Construction and Building Materials 155 (2017) 371–380

Contents lists available at ScienceDirect

Construction and Building Materials

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

An investigation into fracture behavior of geopolymer concrete with

digital image correlation technique
Z.L. Xie a, H.F. Zhou a,⇑, L.J. Lu a, Z.A. Chen b
a
College of Civil Engineering and Architecture, Wenzhou University, Chashan University Town, Wenzhou 325035, China
b
College of Civil Engineering and Architecture, Zhejiang University of Technology, No. 18, Chaowang Street, Xiacheng District, Hangzhou 310014, China

h i g h l i g h t s

DIC was proposed to monitor the crack tip field of fly ash based GPC.

CTOD and crack extension length were achieved with the help of DIC.

Crack growth behavior of fly ash based GPC was elaborated.

Crack propagation can be evidently manifested in COD and crack extension length.

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

Article history: Fly ash based geopolymer concrete (GPC) is gaining significant interest as a greener alternative to ordi-
Received 23 January 2017 nary Portland cement concrete over the past decades. Its fracture behavior, especially the crack growth
Received in revised form 23 July 2017 behavior, has received few attentions, possibly hindered by the difficulties in the monitoring of crack
Accepted 8 August 2017
tip field with conventional measurement technique. Recently, digital image correction (DIC) has opened
Available online 23 September 2017
up the opportunity to gain a deep insight into its fracture behavior. Therefore, an investigation into the
fracture behavior of fly ash based GPC with the use of DIC was performed. First, three-point bending tests
Keywords:
on notched beam specimens were conducted. The displacement field of beam specimen was measured by
Geopolymer concrete
Fracture
DIC. Then, crack opening displacement (COD), crack extension length and mid-span deflection were
Digital image correlation achieved and the crack growth in fly ash based GPC were elaborated. Finally, the fracture energy and frac-
Crack opening displacement ture toughness were calculated. The results show that DIC can be a useful technique for the monitoring of
Crack extension length crack tip field. The crack propagation in fly ash based GPC can be evidently manifested in COD and crack
extension length. The normalized crack extension length may be employed as a fracture criterion to pro-
nounce the onset of unstable crack propagation in fly ash based GPC.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction thermore, they are added with the advantage of environmental

Geopolymers, sometimes termed as ‘‘inorganic polymers”, are
cementitious materials synthesized by combining source materials
which are rich in silica and alumina with strong alkali solutions
and soluble silicates (in most cases) where the dissolved Al2O3
and SiO2 species undergo geopolymerization to form a 3D amor-
phous aluminosilicate network [1]. Recently, they are gaining sig-
nificant interest as a greener alternative to ordinary Portland
cement (OPC). They can provide comparable or superior perfor-
mance to OPC, such as high compressive strength, low shrinkage,
acid resistance, fire resistance and low thermal conductivity, satis-
fying most of requirements for construction materials [2–4]. Fur-
⇑ Corresponding author.
E-mail address: mailtofei@wzu.edu.cn (H.F. Zhou).
benefit as they completely move away from OPC and thereby the
high CO2 emission associated with its production. It was reported
that the use of geopolymers could bring down the overall CO2
emission by up to 60–80% in comparison with the use of OPC [5–
9]. In addition, abundant of industrial byproducts generated in var-
ious industries, which are causing problems in term of finding an
ideal solution for disposal purposes, were found to be suitable to
use as geopolymer source materials. Therefore, the use of geopoly-
mers offers the greatest potential in solving not only the environ-
mental degradation related to the use of OPC as primary binder
material in the construction industry, but also the waste manage-
ment problems related to the aluminosilicate solid waste materials
generated from various industries.
At present, fly ash and blast furnace slag appear to be the most
promising precursors for large-scale industrial production of

