaOH,

Na2SiO3

9.5-19 789 25.50 1.30 1.63 3.70 ---- Cold

bonding

[28] FA NaOH,

Na2SiO3

9.5-19 933 0.70 1.56 1.57 12 ---- Sintering

[10] FA NaOH,

Na2SiO3

8-10 783 18.19 ---- ---- ---- ---- Sintering

[28] FA NaOH,

Na2SiO3

9.5-19 936 0.70 1.59 1.60 9.60 ---- Cold

bonding

[29] FA +

GBFS

NaOH,

Na2SiO3

10-20 903 10.60 ---- 5.70 ---- Cold

bonding

[29] FA +

GBFS

NaOH,

Na2SiO3

10-20 1002 8.30 ---- 15.50 ---- Cold

bonding

[30] BA + Ce-

ment

---- ---- 938 25.00 1.48 4.00 35.70 Sintering

[30] BA + Ce-

ment

---- ---- 1017 21.50 1.57 5.35 29.20 Sintering

[31] FA + Ben-

tonite

---- ---- 933 0.7 1.56 12 28.00 Cold

bonding

[31] FA + Glass

powder

---- ---- 936 0.7 1.59 9.6 30.00 Cold

bonding

[32] Diatomite ---- 500 7.6 ---- ---- ---- Natural

[33] Pumice ---- 475 25.00 0.80 ---- ---- Natural

[33] Expanded perlite ---- 40 70.00 2.20 ---- ---- Natural

[34] Pumice ---- 0.82-2.17 1.49-1.96 ---- Natural

[26] Natural LWA ---- 1490 0.90 2.65 ---- 15.63 Natural

[35] Calcined diatomite

aggregate

4.75-

12.5

417 112.0 2.45 ---- ---- Natural


44

Table 4 – Physical and mechanical properties of produced LWAs (FA20-80SF)

Property Value Property Value

Loose bulk density (kg/m3) 699 Aggregate size 11-17mm

Compacted bulk density (kg/m3) 738 Aggregate impact value (%) 10.24

Voids (%) 5.58 Particle crushing strength (MPa) 3.96 (max)

Porosity (%) 31.93 Specific Gravity (OD, SSD) (1.4, 1.7)

Water absorption (%) 18.98

reported densities of artificial LWA cured by other

methods, where densities ranged between 789-

1017kg/m3 (Table 3); and (2) they were heavier than

some natural LWAs like pumice, expanded perlite

and diatomite with densities equal to 475kg/m3 [33],

40kg/m3 [33] and 500kg/m3 [32], respectively. A

graphical comparison of natural LWAs, synthetic

LWAs from literature and produced LWAs (AGP) is

shown in Fig. 7.

3.1.3 Porosity

Aggregate total porosity test roots for determin-

ing the percentage of total pores in aggregate. Total

porosity of prepared pellets was determined, and the

observed values are presented in Table 5. LWAs are

exposed to heating process during their formation

which causes the expansion of LWAs [37]. This ex-

pansion leads towards the introduction of closed

pores in aggregate’s inner anatomy, causing a signif-

icant increase in its total or true porosity. The total

porosity of prepared pellets was found to be 31.93%

with true density of 2010kg/m3. Maximum total po-

rosity of LWAs can be up to 67% as given in litera-

ture [38]. Thus, AGP porosity fell in the range of pre-

scribed true porosity for LWAs. It can be observed

that AGP exhibited greater value of true porosity than

literature-reported LWAs having true porosity in the

range of 6.20-31.10% [28]. Literature has shown that

some natural LWAs like pumice and scoria have true

porosity equal to 59.06% and 49.04%, respectively

[39]. Therefore, it can be deduced that AGP are heav-

ier than natural LWAs because porosity and density

are inversely related to each other [27] as shown in

Fig. 8.

Fig. 7 – Comparison of loose bulk densities of pro-

duced LWAs and previous LWAs

Fig. 8 – Comparison between porosity and density of

different LWA

Table 5 – Observations for aggregate porosity test

Mix Density True density True porosity

g/cm3 kg/m3 g/cm3 kg/m3 %

FA20-80SF 1.367 1367 2.013 2012 31.93

Table 6 – Observations of specific gravity and water absorption

Mix Type Specific gravity Water absorption

OD SSD Apparent %

FA20-80SF 1.4 1.7 2.0 18.98

417kg/m3475kg/m3

878kg/m3

1017kg/m3

933kg/m3

699kg/m3

0

10

20

30

40

50

60

70

0

200

400

600

800

1000

1200

Diatomite Pumice Oven cured Sintered Coldbonded Microwaved

Lo

ose

Bu

lk D

en

sit

y (l

b/f

t3)

Lo

ose

Bu

lk D

en

sit

y (k

g/m

3)

Aggregate Sample Types

Loose bulk density

ACI 213R Value

L.B lb/ft3

LWAs Briefed in Literature AGP

[33] [31] [24] [28] FA20-80SF[26]

0

150

300

450

600

750

900

1050

0

10

20

30

40

50

60

70

80

Pumice Scoria Microwaved Sintered ColdbondedD

en

sit

y (k

g/m

3)

Tru

e P

oro

sit

y (%

)


True PorosityThreshold Porosity (<67%)Density

AGP Synthetic LWAsNatural LWAs

[37] [37] [26] [26]


45

3.1.4 Water absorption and specific gravity

Water absorption (WA) of prepared pellets with

different aggregate sizes ranging between 11-17mm

was calculated after 24 h immersion in water, and the

observations are presented in Table 6. The WA of

AGP was found to be 18.98% which was within the

normal range for LWAs (<25%) in accordance with

ACI-213R [36]. It was observed that WA of many

previously synthesized LWAs lies in the range of

0.70-30.90% as shown in Figure 9. It was seen that

AGP exhibited lesser WA and higher density as com-

pared to the natural LWAs like expanded perlite and

pumice, as their WA values were equal to 70.00%

and 25%, respectively [33]. Thus, more water ab-

sorption of aggregates is associated to lesser density

of aggregates, which is an indication of porous mi-

crostructure as shown in Figure 10. However, most

of the commercial artificial LWAs exhibit water ab-

sorption within 10-18% [40].

Specific gravities (oven dry (OD), saturated sur-

face dry (SSD) and apparent) of AGP were calculated

using Eq. (4), Eq. (5) and Eq. (6), respectively. Spe-

cific gravity (OD) of AGP was found to be 1.4 which

was within the range of 1.16-1.59 of literature-re-

ported synthetic LWAs as shown in Table 3. Accord-

Fig. 9 – Comparison of water absorption of produced

LWAs and previous LWAs

Fig. 10 – Comparison between water absorption and

density of various LWAs

ing to ACI-213R [36], the specific gravity of LWAs

is 1/3 to 2/3 of normal weight aggregates. So, the

manufactured LWAs fulfill the requirements of

AC1-213R. AGP showed greater specific gravity as

compared to natural LWA (pumice) having specific

gravity (OD) equal to 0.82 [34]; concluding, natural

LWAs exhibit lower specific gravities as compared

with synthetic ones as evident from Figure 11.

3.2 Mechanical properties

3.2.1 Particle crushing strength

Particle crushing strength test was conducted on a

range of produced LWAs (13-17mm) as shown in

Table 7. The highest crushing strength of 3.96 MPa

was recorded for particle size of 15mm. It was ob-

served that particle crushing strength increased as the

size of aggregate increased. However, inconsistency

was witnessed in predicting the trend for particle size

of 17mm, which might be there due to non-uni-

formity of particle shape. It was examined that parti-

cle crushing strength of produced LWAs fell in the

range of 2.03-12.00 MPa observed for literature-re-

ported LWAs with particle sizes ranging between

10-20mm as shown in Table 3. In addition, particle

crushing strength of produced LWAs was greater

than that of natural LWAs (1.49-1.96 MPa) [26]. As

natural LWAs (lighter in nature) have lesser strength

thus, it can be deduced conclusively that density and

strength are directly related with each other as shown

in Figure 12.

Fig. 11 – Comparison between specific gravity and

density of LWAs

Table 7 – Observations of particle crushing strength

test

Particle size Aggregate strength

(mm) (MPa)

13 3.66

14 3.94

15 3.96

17 3.08

7.6%

25.0%

20.2%

16.8%

25.5%

18.98%

0

5

10

15

20

25

30

35

Diatomite Pumice Oven cured Sintered Coldbonded Microwaved

Wate

r A

bso

rpti

on

(%

)


Water Absorption (%)

ACI 213R Value (<25%)

[31]LWAs Briefed in

LiteratureAGP

[30] [25] [24] [26] FA20-80SF

300

400

500

600

700

800

900

1000

5

10

15

20

25

30

35

Coldbonded Sintered Sintered Microwaved Pumice

De

nsit

y (k

g/m

3)

Wate

r A

bso

rtio

n (

%)


Water Absorption

Loose bulk density

Synthetic LWAs AGPNatural

LWA

[9] [32][27] [24] FA20-80SF

1.59 1.561.49

1.4

0.8

300

400

500

600

700

800

900

1000

0.1

0.3

0.5

0.7

0.9

1.1

1.3

1.5

1.7

1.9

Coldbonded Sintered Coldbonded Microwaved Pumice

De

nsit

y (

kg

/m3)

Sp

ecif

ic G

rav

ity


Specific Gravity

Density

[32]


LWA

[26] [26] [10] FA20-80SF


46

Fig. 12 – Comparison between particle crushing

strength and density of LWAs

3.2.2 Aggregate impact value

The aggregate impact value (AIV) test was car-

ried out on produced LWAs. The lower the impact

value is, better will be the resistance of aggregates

against impact loads. BS 882-1992 [41] describes the

maximum impact value as 25% when aggregate is to

be used in heavy duty floors, 30% when aggregate is

to be used in concrete for wearing surfaces, and 45%

for other concrete applications. AGP exhibited good

impact value of 10.24% and they were envisaged to

be comparatively stronger than LWAs formulated

and reported in previous research works having AIV

in the range of 22.10-35.70% as shown in Figure 13.

It was observed that AGP appeared to be stronger than

natural LWA having AIV equal to 15.63% [26] and

this was because of their greater density relative to

natural LWA as shown in Figure 13.

4 Conclusions

In this work, LWAs were produced through ge-

opolymerization by using FA and SF as precursors.

Microwave heating (5 min) was adopted as curing

regime. Physical and mechanical properties of pre-

pared LWAs were investigated and compared with

other synthetic LWAs as well as natural LWAs. The

main conclusions obtained from the experimental

work can be summarized as follows:

(1) The aggregates presented smooth surface with

small tiny pores. Physical properties such as

density, water absorption and specific gravity of

produced LWAs followed the specified ranges

of ACI standard for LWAs.

(2) The loose and compacted bulk densities of

LWAs were found to be 699 kg/m3 and 738

kg/m3, respectively, which was within ACI limit

(<880 kg/m3) mentioned for structural LWA. It

was ensured that the produced LWAs were

lighter than many previously formulated LWAs.

Fig. 13 – Comparison between aggregate impact

value and density of LWAs

However, these were heavier than some natural

LWAs such as pumice, expanded perlite and di-

atomite.

(3) Similarly, total porosity (31.93%) of produced

LWAs was lesser than the synthetic LWAs re-

ported in literature, but higher than the natural

LWAs. Water absorption of produced LWAs

was 18.98% which was lesser than natural

LWAs as well as ACI limit (<25%) for struc-

tural LWAs. It suggests that LWAs can be used

to produce structural concrete. More water ab-

sorption of LWAs is attributed towards lesser

density, which indicates porous microstructure

of LWAs.

(4) The produced LWAs exhibited good mechani-

cal properties. The maximum particle crushing

strength was found to be 3.96 MPa for aggre-

gate size of 15 mm. Higher strength character-

istics were observed for the produced LWAs in

comparison to the natural LWAs thus indicating

the direct relation of density and particle crush-

ing strength.

(5) The impact value of 10.24% was observed for

produced LWAs, which shows its better re-

sistance against impact load than both previ-

ously developed synthetic LWAs and natural

LWA -that have been used in concrete. The ob-

tained impact value confirms the applicability

of produced LWAs for heavy duty floors and in

other concrete applications as well, according to

BS 882-1992.

Proposed curing methodology is able to pro-

duce LWAs in just 5 minutes and may have a strong

potential to be used at industrial scale. Detailed anal-

ysis with respect to time savings and energy savings

must be carried out and is strongly recommended for

future works.

9.6 MPa

5.7 MPa

3.7 MPa 3.66 MPa

1.72 MPa

300

400

500

600

700

800

900

1000

1

2

3

4

5

6

7

8

9

10

11

12

FA FA+GBFS FA FA20-80SF Pumice

De

nsit

y (

kg

/m3)

Part

icle

Cru

sh

ing

Str

en

gth

(M

Pa)


Particle Crushing Strength Density

[26] [32]


LWA

[26] [27]

300

400

500

600

700

800

900

1000

6

12

18

24

30

36

42

48

Sintered Coldbonded Coldbonded Oven cured Microwaved

De

nsit

y (k

g/m

3)

Ag

gre

gate

Im

pact

Valu

e (

%)


Aggregate Impact Value

BS-882 value for heavy duty floor < 25%

BS-882 value for wearing surfaces < 30%

Density

LWAs Briefed in Literature

AGP

[28] [26] [26] [25] FA20-80SF


47

List of Abbreviations

LWA Lightweight aggregate

FA Fly ash

SF Silica fume

GBFS Ground granulated blast furnace slag

AGP Geopolymer light weight aggregates

LBD Loose bulk density (kg/m3)

CBD Compacted bulk density (kg/m3)

SSD Saturated surface dry

OD Oven dry

Cl Chlorine (%)

LOI Loss on ignition (%)

𝑤𝐿𝐴+𝐶 Weight of loose aggregate and con-

tainer (g)

𝑤𝐶𝐴+𝐶 Weight of compacted aggregate and

container (g)

𝑤𝐿𝐴 Weight of loose aggregate (g)

𝑤𝐶𝐴 Weight of compacted aggregate (g)

𝑤𝐶 Weight of empty container (g)

𝑉𝐶 Volume of container (m3)

𝑚𝑂𝐷 Mass of oven dry test sample in air (g)

𝑚𝑆𝑆𝐷 Mass of surface saturated dry test

sample in air (g)

𝑚𝐴𝑃 Apparent mass of saturated test sam-

ple in water (g)

m1 Mass of single aggregate (g)

v1 Volume of single aggregate having

external pores with access of water

and internal pores without access of

water (m3)

w1 Weight of pycnometer filled with wa-

ter (g)

w2 Weight of pycnometer filled with wa-

ter and fine powder aggregate sample

(g)

w3 Weight of pycnometer filled with fine

powder aggregate sample and water

(g)

D1 Diameter of aggregate before curing

(mm)

D2 Diameter of aggregate after curing

(mm)

AIV Aggregate impact value (%)

M1 Weight of sample before compaction

(g)

M2 Weight of sample passing through

2.36mm opening or sieve # 8 (g);

P Failure load (kN)

D Distance between two plates (m)

WA Water absorption (%)

PCS Particle crushing strength (MPa)

SG Specific gravity

𝜌𝑏 Aggregate volume density (kg/m3)

𝜌𝑑 True density (kg/m3)

References

1. Aslam, M.; Shafigh, P.; Jumaat, M.Z.; and

Lachemi M. (2016) “Benefits of using blended

waste coarse lightweight aggregates in structural

lightweight aggregate concrete,” Journal of

Cleaner Production, vol. 119, pp. 108-117.

2. Kisku, N.; Joshi, H.; Ansari, M.; Panda, S.;

Nayak, S.; and Dutta, S.C. (2017) “A critical re-

view and assessment for usage of recycled ag-

gregate as sustainable construction material,”

Construction and Building Materials, vol. 131,

pp. 721-740.

3. Mannan, M.; and Ganapathy, C. (2001) “Mix

design for oil palm shell concrete,” Cement and

Concrete Research, vol. 31, no. 9, pp. 1323-

1325.

4. Oktay, H.; Yumrutaş, R.; and Akpolat, A. (2015)

“Mechanical and thermophysical properties of

lightweight aggregate concretes,” Construction

and Building Materials, vol. 96, pp. 217-225.

5. Mohammed, J.H.; and Hamad, A.J. (2014) “Ma-

terials, properties and application review of

Lightweight concrete,” Technical Review of the

Faculty of Engineering University of Zulia, vol.

37, no. 2, pp. 10-15.

6. Tajra, F.; Elrahman, M.A.; and Stephan, D.

(2019) “The production and properties of cold-

bonded aggregate and its applications in con-

crete: A review,” Construction and Building


7. Priyanka, M.; Karthikeyan, M.; and Sri Rama

Chand, M. (2020) “Development of mix propor-

tions of geopolymer lightweight aggregate con-

crete with LECA,” Materials Today: Proceed-

ings, vol. 27, pp. 958-962.

8. Bijen, J. (1986) “Manufacturing processes of ar-

tificial lightweight aggregates from fly ash,” In-

ternational Journal of Cement Composites and

Lightweight Concrete, vol. 8, no. 3, pp. 191-199.


48

9. Franus, M.; Panek, R.; Madej, J.; and Franus, W.

(2019) “The properties of fly ash derived light-

weight aggregates obtained using microwave ra-

diation,” Construction and Building Materials,

vol. 227 pp. 116677.

10. Nadesan, M.S.; and Dinakar, P. (2017) “Struc-

tural concrete using sintered flyash lightweight

aggregate: A review,” Construction and Build-


11. Gomathi, P.; and Sivakumar, A. (2014) “Cold

Bonded Fly Ash Lightweight Aggregate Con-

taining Different Binders,” Research Journal of

Applied Sciences, Engineering and Technology,

vol. 7, no. 6, pp. 1101-1106.

12. Sharif, S.; Mahmad Nor, A.; Yahya, Z.; Abdul-

lah, M.M.A.B.; Abdul Razak, R.; Jaya Ekaputri,

J.; Faris, M.A.; Nur Hamzah, H.; Abdullah,

M.M.A.B.; Abd Rahim, S.Z.; Ghazali, M.F.;

Mat Saad, N.; Ramli, M.M.; Zainol Murad, S.A.;

and Mat Isa, S.S. (2016) “A Review on the Man-

ufacturing of Lightweight Aggregates Using In-

dustrial By-Product,” MATEC Web of Confer-

ences, vol. 78, pp. 01067.

13. Domagała, L. (2011) “Modification of proper-

ties of structural lightweight concrete with steel

fibres,” Journal of Civil Engineering and Man-

agement, vol. 17, no. 1, pp. 36-44.

14. Topçu, İ.B.; and Uygunoğlu, T. (2007) “Proper-

ties of autoclaved lightweight aggregate con-

crete, Building and Environment, vol. 42, no. 12,

pp. 4108-4116.

15. Reddy, M.V.S.; Nataraja, M.; Sindhu, K.; Ha-

rani, V.; and Madhuralalasa, K. (2016) “Perfor-

mance of light weight concrete using fly ash pel-

lets as coarse aggregate replacement,” Interna-

tional Journal of Engineering Research and

Technology, vol. 9, no. 2, pp. 95-104.

16. Thostenson, E.; and Chou, T.-W. (1999) “Mi-

crowave processing: fundamentals and applica-

tions,” Composites Part A: Applied Science and

Manufacturing, vol. 30, no. 9, pp. 1055-1071.

17. C. ASTM, Standard specification for coal fly

ash and raw or calcined natural pozzolan for use

in concrete, ASTM Int 4 (2012).

18. Wang, Y.; He, F.; Wang, J.; and Hu, Q. (2019)

“Comparison of Effects of Sodium Bicarbonate

and Sodium Carbonate on the Hydration and

Properties of Portland Cement Paste,” Materials,

vol. 12, no. 7, pp. 1033.

19. ASTM C29, Standard test method for bulk den-

sity (“Unit Weight”) and voids in aggregate.

American Society for Testing and Materials:

West Conshohocken, PA, USA. , (2017).

20. ASTM C127–14, Standard Test Method for

Density, Relative Density (Specific Gravity),

and Absorption of Coarse Aggregate, ASTM In-

ternational. West Conshohocken, Pennsylvania

19428–2959, United States (2014).

21. Borys, M.; Kurpińska, M.; Ferenc, T.; Czyż, Z.;

Falkowicz, K.; Kujawska, J.; Kulisz, M.; and

Szala, M. (2017) “Effect of porosity on physical

properties of lightweight cement composite with

foamed glass aggregate,” ITM Web of Confer-

ences, vol. 15, pp. 06005.