http://dx.doi.org/10.1016/j.conbuildmat.2017.08.041
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
372 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380
geopolymers [10,11]. Fly ashes are very fine residues from the
combustion of coal during the operation of thermal power plants.
There is an easy access to a suitable supply of fly ash as its annual
production worldwide is estimated around 500 million tonnes
[12–14]. In the past two decades, considerable research has been
carried out throughout the world to cover the various aspects of
fly ash based geopolymer, providing a large volume of useful data
and important findings. For example, the properties of fly ash
based geopolymer, such as compressive strength, splitting and
flexural tensile strength, chloride resistance, sulfate resistance,
acid resistance, thermal resistance, freeze-thaw resistance, and
efflorescence resistance, have been studied extensively [15–17].
A wide variety of applications has also been identified for fly ash
based geopolymer including geopolymer concrete (GPC), coating
materials, fire resistant materials, adsorption and purification,
and immobilization of toxic metals [17]. Among them, GPC has
attracted most of the attentions. So far, GPC has been recognized
as being equivalent to OPC concrete for non-structural applications
by the state roads authority in Victoria, Australia in a 2010 update
to their design specification Section 703 [18]. Researchers on the
use of geopolymers have extended to the investigation of struc-
tural applications of GPC as well [19,20]. Furthermore, a few actual
field applications of fly ash based GPC have been reported [21].
Therefore, the understanding that has been developed to date pro-
vides indications that fly ash based geopolymer does in fact have
the potential for wide scale utilization in the construction industry,
as well as in other niche applications.
Although fly ash based GPC has shown comparable or superior
performance to OPC concrete in various aspects, it is still far from
problem free. Like OPC concrete, most of fly ash based GPCs are
also quasi-brittle in nature [17,22–24]. Because of the quasi-
brittle nature, fly ash based GPCs are sensitive to cracking, which
not only imposes constraints in their applications, but also affects
their long term durability. Therefore, the fracture behavior of fly
ash based GPCs, which is related to the formation and propagation
of cracks, deserves intensive investigations to ensure their suitabil-
ity for structural applications. Nevertheless, studies on fracture
behavior of fly ash based GPC are still very limited in comparison
with other mechanical properties. Pan et al. studied the fracture
properties of low calcium fly ash based geopolymer paste and con-
crete with focuses on fracture energy and brittleness [22]. Sarker
et al. investigated more fracture characteristics of low calcium fly
ash based GPC, including fracture energy, fracture toughness, and
fracture planes [23]. Recognizing that the curing method plays a
significant role on the fracture properties of geopolymers, Nath
and Sarker developed a low calcium fly ash based GPC, which is
suitable for curing in ambient condition by adding a small percent-
age of ground granulated blast furnace slag (GGBFS), and evaluated
the fracture energy and fracture toughness of the GPC [24]. Like-
wise, the incorporation of OPC in high calcium fly ash based
geopolymer was suggested to improve the strength development
of geopolymer cured at ambient temperature, and the fracture
properties of fracture energy and fracture toughness were exam-
ined correspondingly [25]. In addition, studies on fracture proper-
ties of fiber reinforced fly ash based GPC have also been reported
[26–28]. In all those previous studies, mainly two fracture proper-
ties, i.e., fracture energy and fracture toughness, were determined
for fly ash based GPC, while the crack growth behavior of fly ash
based GPC have received much less attention. The reason possibly
lies in the fact that the monitoring of strain and displacement fields
near the crack tip through conventional contact measurement
techniques is very difficult if not impossible. For instance, there
have been a number of attempts to use different techniques such
as strain gauge, clip gauge to measure crack opening displacement
(COD), which usually require the crack position known prior. In
reality, however, this is not always the case, making them unwork-
able sometimes. Therefore, it is desired that new technique suit-
able for the monitoring of crack tip field be proposed so as to
achieve a full understanding of the facture behavior of fly ash
based GPC, especially its crack growth behavior.
Recent years have seen a great development in full-field optical
measurement methods [29,30]. Among them, digital image correc-
tion (DIC) has emerged as a practical tool in the experimental
mechanics. DIC uses the digital images of the surface of the object
before and after deformation to obtain the displacement field of
the object. A matching algorithm is employed to recognize and
track a specific subset (a group of pixels) on a series of images.
The concept behind the matching algorithm is that the distribution
of grey values in a subset of the image taken at the undeformed
state corresponds to the distribution of grey values of the same
subset on the image taken at the deformed state. By tracking the
subsets in each successive image and using photogrammetric prin-
ciples of triangulation and bundle adjustment, the 3-D position or
coordinate of each subset across the surface of the object are
obtained and thereby the resulting relative displacements and
strains can be calculated. For this technique work well, an arbitrary
speckle pattern that is randomly sprayed onto the object surface or
that is offered by the texture of the object’s material is needed. The
objective is to obtain an image with a varied and distinctive pat-
tern, in order to enable discrimination between different subsets.
DIC technique was developed in the 1980s independently by Yam-
aguchi [31] and Peters and Ranson [32]. But the technology has
only recently been exploited in industry and research, benefiting
from the availability of high resolution digital cameras as well as
the rapid developments in image correlation algorithms. The tech-
nique is currently being explored to study the fracture behavior of
some materials and can provide a wealth of additional information
that was previously unobtainable [33–37]. To the best of the
authors’ knowledge, the use of DIC for the study of fracture behav-
ior of geopolymers has not been reported yet. Recognizing the
demand for more studies on fracture behavior of fly ash based
GPC, as well as the opportunity offered by DIC to gain a deep
insight into it, this study therefore performs an investigation into
the fracture behavior of fly ash based GPC with the help of DIC.
First, three-point bending tests on notched beam specimens of
fly ash based GPC were conducted. To parenthetically examine
the effect of notch depth on the fracture behavior of fly ash based
GPC, beam specimens with same dimension but different notch
depths were tested. DIC was employed to measure the displace-
ment field of beam specimen throughout the test. Then, the crack
mouth opening displacement (CMOD), crack tip opening displace-
ment (CTOD), and crack extension length, which are not readily
available through conventional measurement techniques, were
achieved with the use of the displacement field measured by DIC.
Upon them, the crack formation and propagation in the fly ash
based GPC were elaborated. Finally, the fracture properties of fly
ash based GPC namely fracture energy and fracture toughness
were calculated and discussed.
2. Experimental program
2.1. Materials
Geopolymer binder was prepared with low calcium fly ash col-
lected from a local thermal power plant in Wenzhou, China. In
addition, GGBFS, which has been reported to be able to improve
the workability and regulate the setting of the fly ash based GPC
[17], was added as a small part of the binder. The chemical compo-
sition of fly ash and GGBFS determined by X-ray fluorescence (XRF)
analysis are shown in Table 1. The alkaline liquid used was a mix-
ture of sodium hydroxide (NaOH) and sodium silicate (Na2SiO3)
 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380 373