22. B. British Standard, Testing aggregates. Method

for determination of aggregate impact value

(AIV), Vol. BS (1990) 812-112.

23. Alqahtani, F.K.; Rashid, K.; Zafar, I.; and Iqbal

Khan, M. (2021) “Assessment of morphological

characteristics and physico-mechanical proper-

ties of geopolymer green foam lightweight ag-

gregate formulated by microwave irradiation,”

Journal of Building Engineering, vol. 35, pp.

102081.

24. Li, Y.; Wu, D.; Zhang, J.; Chang, L.; Wu, D.;

Fang, Z.; and Shi, Y. (2000) “Measurement and

statistics of single pellet mechanical strength of

differently shaped catalysts,” Powder technol-

ogy, vol. 113, no. 1-2, pp. 176-184.

25. Van der Wegen, G.; and Bijen, J. (1985) “Prop-

erties of concrete made with three types of arti-

ficial PFA coarse aggregates,” International

Journal of Cement Composites and Lightweight

Concrete, vol. 7, no. 3, pp. 159-167.

26. Wang, H.-T.; and Wang, L.-C. (2007) “Experi-

mental study on mechanical properties of steel

fiber reinforced lightweight aggregate concrete,”

Jianzhu Cailiao Xuebao(Journal of Building

Materials), vol. 10, no. 2, pp. 188-194.

27. Ul Rehman, M.; Rashid, K.; Ul Haq, E.; Hussain,

M.; and Shehzad, N. (2020) “Physico-mechani-

cal performance and durability of artificial light-

weight aggregates synthesized by cementing

and geopolymerization,” Construction and

Building Materials, vol. 232, pp. 117290.

28. Kockal, N.U.; and Ozturan, T. (2011) “Durabil-

ity of lightweight concretes with lightweight fly

ash aggregates,” Construction and Building Ma-

terials, vol. 25, no. 3, pp. 1430-1438.

29. Bui, L.A.T.; Hwang, C.L.; Chen, C.T.; Lin, K.L.;

and Hsieh, M.Y. (2012) “Manufacture and per-

formance of cold bonded lightweight aggregate

using alkaline activators for high performance

concrete,” Construction and Building Materials,

vol. 35, pp. 1056-1062.

30. Saad, M.; Baalbaki, O.; Khatib, J.; El Kordi, A.;

and Masri, A. (2019) “Manufacturing of Light-

weight Aggregates from Municipal Solid Waste

Incineration Bottom Ash and their Impacts on

Concrete Properties,” 2nd international Con-

gress on Engineering and Architecture, At Mar-

maris / Turkey, pp. 1638-1649.


49

31. Bui, L.A.T.; Hwang, C.L.; Chen, C.T.; and

Hsieh, M.Y. (2012) “Characteristics of Cold-

Bonded Lightweight Aggregate Produced with

Different Mineral Admixtures,” Applied Me-

chanics and Materials, vol. 174-177, pp. 978-

983.

32. Mehmedi Vehbi GÖ KÇ , E. (2012) “Use of diat-

omite in the production of lightweight building

elements with cement as binder,” Scientific Re-

search and Essays, vol. 7. No. 7, pp. 774-781.

33. Karakurt, C.; and Ö zen, U. “Influence of natural

lightweight aggregates on the properties of con-

crete,” Journal of the Turkish Chemical Society

Section B: Chemical Engineering 1(Sp. is. 2), pp.

9-16.

34. Rashad, A.M. (2019) “A short manual on natu-

ral pumice as a lightweight aggregate,” Journal

of Building Engineering, vol. 25, pp. 100802.

35. Posi, P.; Lertnimoolchai, S.; Sata, V.; and

Chindaprasirt, P. (2013) “Pressed lightweight

concrete containing calcined diatomite aggre-

gate,” Construction and Building Materials, vol.

47, pp. 896-901.

36. Akers, D.J.; Gruber, R.D.; Ramme, B.W.; Boyle,

M.J.; Grygar, J.G.; Rowe, S.K.; Bremner, T.W.;

Kluckowski, E.S.; Sheetz, S.R.; and Burg, R.G.

(2003) “Guide for Structural Lightweight-Ag-

gregate Concrete,” ACI 213R-03. American

Concrete Institute (ACI), Michigan.

37. Abdulkareem, O.A.; Abdullah, M.; Hussin, K.;

Ismail, K.N. and Binhussain, M. (2013) “Me-

chanical and Microstructural Evaluations of

Lightweight Aggregate Geopolymer Concrete

before and after Exposed to Elevated Tempera-

tures,” Materials, vol. 6, no. 10, pp. 4450-4461.

38. Kurpińska, M.; and Ferenc, T. (2017) “Effect of

porosity on physical properties of lightweight

cement composite with foamed glass aggregate,”

ITM Web of Conferences, EDP Sciences, vol.

15, pp. 06005.

39. Kılıc, A.; and Teymen, A. (2009) “The effects

of scoria and pumice aggregates on the strengths

and unit weights of lightweight concrete,” Sci-

entific Research and Essay, vol. 4, no. 10, pp.

961-965.

40. Colangelo, F.; Messina, F.; and Cioffi, R. (2015)

“Recycling of MSWI fly ash by means of ce-

mentitious double step cold bonding pelletiza-

tion: Technological assessment for the produc-

tion of lightweight artificial aggregates,” Jour-

nal of hazardous materials, vol. 299, pp. 181-

191.

41. Manikandan, R.; and Ramamurthy, K. (2008)

“Effect of curing method on characteristics of

cold bonded fly ash aggregates,” Cement and

concrete composites, vol. 30, no. 9, pp. 848-853.



ISSN 2465-7964 / eISSN 2465-7972

https://doi.org/10.18702/acf.2020.12.6.2.50

50

Technical Paper

Evaluation of mechanical and durability properties of con-

crete made with Indian bottom ash as replacement of fine ag-

gregate

P. N. Ojha, Amit Trivedi, Brijesh Singh*, Abhishek Singh

(Received June 5, 2020; Revised October 28, 2020; Accepted December 5, 2020; Published December 31, 2020)

Abstract: Bottom ash is a major by-product of the coal-based power generation process and it has particle

size ranging from 45 μm to 150 μm. As per current provisions of IS: 383-2016, bottom ash can be used

as replacement of natural fine aggregate up to 25% in case of lean concrete (less than M15 Grade) only.

However, its use in reinforced and plain concrete is not permitted. Therefore, it is imperative to study the

feasibility of using coal based bottom ash as a replacement of conventional fine aggregates (i.e., natural

and crushed sand) in plain and reinforced concrete to increase the utilization of this industrial byproduct.

In this study, natural and crushed sand were replaced with bottom ash at various percent-ages for prepa-

ration of concrete and study its effect on fresh, hardened and durability properties of concrete. Bottom

ash was collected from Vindhyachal thermal power plant of India. Experimental studies were conducted

at w/c ratio of 0.65 and 0.40. Concrete mixes were studied and analyzed for various mechanical and

durability properties. Based on fresh concrete properties i.e., workability, slump retention and strength

development, it was observed that up to 50% replacement of conventional fine aggregate with bottom

ash is technically feasible.

Keywords: Bottom ash; fine aggregate; characterization; mechanical property; durability.

1 Introduction

Bottom ash is a major by-product of the coal-

based power generation process. In coal based Ther-

mal Power Plant, at the bottom of the furnace, there

is a hopper for collection of bottom ash. The bottom

ash can be collected by wet cooling and wet removal

process or dry cooling and dry removal process from

the bottom of boilers. Characteristics of bottom ash

depend on the process of removal of bottom ash from

the boiler. In wet cooling and wet removal process,

a hopper is always filled with water to quench the

ash. Bottom ash consists of heavier particles that fall

to the bottom of the furnace. Bottom ash is composed

primarily of amorphous or glassy alumino-silicate

materials derived from the melted mineral phases.

Bottom ash differs from fly ash collected from elec-

trostatic precipitators in a dry form in that it contains

significant amount of relatively coarser particles

(greater than 45 μm and up to 150 μm). Coal bottom

ash has angular, irregular, porous and rough surface

textured particles. Coal bottom ash is lighter and

more brittle as compared to natural river sand. The

specific gravity of coal bottom ash varies from 1.8 to

2.6 depending upon the source and type of coal. Coal

bottom ash derived from high Sulphur coal and low

rank coal is not very porous and is quite dense. In

India, over 70% of electricity generated is by coal

fired plants. As per Central Electricity Authority [1]

data 2014-15 the annual production of Ash is 180MT

out of which 30-35MT is bottom ash and rest is fly

ash.

BIS has incorporated the provision of manufac-

tured aggregates to be used in concrete in IS: 383-

2016. It mentions that bottom ash can be used as re-

placement of natural fine aggregate up to 25% in

case of lean concrete (less than M15 Grade) only.

However, it is not permitted to use bottom ash in re-

inforced and plain concrete. At national and interna-

tional level, researchers have carried out study on us-

age of bottom ash as a replacement of fine aggregate

at different percentage levels ranging from 10% to

100%. Fresh and hardened concrete properties

P. N. Ojha is a Joint Director at National Council for Cement

& Building Materials, India.

Amit Trivedi is a General Manager at National Council for Ce-


Corresponding author Brijesh Singh is a Group Manager at

National Council for Cement & Building Materials, India.

Abhishek Singh is a Project Engineer at National Council for

Cement & Building Materials, India.


51

(strength-based properties) have been studied. How-

ever, limited studies are available on durability prop-

erties of concrete. Abdulhameed et al. [2] have car-

ried out the studies on usage of bottom ash as re-

placement of fine aggregate in concrete. The coal

bottom ash was obtained from Tanjung Bin power

plant. Natural sand was partially replaced with bot-

tom ash in the range of 5, 10, 15 and 20%. Decrease

in the workability of fresh concrete in terms of slump

and compacting factor was observed with increase in

percentage of bottom ash. Reduction in compressive

strength and density of concrete was observed with

increase in percentage of bottom ash.

Ratchayut et al. [3] have carried out the studies

on usage of bottom ash as replacement of fine aggre-

gate in concrete. The study was conducted on self-

compacting concrete having w/binder ratio of 0.31.

Bottom ash was obtained from Saraburi Power Plant,

Thailand. Natural sand was replaced with bottom ash

up to 30%. Slump flow decreased continuously with

increase in bottom ash content. It was considered that

such decrease was due to increased aggregate to ag-

gregate friction from highly irregular shape and

rough texture of bottom ash particles. Mehdi et al. [4]

carried out the studies on usage of bottom ash as re-

placement of fine aggregate in concrete. Coal bottom

ash was obtained from Malaysian Power Plant. Con-

crete specimens were prepared incorporating 0, 20,

50, 75 and 100% of bottom ash replacing sand and

20% of coal fly ash by mass, as a substitute for ordi-

nary Portland cement. Workability of concrete re-

duces on increasing the percentage of bottom ash. At

the age of 28 days, no significant effect was observed

in compressive, flexural and tensile strengths of all

concrete samples. The drying-shrinkage of experi-

mental concrete mixtures containing 50%, 75% and

100% Bottom ash and 20% fly ash was lower than

the control mix.

Kim et al. [5] have carried out the studies on

chloride resistance of high-strength concrete incor-

porating bottom ash. The results showed that, alt-

hough there was no significant effect on the chloride

diffusion, bottom ash in high-strength concrete can

significantly reduce the amount of chloride diffusion

as chloride did not readily diffuse to the cement paste

in the vicinity of bottom ash. Malkit et al. [6] carried

out the studies on usage of bottom ash as replace-

ment of fine aggregate in concrete. At fixed water

cement ratio, workability and loss of water from

bleeding decreased with the use of coal bottom ash

as a replacement of river sand in concrete. Compres-

sive strength of bottom ash concrete at the curing age

of 28 days was not significantly affected. However,

after 90 days of curing age, compressive strength of

bottom ash concrete surpassed that of conventional

concrete. Splitting tensile strength of concrete im-

proved at all the curing ages. The modulus of elas-

ticity decreased with the use of coal bottom ash at all

the curing ages. Andrade et al. [7] carried out the

studies on the influence of the use of coal bottom ash

as a replacement for natural fine aggregates on the

properties of concrete in the fresh state. In the fresh

state the concretes produced with the bottom ash are

susceptible to water loss by bleeding and the higher

the percentage of bottom ash used as a natural sand

replacement the lower the deformation through plas-

tic shrinkage. Aggarwal et al. [8] carried out studies

on concrete (w/c of 0.43) by replacing up to 40% fine

aggregate (by weight) with coal bottom ash obtained

from thermal power plant in Panipat, Haryana (India)

was used in the investigation. The density of con-

crete decreased with the increase in bottom ash con-

tent due to the low specific gravity of bottom ash as

compared to fine aggregates. Mix containing 30%

and 40% bottom ash, at 90 days, attains the compres-

sive strength equivalent to 108% and 105% of com-

pressive strength of normal concrete at 28 days and

attains flexural strength in the range of 113-118% at

90 days of flexural strength of normal concrete at 28

days. Kadam et al. [9] carried out studies on concrete

(w/c of 0.45) by replacing up to 100% fine aggregate

(by weight) with coal bottom ash from Eklahare ther-

mal power plant in India. The compressive strength

for 7, 28, 56 and 112 days was increased up to 20%

replacement and after that compressive strengths

were decreased from 30% to 100% replacement. The

split tensile and flexural strength was increased at 7,

28, 56 and 112 days for 10% to 30% replacement and

after that it was decreased for remaining replacement.

Arumugam et al. [10] carried out studies on concrete

(w/c of 0.5) by replacing up to 60% fine aggregate

(by weight) with coal bottom ash study. The unit

weight of concrete gets reduced through the addition

of bottom ash as replacement of fine aggregate since

it has lesser specific gravity than fine aggregate. The

7 days, 28 days and 56 days strength shows that the

strength increases from standard concrete up to the

addition of 20% replacement of fine aggregate with

bottom ash.

Raju et al. [11] carried out studies on concrete

by replacing up to 30% fine aggregate (by weight)

with coal bottom ash obtained from Hindustan News

Print Limited, Kottayam, Kerala (India). Slump re-

duced with increase in percentage of bottom ash due

to higher water absorption of bottom ash. Compres-

sive strength, split tensile strength and flexural

strength increased up to 20% replacement of Bottom

ash. Based on the review of existing literature, it can

be inferred that there is a potential for use of bottom

ash as replacement of fine aggregate in concrete.


52

2 Experimental plan

2.1 Sampling, screening and separation of bot-

tom ash into different fractions

In this study ‘B’ stands for bottom ash and ‘Y’

stands for Vindhyachal. About 200 bags containing

Bottom Ash were collected from Vindhyachal site.

Bottom ash “as such” i.e.,, without screening/sieving

has been referred as ‘BY’ in this study. After that,

separation of bottom ash samples was carried out by

mechanical sieving. No part of sample was retained

on sieve size 4.75mm. Material finer than 75 µm was

found to be 9.6% by wet sieving method, while it was

8-11% by dry sieving method. When, material finer

than 75 µm was removed from the BY (as such sam-

ple), it was designated as ‘BY2’ (fraction between

4.75 mm and 75 µm) for this study. Both bottom ash

samples BY and BY2 were used to replace fine ag-

gregates for preparation of concrete for further study.

2.2 Study of fundamental properties of bottom

ash and other concrete making materials

Studies on fundamental properties of bottom

ash samples (BY and BY2) by conducting physical,

chemical and microstructural characterization of all

the three fractions of bottom ash separately in order

to assess the feasibility of the use of bottom ash as

construction material were carried out. Characteriza-

tion by means of analysis of engineering properties

of bottom ash as fine aggregate in concrete includes

properties such as specific gravity, fineness, grada-

tion, texture, physical and chemical characteristics

etc. This also included petrographic examinations,

Scanning electron microscopy (SEM) examination

and X-ray diffraction (XRD) analysis. Along with

evaluation of bottom ash samples, other concrete

making materials such as OPC 43, aggregates 20 mm,

10 mm, natural sand and crushed sand and PC

based/naphthalene-based superplasticizer were also

evaluated.

2.3 Replacement of conventional sand by bot-

tom ash in concrete mixes

Varying proportions of bottom ash and fine ag-

gregate were tried in an effort to determine the opti-

mum ratio of bottom ash to fine aggregate. The per-

formance of concrete was evaluated in terms of fresh

concrete properties, mechanical properties and dura-

bility properties. The present study shall include 25,

50, 75 and 100% replacement of natural sand and

crushed sand by bottom ash “as such” (BY) and frac-

tion between 4.75 mm and 75 µm (BY2). The con-

crete mixes given in Table 1 below shall be studied:

2.4 Casting and testing of concrete samples

Casting and testing of concrete samples as per

relevant IS/ASTM/DIN/ISO methods were carried

out to determine the engineering properties/charac-

teristics of mixes. Fresh concrete properties such as

slump, air content, wet density and initial & final set-

ting time of concrete along with compressive

strength at 3, 7, 28 and 56 days were evaluated for

all the 32 mixes. Hardened concrete properties such

as flexural strength, static modulus of elasticity

along with drying shrinkage and moisture movement

were evaluated at the age of 28 days for 5 selected

experimental mixes and 2 control mixes. Evaluation

of durability properties of concrete such as pH value,

water permeability, volume of permeable voids, Wa-

ter absorption, rapid chloride penetration test, elec-

trical resistivity using four-point Wenner probe, air

permeability and accelerated carbonation test were

carried out for 5 selected experimental mixes and 2

control mixes.

Table 1 – Details of level of replacements and total number of mixes

w/c % fine aggregate replacement by bottom ash No. mixes

0.65 and

0.40

Without Bottom Ash i.e., Control Mixes, with natural river sand (100 %)

& crushed sand (100 %) 4

0.65 and

0.40

100 % of BY2 (i.e., after removing particles greater than 4.75 mm and

less than 75µm) as fine aggregate. 2

0.65 and

0.40 100 % BY (as such) as fine aggregate. 2

0.65 and

0.40

25, 50 and 75 % replacement of natural river sand by BY2 (i.e., After re-

moving particles greater than 4.75 mm and less than 75µm) at 6

0.65 and

0.40

25, 50 and 75 % replacement of crushed sand by BY2 (i.e., after removing

particles greater than 4.75 mm and less than 75µm) at 6

0.65 and

0.40 25, 50 and 75 % replacement of natural river sand by BY(as such) 6

0.65 and

0.40 25, 50 and 75 % replacement of crushed sand by BY(as such) 6

Total no. of mixes 32


53

3 Characterization of bottom ash and

concrete making materials

Only bottom ash samples bottom ash “as such”

i.e.,, without screening/sieving (BY) and fraction be-

tween 4.75 mm and 75 µm (BY2) which were used

to replace fine aggregates for preparation of concrete

were evaluated for sieve analysis, physical and

chemical properties. BY and BY-2 samples were

also subjected to petrographic examination, X-ray

diffractometry and analyzed using scanning electron

microscope.

3.1 Sieve analysis of bottom ash fractions

Sieve analysis of Bottom Ash “as such” (i.e.,,

BY) and fraction between 4.75 mm and 75 µm (i.e.,,

BY2) was carried out as per IS 383:2016 and the re-

sults are given in Table-2 and gradation curve has

been shown in Fig. 1.

For BY (bottom ash ‘as such’) sample the per-

centage passing through sieve size 600 µm is 82%

which corresponds to Zone-IV as per IS: 383-2016.

However, the percentage passing through 300 µm &

150 µm are 58% & 29% respectively which are more

than the grading requirement of Zone-IV as per IS:

383-2016 and therefore bottom ash “as such” (i.e.,,

BY) is finer than Zone-IV. For BY2 (fraction be-

tween 4.75 mm & 75 µm) sample, the percentage

passing through sieve size 600 µm is 82% which cor-

responds to Zone-IV as per IS: 383-2016. However,

the percentage passing through 300 µm and 150 µm

are 57% & 19% respectively which are more than the

grading requirement of Zone-IV as per IS: 383-2016

and therefore bottom ash BY2 (fraction between

4.75 mm & 75 µm) is finer than Zone-IV.

3.2 Physical characterization of bottom ash

The results of physical characterization of bot-

tom ash sample BY (as such) after screening from

4.75mm and BY2 (fraction between 4.75 mm and

75µm) sieve are given in Table 3.