Table 1
Chemical composition of fly ash and GGBFS determined by XRF analysis.

SiO2 Al2O3 CaO Fe2O3 TiO2 K2O SO3 MgO Na2O LOI
Fly ash 49.10 36.70 4.96 3.67 1.39 0.943 0.49 0.37 0.20 2.08
Slag 32.28 13.80 37.85 2.74 1.30 0.96 2.90 3.73 0.70 1.30

LOI: loss of weight after ignition.

solutions, in which the sodium hydroxide solution had a concen-

tration of 18 M and the sodium silicate solution had a relative den-
sity of 1.35 and a modulus ratio (mole ratio of SiO2 to Na2O) of 3.3.
The sodium hydroxide solution was mixed with sodium silicate
solution by a mass ratio of 1:3. The use of a high concentration
of NaOH was to achieve a high compressive strength of the fly
ash based GPC. The increase in the compressive strength of fly
ash based GPC with the increasing concentration of NaOH has also
been reported in the literature [3]. The fine aggregate used was
medium sand with a fineness modulus of 2.4, and the coarse aggre-
gate was crushed stone with a maximum diameter no more than
20 mm. Both coarse and fine aggregates were in saturated-
surface-dry (SSD) condition.

2.2. Test specimens

Fig. 1. Fresh fly ash based GPC.