3.3 Chemical characterization of bottom ash

The results of chemical characterization of bot-

tom ash sample (As such) after screening from

4.75mm sieve and BY2 (fraction between 4.75 mm

and 75 µm) sieve are given in Table 4.

Table 2 – Sieve analysis of bottom ash fractions BY and BY2

IS Sieve Size Percentage Passing (%) Percentage Passing for Grading Zone

IV as per IS: 383-2016 Table 9 BY BY2

10 mm 100 100 100

4.75 mm 100 100 95 – 100

2.36 mm 96 97 95 – 100

1.18 mm 92 92 90 – 100

600 µm 82 82 80 – 100

300 µm 58 57 15 – 50

150 µm 29 19 0 – 15

Fig. 1 – Gradation curve for sieve analysis of bottom ash BY and BY2


54

3.4 Mineralogical and microstructural investi-

gations

In order to investigate the possibility of occur-

rence of undesirable and deleterious phenomenon

such as alkali silica reaction or alkali aggregate reac-

tion due to presence of reactive forms of silica, both

bottom ash samples (BY and BY2) were subjected to

petrographic examination, X-ray diffractometry and

analyzed using scanning electron microscope.

3.4.1 Petrographic examination of bottom ash BY (As such): The glass content in this sample

is 11%. The other mineral constituents are quartz, or-

thoclase –feldspar, plagioclase-feldspar, muscovite,

iron oxide and other opaque minerals. Grain size of

glass varies from 4 µm to 74 µm. Glass grains are of

various shapes and sizes. Grain size variation in glass

is too large. Common form of glass grains is rounded,

sub-rounded, polygonal, lath and micro globular.

Subhedral to anhedral quartz grains with rounded

grain margins are uniformly distributed in the sample.

Grain size of quartz varies from 6 µm to 171 µm.

Subhedral to anhedral opaque minerals with cor-

roded margins are also uniformly distributed. The

microphotograph of this Bottom Ash sample is given

in Fig. 2. BY2 (fraction between 4.75 mm and 75

µm): The glass content in this sample is 16%. The

other mineral constituents are quartz, orthoclase –

feldspar, plagioclase-feldspar, iron oxide and other

opaque minerals. Grain size of glass varies from 3

µm to 52 µm. Glass grains are of various shapes and

sizes. Common form of glass grains is rounded, sub

rounded, lath, rectangular, polygonal and micro

globular. Subhedral to anhedral quartz grains with

sharp angular grain margins are uniformly distrib-

uted in the sample. Grain size of quartz varies from

4 µm to 150 µm. The microphotograph of this Bot-

tom Ash sample is given in Fig. 3.

3.4.2 X-Ray diffraction analysis (XRD) of bottom

ash

BY (as such): XRD studies of the random sam-

ple revealed the presence of quartz, mullite, tri-

dymite and hematite phases. These minerals are fur-

ther classified as predominant, major and minor con-

stituents. BY2 (fraction between 4.75 mm and 75

µm): XRD studies of the random sample revealed the

presence of quartz, mullite, tridymite and hematite

phases. These minerals are further classified as pre-

dominant, major and minor constituents. The list of

phases identified, their chemical formulae and rela-

tive abundance is given in the Table 5.

Table 3 – Results of physical properties of bottom ash samples BY and BY2

Sl.No. Test Carried out BY (As such) BY2 (fraction between 4.75

mm and 75 µm)

1 Specific gravity 2.08 2.06

2 Water absorption, % 1.5 1.7

3 Material finer than 75 µm % (wet sieving) 9.45 2.1

4 Soundness, MgSO4 % 9.23 11.4

5 Fineness modulus 1.465 1.541

6 Organic impurities % Nil Nil

7 Clay lumps % Nil Nil

8 Total deleterious material, % (except coal

& lignite) Nil Nil

9 Lime reactivity (N/mm2) 0.74 0.194

Table 4 – Results of chemical properties of bottom ash samples BY and BY2

Sl. No. Test Carried out BY (As such) BY2 (fraction between 4.75

mm and 75 µm)

1 Loss on ignition (LOI) % 1 1.58

2 Silica (SiO2) % & iron oxide (Fe2O3) % 67.2 & 12.29 56.74 & 18.84

3 Aluminum oxide (Al2O3) % 15.76 17.78

4 Calcium oxide (CaO) % 1.03 1.74

5 Magnesium oxide (MgO) % 1.11 1.25

6 Sulphate (SO3) % Nil 0.07

7 Total alkalis %: Na2O & K2O 0.09 & 0.55 0.16 & 0.66

8 Chloride (Cl), Acid soluble % 0.013 0.011

9 Reactive SiO2 % 29.15 25.38

10 Water soluble Cl % 0.004 0.004

11 Sulphide Sulphur % 0.032 0.02


55

Fig. 2 – BY- Bottom Ash (As such) distribu-

tion of mineral grains (5x)

Fig. 3 – BY2- Bottom Ash (fraction between 4.75 mm

and 75 µm) Distribution of mineral grains (5x)

3.4.3 Scanning electron microscopy examination

(SEM) of bottom ash

BY (As such): The sample contained coarse

particles with sizes ranging from 10 to 550 micron.

Large particles of unburnt carbon were abundant in

the sample. Most of the particles are irregularly

shaped and agglomerated. Few amorphous particles

were also observed. The surface of the glassy mate-

rial has rough texture. The microphotograph of this

Bottom Ash sample is given in Fig. 4. BY2 (fraction

between 4.75 mm and 75 µm): Sample predomi-

nantly contained irregular shaped crystalline com-

pounds of quartz, hematite and magnetite. The parti-

cles were having the sizes in the range of 10 to 400

microns. Most of the particles were in agglomerated

form. The microphotograph of this bottom ash sam-

ple is given in Fig. 5.

After analysis of results of petrographic exami-

nation, X-ray diffractometry and study using scan-

ning electron microscope, it was observed that no

deleterious minerals or compounds were present in

both bottom ash samples which can cause long term

durability related issues in concrete prepared using

bottom ash as a replacement of fine aggregate.

3.5 Characterization of other concrete making

materials

Cement (OPC-43), coarse aggregate (10 and 20

mm), fine aggregate (natural and crushed) and chem-

ical admixture – PC based (BASF Master Glenium

Sky 8777) and Naptha (BASF Rheobuild 1100) were

used in this study. These concrete making materials

were tested as per relevant Indian Standards and

showed conformance to the required standards.

a) Cement OPC-43: The cement sample of

OPC-43 (Ultratech Brand) was tested for

various physical and chemical properties

and the test results are presented in Table 6.

Results of OPC-43 (Ultratech Brand)

showed conformance to the requirements of

IS 269:2015.

b) Coarse aggregates (10 mm and 20 mm):

coarse aggregates (10 mm and 20 mm)

samples were evaluated for various proper-

ties as per IS: 2386-1963. The test results

(Tables 7 and 8) of coarse aggregate sam-

ples (CA 10 mm and 20 mm) showed con-

formance to the requirements of IS: 383-

2016.

c) Fine aggregate (natural & crushed): fine

aggregate (natural & crushed) samples

were evaluated for several properties as

per IS: 2386-1963. Test results (Table 9

and 10) showed that the fine aggregate

(natural & crushed) samples meet the vari-

ous physical requirements of IS: 383-2016.

d) Chemical admixtures: chemical admixtures

BASF Rheobuild 1100 (Naptha based) and

BASF Master Glenium Sky 8777 (PCE

based) sample met the various physical re-

quirements of IS: 9103-1999.

Fig. 4 – SEM image of BY- bottom ash (as such)

Fig. 5 – SEM image of BY2- bottom ash (fraction

between 4.75 mm and 75 µm)


56

Table 5 – Phases and their relative abundance in BY and BY2 samples

Phase Chemical Formula Relative Abundance

BY BY-2

Quartz SiO2 Predominant Predominant

Mullite Al6Si2O13 Major Major

Tridymite SiO2 Major Major

Hematite Fe2O3 Minor Minor

Table 6 – Test results of cement sample (OPC-43 Grade)

Sl. No. Properties Test results Limits as per IS 269:2015

(A) Physical analysis:

1 Blain’s fineness, m2/kg 309 225.0 (min.)

2 Setting time, minutes

Initial & final

155 & 215

30.0 (min.) & 600.0 (max.)

3 Compressive strength, N/mm2

3 days

7 days

28 days

32

43

54.5

23.0 (min.)

33.0 (min.)

43.0 (min.)

4 Soundness

Autoclave, %

Le Chatelier Exp. (mm)

0.06

2.0

0.8 (max.)

10.0 (max.)

(B) Chemical analysis:

1 Loss of ignition (LOI) % by mass 2.91 5.0 (max.)

2 Silica (SiO2) % by mass 20.00 --

3 Iron oxide (Fe2O3) % by mass 4.08 --

4 Aluminum oxide (Al2O3) % by mass 4.81 --

5 Calcium oxide (CaO) % by mass 60.15 --

6 Magnesium oxide (MgO) % by mass 4.50 6.0 (max.)

7 Sulphate (SO3) % by mass 1.89 3.5 (max.)

8 Alkalies: Na2O & K2O % 0.45 & 0.55 --

9 Chloride content % by mass 0.028 0.1

10 Insoluble residue % by mass 1.76 5.0

Table 7 – Physical test results of coarse aggregates (10 mm & 20 mm) samples

Sl. No. Test Carried out Result Obtained Permissible Limits as Per

IS: 383-2016 CA 10 mm CA 20 mm

1 Specific gravity 2.73 2.75 --

2 Water absorption (%) 0.3 0.3 --

3 Crushing value % 27 26 30 (For wearing surface)

4 Impact value % 20 19 30 (For wearing surface)

5 Flakiness index % 4.8 6.2 (40%) Combined limit for flaki-

ness and elongation index 6 Elongation index % 10.5 19.2

8 Deleterious materials %

(except coal & lignite)

0.15

0.2

2

4 Evaluation of fresh, hardened and

durability properties of concrete

Preparation, casting and testing of concrete

mixes was carried out as per relevant

IS/ASTM/DIN/ISO methods to determine the engi-

neering properties of various concrete mixes. Fresh

concrete properties such as slump, air content, wet

density and initial & final setting time of concrete (as

per IS: 1199) along with compressive strength (as

per IS: 516) at 3, 7, 28, and 56 days and cylindrical

compressive strength as per ASTM C39 at 28 days

were evaluated for all the 32 mixes. Hardened con-

crete properties such as flexural strength (as per IS:


57

Table 8 – Sieve analysis of coarse aggregates (10 mm & 20 mm)

IS sieve size

(mm)

CA 10 mm CA 20 mm

% passing As per Table 7 (Clause

6.1 & 6.2) IS 383-2016

% passing As per Table 7 (Clause 6.1 &

6.2) IS 383-2016

40 100 100 100 100

20 100 100 95 85-100

12.5 100 100 - -

10 86 85-100 2 0-20

4.75 5 0-20 0 0-5

2.36 0 0-5 0 -

Table 9 – Physical test results of fine aggregate (natural & crushed) sample

Sl

No.

Test Carried out

Result obtained Permissible Limits as Per

IS: 383-2016

Natural Crushed Natural Crushed

1 Specific gravity 2.64 2.73 -- --

2 Water absorption, % 0.4 0.6 -- --

3 Material finer than 75-micron, IS Sieve % 0.2 5.9 3 15

Table 10 – Sieve analysis of fine aggregate (crushed) sample

Sieve Size Percentage passing Percentage passing for Grading Zone III as per IS

383:2016 Table 9 Natural Crushed

10 mm 100 100 100

4.75 mm 100 100 90-100

2.36 mm 100 90 85-100

1.18 mm 97 78 75-100

600 micron 74 62 60-79

300 micron 25 38 12-40

150 micron 5 19 0-10 (but for crushed stone sands, the permissible

limit on 150 micro IS Sieve is increased to 20 %)

Zone as per IS: 383-2016 Zone III

516), static modulus of elasticity (as per IS: 516)

along with drying shrinkage (as per IS: 1199) and

moisture movement (as per IS: 1199) were evaluated

at the age of 28 days for 5 selected experimental

mixes and 2 control mixes. Evaluation of durability

properties of concrete such as pH value, water per-

meability (as per DIN 1048), volume of permeable

voids, water absorption (as per ASTM C 1585), rapid

chloride penetration test (as per ASTM C 1202),

electrical resistivity using four-point Wenner probe,

air permeability and accelerated carbonation test (as

per ISO 1920 Part 12) were carried out for 5 selected

experimental mixes and 2 control mixes.

The concrete mix trials have been carried out

using OPC-43 grade cement, natural fine aggregate

(sand), crushed fine aggregate (sand) at w/c 0.4 and

0.65 and using bottom ash samples in different pro-

portions. There are four control mixes i.e,, M1, M6,

M11 and M15 having 100% Fine Aggregate

(Crushed/Natural) with w/c 0.4 and 0.65. Study was

conducted on two fractions of bottom ash BY (As

such) and BY2 (between 4.75 mm and 75 µm). 14

Nos. out of 28 Nos. concrete mix trials conducted on

bottom ash (As such) and 14 no’s conducted on bot-

tom ash (between 4.75 mm and 75µm). In concrete

mixes, fine aggregate is being replaced with each

fraction of bottom ash by 25%, 50%, 75%, and 100%

respectively. The concrete mixes were designed for

the workability range of 90 – 120 mm slump. Mix-

proportions and test results of 32 concrete mixes are

given in Tables 11 and 12, respectively.

4.1 Evaluation of fresh concrete properties and

compressive strength of hardened concrete

32 mixes were prepared and analyzed for differ-

ent fresh properties of concrete along with compres-

sive strength at different ages. Mix proportions and

fresh concrete properties along with compressive

strength results for all the mixes are given in Tables

11 and 12, respectively. Comparison of 28-day com-

pressive strength for all the mixes has been shown in

figure 6. Observations related to experimental mixes


58

Table 11 – Mix design details of all the mixes Mix Bottom ash

(% replacement

of fine aggre-

gate)

Dosage of

Admixture

(% by weight of ce-

ment)

Cement

(kg/m3)

Water

(kg/m3)

Fine aggregate

(kg/m3)

(Type)

Bottom

Ash

(kg/m3)

Coarse

Aggregate

(kg/m3)

10 mm 20 mm

M1 0 0.2 (PCE based) 425 170 739 (crushed) 0 464 696

M2 25 (BY) 0.3 (PCE based) 425 170 554 (crushed) 139 464 696



M5 100 (BY) 1.0 (PCE based) 425 170 0 503 492 739

M6 0 0 300 195 814 (crushed) 0 451 677

M7 25 (BY) 0.5 (Naptha based) 300 195 609 (crushed) 155 450 670

M8 50 (BY) 1.2 (Naptha based) 300 195 405 (crushed) 309 449 674


M10 100 (BY) 1.0 (PCE based) 300 195 0 616 448 673

M11 0 0.2 (PCE based) 425 170 697 (Natural) 0 472 708

M12 25 (BY) 0.3 (PCE based) 425 170 508 (Natural) 134 479 719



M15 0 0.5 (Naptha based) 300 195 748 (Natural) 0 466 699

M16 25 (BY) 1.0 (Naptha based) 300 195 560 (Natural) 147 465 698

M17 50 (BY) 2.0 (Naptha based) 300 195 372 (Natural) 293 463 695


M19 25 (BY-2) 0.2 (PCE based) 425 170 554 (crushed) 139 464 696



M22 100 (BY-2) 1.0 (PCE based) 425 170 0 498 492 739

M23 25 (BY-2) 0 300 195 610 (crushed) 154 451 677

M24 50 (BY-2) 1.0 (Naptha based) 300 195 396 (crushed) 299 457 686

M25 75 (BY-2) 2.0 (Naptha based) 300 195 192 (crushed) 436 464 695

M26 100 (BY-2) 1.0 (Naptha based) 300 195 0 554 481 721

M27 25 (BY-2) 0.3 (PCE based) 425 170 522 (Natural) 136 472 707



M30 25 (BY-2) 0.6 (Naptha based) 300 195 561 (Natural) 146 466 699

M31 50 (BY-2) 1.5 (Naptha based) 300 195 354 (Natural) 276 480 720


Fig. 6 – Comparison of 28-day compressive strength results for all the mixes


59

Table 12 – Fresh concrete properties and compressive strength of hardened concrete of all the mixes Mix Slump

(mm)

Air

content

(%)

Wet

Den-

sity

(kg/m3)

Setting time

(minutes)

Cube Compressive Strength

(MPa)

Cylindrical

Compressive

Strength

(MPa)

as per ASTM

C39

Standard De-

viation for

compressive

strength at 28

days age for

six specimen

for each mix Ini-

tial

Fi-

nal

3

Day

7

Day

28

Day

56

Day

28

Day

M1 100 0.8 2542 380 560 32.65 39.52 40.99 42.20 33.59 1.85

M2 100 1.2 2530 440 550 38.56 37.77 44.36 47.99 35.77 2.04

M3 120 0.8 2489 470 610 46.05 46.35 48.66 54.88 39.24 1.84

M4 110 1.3 2425 510 690 35.15 42.35 57.71 60.81 48.49 1.65

M5 Zero Properties were not evaluated

M6 110 1.9 2498 420 610 16.94 19.65 27.49 27.66 21.65 1.88

M7 110 1.3 2525 450 580 20.58 22.55 30.11 33.42 24.09 1.79

M8 110 1.4 2471 480 640 22.79 25.47 33.47 37.79 26.78 1.68

M9 100 1.3 2423 520 660 26.30 28.97 40.99 46.70 34.16 1.54


M11 100 1.9 2447 420 620 39.99 40.36 45.76 46.64 37.20 1.89

M12 110 1.8 2495 450 640 39.30 47.24 51.80 52.78 43.53 1.99

M13 100 1.5 2477 470 670 41.82 45.30 55.19 57.01 46.77 2.01

M14 90 1.3 2427 480 710 38.60 41.71 58.49 65.96 49.56 1.84

M15 100 1.7 2435 510 750 15.69 21.18 25.31 27.25 19.93 1.69

M16 110 1.5 2427 540 770 15.94 21.94 25.91 28.56 21.06 1.74

M17 90 1.2 2394 580 800 15.91 17.75 24.38 25.93 19.50 2.01

M18 120 1.0 2403 620 830 24.58 27.07 35.79 38.97 29.09 1.99

M19 100 1.4 2587 450 560 36.25 40.58 44.13 47.74 36.78 2.06

M20 110 1.3 2409 480 640 43.20 48.36 52.59 55.05 44.19 1.85



M23 100 1.2 2447 460 565 16.31 21.88 26.11 29.80 20.39 1.96

M24 110 1.0 2409 490 610 20.64 24.59 28.83 32.94 22.70 1.85

M25 120 0.8 2370 530 640 18.35 23.83 31.30 31.99 25.04 1.79


M27 100 1.4 2519 470 660 42.32 44.96 48.14 50.49 39.13 1.84

M28 120 1.1 2489 490 690 44.35 50.45 53.54 56.96 45.37 2.01

M29 110 1.2 2462 500 710 43.10 49.19 60.62 63.49 50.94 2.34

M30 100 1.7 2439 560 790 23.26 26.36 32.35 35.22 26.09 1.89

M31 110 1.4 2409 600 830 20.67 22.44 27.77 30.49 21.87 2.22

M32 120 1.0 2400 630 850 27.60 30.03 34.66 37.11 27.08 1.67

(containing bottom ash) of similar type along with

their corresponding control mixes has been dis-

cussed individually under sections 4.1.1 to 4.1.8.

4.1.1 Mix using bottom ash BY (As such), w/c 0.4

and crushed fine sand

Concrete trial mixes were carried out at w/c ra-

tio of 0.4 using crushed fine aggregate and Bottom

Ash BY (as such) sample. The crushed fine aggre-

gate has been replaced in four concrete mixes (i.e.,,

M2, M3, M4, and M5) by 25%, 50%, 75%, and

100%, respectively, with bottom ash BY. Mix pro-

portions and fresh concrete properties along with

compressive strength results for all the mixes are

given in Tables 11 and 12, respectively, where M1 is

control mix using crushed fine aggregate at w/c 0.4.