Beam specimens were prepared to investigate the fracture

properties of fly ash based GPC. The mixture proportions of GPC
were selected by trial and error method. In the GPC mixtures stud-
ied in this work, the content of coarse aggregate, fine aggregate,
low calcium fly ash, GGBFS, and alkaline liquid was 1,050, 565,
420, 180, and 200 kg/m3, respectively. First, the aggregates, fly
ash and GGBFS were dry mixed for 5 min. The alkaline liquid was
then added gradually to the dry mixture of aggregates and binders,
and wet mixing was carried out for another 5 min. After that, the
fresh GPC mixture was poured into moulds and compacted on a
vibration table for 1 min. Upon the completion of casting, the
moulds with finished GPC specimens were covered by polyethy-
lene sheet and placed immediately into a preheated oven at
60 °C for heat curing. After 24 h, the specimens were then
demoulded and cured in ambient condition. At the age of 28 days,
the beam specimens were cut with a single edge notch in the mid
span. As far as the notch-depth ratio is concerned, ratios ranging
from 0.2 to 0.5 have been widely used in the literature [22–
24,38,39], though a ratio of 0.5 is specified by RILEM [40]. Recog-
nizing the non-uniformity in the notch-depth ratios used in the lit-
erature, this study therefore attempted to explore the effect of
notch-depth ratio on the fracture behavior of fly ash based GPC
by means of single size variable notch test method. To this end, a
total of four notch-depth ratios varying from 0.2 to 0.5 at an incre-
ment of 0.1, which covered most of the notch-depth ratios used in
the literature, were considered in this study. For each case of
notch-depth ratio, three beam specimens were fabricated in order
to check the repeatability of test results. Meanwhile, three com-
panion cylinder specimens with a diameter of 150 mm and a
height of 300 mm were also cast for compressive strength tests.
At the age of 28 days, the compressive strength of fly ash based
GPC was 61.3, 62.8, and 62.1 MPa for the three cylinder specimens,
respectively, resulting in an average compressive strength of
62.1 MPa.
In the mean time, slump and setting time were determined to
evaluate the fresh properties of the GPC mixtures. Fig. 1 shows a
picture of the fresh GPC mixtures. Inheriting the color of fly ash,
the fresh GPC mixtures were dark in color. Furthermore, it can be
seen that the mixtures are very cohesive. Meanwhile, they exhibit
satisfactory workability as well. Slump test was conducted to mea-
sure the workability of the mixtures, resulting in a slump of
250 mm which was similar to those obtained in previous studies
[12,41]. It therefore validates that the fresh fly ash based GPC pro-
duced in this study has adequate workability. Setting time test was
performed as per ASTM C191–13 [42]. The paste was molded and
placed in a moist cabinet and allowed to setting. Periodic penetra-
tion tests were performed by allowing 1 mm Vicat needle to settle
into the paste. It turned out that the fly ash based GPC produced in
this study had the initial and final setting times of 50 min and
95 min, respectively. It has been reported that the final setting of
the fly ash based GPC usually can be achieved within 1–2 h at room
temperature [17], indicating that moderate setting time was
achieved for the fly ash based GPC produced in this study. The ade-
quate workability and moderate setting time achieved in this study
is mainly attributed to the addition of a small percentage of GGBFS
in the mixture proportions, which has been found to be useful to
improve the workability and regulate the setting of the fly ash
based GPC [17]. In addition, as mentioned earlier, both coarse
and fine aggregates were in SSD condition, which also contributed
to the low liquid to binder ratio used in this study to some extent.
2.3. Fracture test
Fracture tests were carried out after the specimens were cured
for 28 days. Three-point bending tests on single edge notched
beam specimens were conducted to determine the fracture proper-
ties of fly ash based GPC. Fig. 2 shows the schematic diagram of
three-point bending test. The beam specimens have same dimen-
sion, i.e., 515 mm in length and 100  100 mm in cross section.
P
y
ROI
d =100 mm
x
a0
b = 100 mm
Clip gauge
l = 400 mm
Fig. 2. Schematic diagram of three-point bending test.
374 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380