The study indicates that with increase in percentage

of bottom ash in the concrete mix to maintain given

workability in the range of 90 – 120 mm, the chemi-

cal admixture dosage increases. When the crushed

fine aggregate is replaced by 100%, the workability

could not be achieved despite using 1% PC based ad-

mixture/super-plasticizer (M5) and therefore its con-

crete properties were not evaluated further. Wet

Density results for M2 (25% replacement), M3 (50%

replacement) and M4 (75% replacement) are compa-

rable with that of control mix (M1). Compressive

strength results for M2 (25% replacement), M3 (50%

replacement) and M4 (75% replacement) are higher

than that of M1 (control mix) at 3, 7, 28, and 56 days.


60

4.1.2 Mix using bottom ash BY (As such), w/c

0.65 and crushed fine sand

Concrete trial mixes were carried out at w/c of

0.65 using crushed fine aggregate and bottom ash

BY (as such) sample. The crushed fine aggregate has

been replaced in four concrete mixes (i.e.,, M7, M8,

M9, and M10) by 25%, 50%, 75%, and 100%, re-

spectively, with bottom ash BY. Mix proportions and



11 and 12, respectively, where M6 is control mix us-

ing crushed fine aggregate at w/c 0.65. The study in-

dicates that with increase in percentage of bottom

ash in the concrete mix to maintain given workability

in the range of 90 – 120 mm, the chemical admixture

dosage increases. When the crushed fine aggregate is

replaced by 100%, the workability could not be

achieved despite using 1% PC based admixture/su-

per-plasticizer (M10) and therefore its concrete

properties were not evaluated further. Wet Density

results for M7 (25% replacement), M8 (50% replace-

ment) and M9 (75% replacement) are comparable

with that of control mix (M6). Compressive strength

results for M7 (25% replacement), M8 (50% replace-

ment) and M9 (75% replacement) are higher than

that of M6 (control mix) at 3, 7, 28, and 56 days.


0.4 and natural fine sand


tio of 0.40 using natural fine aggregate and bottom

ash BY (As such) sample. The natural fine aggregate

has been replaced in four concrete mixes (i.e.,, M12,

M13, M14, and M5) by 25%, 50%, 75%, and 100%,

respectively, with bottom ash. Mix proportions and



11 and 12, respectively, where M11 is control mix

using natural fine aggregate at w/c 0.40. The study

indicates that with increase in percentage of bottom



dosage increases. When the natural fine aggregate is



per-plasticizer (M5) and therefore its concrete prop-

erties were not evaluated further. Wet density results

for M12 (25% replacement), M13 (50% replacement)

and M14 (75% replacement) are comparable with

that of control mix (M11). Compressive strength re-

sults for M12 (25% replacement), M13 (50% re-

placement) and M14 (75% replacement) are higher

than that of M11 (control mix) at 3, 7, 28, and 56

days.


0.65 and natural fine sand

Concrete trial mixes were carried out at w/c of

0.65 using natural fine aggregate and bottom ash BY

(As such) sample. The natural fine aggregate has

been replaced in four concrete mixes (i.e.,, M16,

M17, M18, and M10) by 25%, 50%, 75%, and 100%,

respectively, with bottom ash. Mix proportions and



11 and 12, respectively, where M15 is control mix

using natural fine aggregate at w/c 0.65. The study

indicates that with increase in percentage of bottom



dosage increases. When the natural fine aggregate is





results for M16 (25% replacement), M17 (50% re-

placement) and M18 (75% replacement) are compa-

rable with that of control mix (M15). Compressive

strength results for M16 (25% replacement), M17

(50% replacement) and M18 (75% replacement) are

higher than that of M15 (control mix) at 3,7,28 & 56

days.

4.1.5 Mix using bottom ash BY-2 (fraction be-

tween 4.75 mm and 75 µm), w/c 0.40 and

crushed fine sand



Ash BY-2 (fraction between 4.75mm & 75µm) sam-

ple. The crushed fine aggregate has been replaced in

four mixes (i.e., M19, M20, M21 & M22) by 25%,

50%, 75% & 100% respectively with Bottom Ash.

Mix proportions and fresh concrete properties along

with compressive strength results for all the mixes

are given in Tables 11 and 12, respectively, where

M1 is Control Mix using Crushed Fine Aggregate at

w/c of 0.40. The study indicates that with increase in

percentage of Bottom Ash in the concrete mix to

maintain given workability in the range of 90-

120mm, the chemical admixture dosage increases.

When the Crushed Fine Aggregate is replaced by 75%

& 100%, the workability could not be achieved de-

spite using 1% PC based admixture/super-plasticizer

(M21 & M22) and therefore their concrete properties

were not evaluated further. Wet Density results for

M19 (25% replacement) and M20 (50% replacement)

are comparable with that of control mix (M1). Com-

pressive strength results for M19 (25% replacement)

and M20 (50% replacement) are higher than that of

M1 (control mix) at 3, 7, 28, and 56 days.


61


tween 4.75mm & 75µm), w/c=0.65 and

crushed fine sand




ple. The crushed fine aggregate has been replaced in

four concrete mixes (i.e., M23, M24, M25, and M26)

by 25%, 50%, 75% & 100% respectively with Bot-

tom Ash. Mix proportions and fresh concrete prop-

erties along with compressive strength results for all

the mixes are given in Tables 11 and 12 respectively,

where M6 is Control Mix using Crushed Fine Aggre-

gate at W/C ratio 0.65. The study indicates that with

increase in percentage of Bottom Ash in the concrete

mix to maintain given workability in the range of 90-


When the Crushed Fine Aggregate is replaced by

100%, the workability could not be achieved despite

using 1% PC based admixture/super-plasticizer

(M26) and therefore its concrete properties were not

evaluated further. Wet Density results for M23 (25%

replacement), M24 (50% replacement) and M25 (75%

replacement) are comparable with that of control mix

(M6). Compressive strength results for M23 (25%


replacement) are higher than that of M6 (control mix)

at 3, 7, 28, and 56 days.



natural fine sand


tio of 0.40 using natural fine aggregate and Bottom


ple. The natural fine aggregate has been replaced in

four concrete mixes (i.e., M27, M28, M29 & M22)




the mixes are given in Tables 11 and 12, respectively,

where M11 is Control Mix using Natural Fine Ag-

gregate at W/C ratio 0.40. The study indicates that

with increase in percentage of Bottom Ash in the

concrete mix to maintain given workability in the

range of 90-120mm, the chemical admixture dosage

increases. When the Natural Fine Aggregate is re-

placed by 100%, the workability could not be




results for M27 (25% replacement), M28 (50% re-

placement) and M29 (75% replacement) are compa-

rable with that of control mix M11. Compressive

strength results for M27 (25% replacement), M28

(50% replacement) and M29 (75% replacement) are

higher than that of M11 (control mix) at 3, 7, 28, and

56 days.



natural fine sand


tio of 0.65 using natural fine aggregate and Bottom

Ash BY-2 (fraction between 4.75 mm and 75 µm)

sample. The natural fine aggregate has been replaced

in four concrete mixes (i.e., M30, M31, M32 & M26)




the mixes are given in Tables 11 and 12 respectively,

where M15 is control mix using natural fine aggre-

gate at w/c 0.65. The study indicates that with in-

crease in percentage of bottom ash in the concrete

mix to maintain given workability in the range of 90-


When the Natural Fine Aggregate is replaced by

100%, the workability could not be achieved despite

using 1% PC based admixture/super-plasticizer

(M26) and therefore its concrete properties were not

evaluated further. Wet Density results for M30 (25%


replacement) are comparable with that of control mix

M15. Compressive strength results for M30 (25% re-

placement), M31 (50% replacement) and M32 (75%

replacement) are higher than that of M15 (control

mix) at 3, 7, 28, and 56 days.

4.2 Evaluation of hardened concrete and dura-

bility properties in selected mixes

On analysis of fresh concrete properties (work-

ability, air content and wet density) and compressive

strength results for all the concrete mixes, it was ob-

served that the mixes incorporating crushed sand

showed less strength as compared to mixes having

natural sand. Also, the admixture dosage require-

ment is higher in case of crushed sand mixes than

that of mixes with natural sand.

Studies on other hardened concrete properties

and durability studies of concrete were carried out on

selected mixes. Five mixes (M3, M4, M8, M9 &

M24) incorporating bottom ash replacing crushed

sand and two control mixes (M1 & M6 with crushed

sand at w/c 0.4 and 0.65, respectively) were selected

for the same. Comparison of flexural strength, MOE,

RCPT and accelerated carbonation test results has

been shown in Figures 7 to 10.


62

Fig. 7 – Comparison of 28-day flexural strength re-

sults for selected mixes

Fig. 8 – Comparison of 28-day modulus of elasticity

results for selected mixes

Fig. 9 – Comparison of 28-day RCPT results for se-

lected mixes

Fig. 10 – Comparison of carbonation depth results

for selected mixes

Table 13 – Hardened and durability properties of selected concrete mixes (Part-I)

Mix

Bottom Ash

(% replace-

ment of fine

aggregate

Test results at the age of 28 days

Compressive

strength

(MPa)

Flexural

strength

(MPa)

Modulus

of elasticity

(MPa)

Drying

shrinkage

(%)

Moisture

move-

ment

(%)

pH

value

Water

permeability

(mm)

M1 0 40.99 5.80 35055 0.0159 0.0177 12.83 8.33

M3 50 (BY) 48.66 6.13 39468 0.0150 0.0165 12.61 6.00

M4 75 (BY) 57.71 6.94 38512 0.0162 0.0175 12.75 3.33

M6 0 27.49 3.20 33322 0.0169 0.0186 12.63 26.0

M8 50 (BY) 33.47 3.85 30100 0.0165 0.0179 12.35 18.0

M9 75 (BY) 40.99 4.62 29636 0.0158 0.0174 12.40 13.0

M24 50 (BY-2) 28.83 3.42 32128 0.0160 0.0175 12.45 24.5

Table 14 – Hardened and durability properties of selected concrete mixes (Part-II)

Mix

Bottom Ash

(% replace-

ment of fine

aggregate)

Test results at the age of 28 days

Acceler-

ated car-

bonation

test (mm)

RCPT

(Cou-

lumbs)

Volume

of per-

meable

voids

(%)

Sorptivity index (%)

Electrical

resistivity

kOhm.cm

Air perme-

ability

( 10-16

m2)

Initial

mm/sqrt

(sec)

Secondary

mm/sqrt

(sec)

M1 0 2973 6.88 0.004 0.0010 12.8 0.044 3.9

M3 50 (BY) 1754 6.25 0.003 0.0005 17.2 0.030 3.5

M4 75 (BY) 806.7 5.84 0.002 0.0003 25.0 0.018 3.4

M6 0 3531 9.61 0.004 0.0010 21.2 0.090 9.6

M8 50 (BY) 1796 9.82 0.006 0.0015 35.2 0.065 10.2

M9 75 (BY) 859 8.19 0.002 0.0008 41.9 0.054 9.8

M24 50 (BY-2) 2291 8.99 0.004 0.0006 25.2 0.080 9.7


63

4.2.1 Mix using bottom ash BY (as such), w/c



tio of 0.40 using Crushed Fine Aggregate and Bot-

tom Ash BY (As such) sample. The crushed fine ag-

gregate was replaced by 0%, 50% & 75% of Bottom

Ash (As such) sample (M1, M3 & M4). The results

of hardened and durability properties of concrete are

given in Tables 13 and 14, where M1 is control mix

using crushed fine aggregate at w/c 0.40. The study

indicates that the test results of compressive strength,

flexural strength and static modulus of elasticity are

either comparable or higher at 50% & 75% replace-

ment of crushed sand by bottom ash in comparison

to control mix (M1). Drying shrinkage and moisture

movement test results of mixes (M3 and M4) at 50%

and 75% replacement of crushed sand by bottom ash

are comparable to that of control mix (M1). Ph value

for mixes (M3 and M4) at 50% and 75% replacement

of crushed sand by bottom ash is comparable with

that of control mix (M1). Water permeability, vol-

ume of permeable voids, air permeability, RCPT and

initial and secondary sorptivity values are lower in

case of mixes (M3 and M4) at 50% and 75% replace-

ment of crushed sand by bottom ash with that of con-

trol mix (M1). Electrical resistivity is higher in case

of mixes (M3 and M4) at 50% and 75% replacement

of crushed sand by bottom ash with that of control

mix (M1). Carbonation depth results measured by

accelerated carbonation test are lower in case of

mixes (M3 and M4) at 50% and 75% replacement of

crushed sand by bottom ash with that of control mix

(M1).

4.2.2 Mix using bottom ash BY (as such), w/c




Ash BY (As such) sample. The crushed fine aggre-

gate was replaced by 0%, 50%, and 75% of Bottom

Ash BY (As such) sample (M6, M8 & M9). The re-

sults of hardened and durability properties of con-

crete are given in Tables 13 and 14, where M6 is con-

trol mix using crushed fine aggregate at w/c ratio of

0.65. The study indicates that the test results of com-

pressive strength, flexural strength and static modu-

lus of elasticity are either comparable or higher for


crushed sand by bottom ash in comparison to control

mix (M6). Drying shrinkage and moisture movement

test results of mixes (M8 and M9) at 50% and 75%

replacement of crushed sand by bottom ash are com-

parable to that of control mix (M6). Ph value for


crushed sand by bottom ash is comparable with that

of control mix (M6). Water permeability, volume of

permeable voids, air permeability, RCPT and initial

and secondary sorptivity values are lower in case of



(M6). Electrical resistivity is higher in case of mixes

(M8 and M9) at 50% and 75% replacement of


(M6). Carbonation depth results measured by accel-

erated carbonation test are lower in case of mixes at

(M8 and M9) 50% and 75% replacement of crushed

sand by bottom ash with that of control mix (M6).

4.2.3 Mix using Bottom Ash BY-2 (fraction be-


crushed fine sand

Concrete trial mix were carried out at w/c of

0.65 using crushed fine aggregate and Bottom Ash

BY-2 (fraction between 4.75 mm and 75 µm) sample.

The crushed fine aggregate was replaced by 0% and

50% of Bottom Ash BY-2 (fraction between 4.7

5mm and 75 µm) sample (M6 and M24). The results

of hardened and durability properties of concrete are

given in Tables 13 and 14, where M6 is control mix

using crushed fine aggregate at w/c of 0.65. The

study indicates that the test results of compressive

strength, flexural strength and static modulus of elas-

ticity are higher for mix (M24) at 50% replacement

of crushed sand by bottom ash in comparison to con-

trol mix (M6). Drying shrinkage and moisture move-

ment test results of mix (M24) at 50% replacement

of crushed sand by bottom ash are comparable to that

of control mix (M6). Ph value for mix (M24) at 50%

replacement of crushed sand by bottom ash is com-

parable with that of control mix (M6). Water perme-

ability, volume of permeable voids, air permeability,

RCPT and initial and secondary sorptivity values are

either lower or comparable in case of mix (M24) at

50% replacement of crushed sand by bottom ash

with that of control mix (M6). Electrical resistivity is

higher in case of mix (M24) at 50% replacement of


(M6). Carbonation depth results measured by accel-

erated carbonation test are lower in case of mix (M24)

at 50% replacement of crushed sand by bottom ash

with that of control mix (M6).

5 Conclusions

Based on the test results and discussion of char-

acterization of bottom ash, fresh concrete, hardened

concrete and durability properties of concrete mixes

following are the conclusions:

(1) With addition of Bottom Ash, there is increase

in admixture dosage in concrete mixes for main-

taining the same workability as compared to


64

control mixes. However, at more than 50% re-

placement of Bottom Ash, it does not give a

workable mix even at the higher admixture dos-

age than the permissible limits.

(2) Strength of the mixes made with Bottom Ash

‘as such’ is more than the mixes made with Bot-

tom Ash ‘fraction between 4.75 mm and 75µm’.

However, for both the fractions i.e., ‘as such’

and ‘fraction between 4.75 mm and 75µm’ of

Bottom Ash, their behavior in fresh, hardened

and durability properties of concrete is compa-

rable. Presence of material finer than 75µm is

beneficial in concrete mixes in terms of pore re-

finement and pozzolanic reactivity, which re-

sults in development of higher compressive

strength of concrete as compared to control mix.

Therefore, use of “as such” fraction of bottom

ash as replacement to fine aggregate in concrete

is technically logical.

(3) Air content in all fresh concrete mixes is less

than 2%. In all the mixes, with replacement of

fine aggregate with Bottom Ash, the wet density

of fresh concrete decreases as the specific grav-

ity of Bottom Ash is less than that of fine aggre-

gate. In all the mixes, with replacement of fine

aggregate with Bottom Ash, setting time got

marginally delayed. However, as seen in the

hardened concrete results, it does not affect the

strength development.

(4) With addition of Bottom Ash in the mixes made

using bottom ash as a replacement to natural and

crushed sand, strength parameters such as com-

pressive strength and flexural strength increases

for both the fractions as compared with that of

control mixes. Static Modulus of Elasticity, dry-

ing shrinkage, moisture movement, pH of con-

crete and accelerated carbonation test results are

comparable for both the fractions of Bottom

Ash as compared with control mix.

(5) Water permeability, Volume of permeable

voids, RCPT and Air permeability are lower for

both the sources and both the fractions of Bot-

tom Ash as compared with control mix due to

pore refinement in hardened concrete. Electrical

Resistivity is higher for both the sources and

both the fractions of Bottom Ash as compared

with control mix which shows higher resistance

to corrosion.

(6) Since the replacement of fine aggregate with

Bottom Ash is more than replacement of ce-

mentitious material with fly ash in concrete, the

total quantity of alkali may be higher. Therefore,

a study needs to be conducted to verify the po-

tential alkali-aggregate reaction in such a con-

crete system. Presence of higher alkali may af-

fect the setting time of concrete. It is observed

that in all the mixes, with replacement of fine

aggregate with Bottom Ash, setting time of con-

crete gets marginally delayed. However, as seen

in the hardened concrete results, it does not af-

fect the strength development.

References

1. Central Electricity Authority, http://www.cea.

nic.in/

2. Abubakar, A. U.; and Baharudin, K. S. (2012)

“Properties of concrete using Tanjung bin power

plant coal bottom ash and fly ash,” International

Journal of Sustainable Construction Engineering

& Technology, vol. 3, no. 2, pp.56-69.

3. Kasemchaisiri, R.; and Tangtermsirikul, S. (2008)

“Properties of self compacting concrete incorpo-

rating bottom ash as a partial replacement of fine

aggregate”, Science Asia Research Article, vol. 34

pp. 087-095.

4. Rafieizonooz, M.; Mirza, J.; Salim, M. R.; Hussin,

M. W.; and Khankhaje, E. (2016) “Investigation of

coal bottom ash and fly ash in concrete as replace-

ment for sand and cement,” Construction and


5. Kim, H. K.; Jang, J. G.; Choi, Y. C. and Lee, H. K.

(2014) “Improved chloride resistance of high

strength concrete amended with coal bottom ash

for internal curing,” Construction and Building


6. Singh, M.; and Siddique, R. (2014) “Strength

Properties and microstructural properties of con-

crete containing coal bottom ash as partial replace-

ment of sand,” Construction and Building Materi-

als, vol. 68, pp. 39-48.

7. Andrade, L. B.; Rocha, J. C. and Cheriaf, M. (2009)

“Influence of coal bottom ash as fine aggregate on

fresh properties of concrete,” Construction and

Building Materials, vol. 23, no. 2, pp. 609-614.

8. Aggarwal, P.; Aggarwal, Y.; and Gupta, S. M.

(2007) “Effect of Bottom Ash as replacement of

fine aggregate in concrete,” Asian Journal of Civil

Engineering (Building and Housing), vol. 8, no. 1,

pp. 49-62.

9. Kadam, M. P.; and Patil, Y. D. (2013) “Effect of

coal bottom ash as sand replacement on the prop-

erties of concrete on different w/c ratio,” Interna-

tional Journal of Advanced Technology in Civil

Engineering, vol. 2, no. 1, pp. 45-50.

10. Arumugam, K.; Ilangovan, R.; and Manohar, J. D.

(2011) “A study on characterization and use of

pond ash as fine aggregate in concrete,” Interna-

tional Journal of Civil & Structural Engineering,

vol. 2, no. 2., pp. 466-474.


65

11. Raju R.; Paul M. M.; and Abubaker K. A. (2014)

“Strength Performance of concrete using Bottom

Ash as fine aggregate,” International Journal of

Research in Engineering and Technology, vol. 2,

no. 9, pp. 111-122.

Mechanical characterization of a FRCM system with aramid ...

Documents