Their notch-depth ratios a0/d vary from 0.2 to 0.5 at an increment
of 0.1. The beam specimen is simply supported over a span of
400 mm with the notched face on the tensile side and a load P is
applied at the center of beam. The load was applied using a univer-
sal testing machine (UTM) under displacement control up to the
complete failure of beam specimen. Intended to determine the
fracture energy, the RILEM guidelines recommend that the maxi-
mum load is reached within about 30–60 s after the start of the test
[40]. Nevertheless, a slower loading rate was employed in this
study in order to enable the observation of the crack growth
behavior in the fly ash based GPC. As mentioned earlier, the fly
ash based GPC is quasi-brittle in nature, which results in a high
crack growth rate. It thereby requires a high frame rate of the
DIC in order to capture the fast crack propagation process of the
fly ash based GPC. Nevertheless, the DIC employed in this study
is capable of working reliably at a maximum frame rate of 10 Hz
only. If a loading rate satisfying the RILEM guidelines is used, the
frame rate of the DIC may be insufficient to capture the rapid crack
propagation process, especially the snap of the moment that corre-
sponds to the peak load. Therefore, a compromise between the
loading rate and the frame rate has to be achieved. After the trial
and error, a loading rate of 0.05 mm/min was finally used in this
study. At this rate, the peak load was reached in 2–4 min in this
study, which was of the same order as that specified in the RILEM
guidelines. By so doing, the effect of the loading rate on the deter-
mination of the fracture energy may also be ignored. In total, it
took 10–15 min for the complete failure of the beam specimen.
Load data were recorded throughout the entire test period by an
automatic data acquisition system. As mentioned earlier, three
beam specimens were tested for each case of notch-depth ratio.
Throughout the entire load history, DIC was employed to mon-
itor the crack growth of fly ash based GPC. Fig. 3 shows the layout
of DIC measurement system. A speckle pattern (black spray paint
on a white background) was first applied to the lateral surface of
beam specimen and a region of interest (ROI) around the notch
tip (as shown in Fig. 2) was selected as the monitoring target of
DIC. Then, a pair of digital cameras was positioned in front of the
beam specimen to capture its deformation under loading. The dis-
ital cameras. To keep the load and deformation in synchronization,
the load generated by UTM was fed into the DIC measurement sys-
tem as an external parameter so that the load corresponding to
each frame could be recorded. With the use of DIC, the 3D displace-
ments and strains of the beam specimen could be obtained, upon
which more fracture parameters could be further determined. To
validate the accuracy of DIC, a clip gauge was also employed in this
study to measure CMOD, as shown in Fig. 2. The clip gauge was
attached to the beam specimen with wedge blocks of 1.7 mm in
thickness.
3. Results and discussions
For briefness of presentation, only the test results of one of the
three beam specimens in each case of notch-depth ratio are pre-
sented unless otherwise specified.
3.1. Full-field displacements
Fig. 4 plots the horizontal displacement u obtained by DIC dur-
ing different stages of loading for the beam specimen with a notch-
depth ratio of 0.4. The origin of x-y coordinate plane is located at
the tip of precut notch. As it is seen, the displacement jump, which
is an outcome of the displacement of the two faces of crack in
opposite direction, appears and becomes increasingly evident as
the loading proceeds. Therefore, it is indicated that the horizontal
displacement field obtained by DIC can be employed to quantify
COD at any position along the crack. Among them, CMOD and
CTOD are the two most widely used displacements to characterize
the opening of crack. Fig. 5 shows the schematic diagram of crack
0.15
0.10
u (mm)

tance between beam specimen and digital camera was selected so 0.05

as to achieve a compromise between the field of view of digital 0

camera and the size of ROI. The intrinsic and extrinsic parameters -0.05
of the digital camera were calibrated with the use of a planar grid 60
pattern. After that, they worked uninterruptedly at a frame rate of 40 25
20
10 fps. As an aid, a flash light was employed to illuminate the beam 20 15
(a) 10
specimen as well as prevent motion blur in capturing fast motion. 0 5
0
-5
The flash light was positioned in front of the beam specimen facing y( -20 -10
mm -15 )
at a slightly downward angle in order to reduce glare into the dig- ) -20 mm
-40 -25 x(

0.15
0.10
u (mm)

0.05
0
-0.05
60
40 25
20
(b) 20
10
15
0 5
0
-5
y( -20 -10
mm -15 )
) -20 mm
-40 -25 x(

Fig. 4. Horizontal displacement obtained by DIC for beam specimen with a0/d = 0.4
Fig. 3. Layout of DIC measurement system. during different stages of loading.
 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380 375

specimen decreases while the CMOD that corresponds to the peak

y load increases as the notch depth increases. Except for this differ-
a
x evidence this, the normalized P-CMOD curves are also plotted in
M N
CTOD
Fig. 5. Schematic diagram of crack tip.
tip. The CMOD and CTOD are respectively the COD at the mouth
and tip of the precut notch. To compute CMOD/CTOD from the hor-
izontal displacement field obtained by DIC, a horizontal line across
the notch mouth or tip is employed as an aid, as shown in Fig. 5.
The horizontal displacements on the left and right line segments
of crack are then averaged respectively to obtain the mean hori-
zontal displacement on each side of crack. Finally, CMOD/CTOD is
obtained as the difference of the mean horizontal displacements
on the two sides of crack. As an illustration, Fig. 6 plots the P-
CMOD curves of fly ash based GPC obtained by DIC for the four
batches of beam specimens. It is seen that the peak load of beam
(a) 7
a 0 /d = 0.2
6 a 0 /d = 0.3
a 0 /d = 0.4
5 a 0 /d = 0.5
4 field measured by DIC. To compute the crack extension length, the
P (kN)
ence, the P-CMOD curves of fly ash based GPC for beam specimens
with different notch-depth ratios are of similar shape. To further
Fig. 6(b). The load and CMOD of each beam specimen were individ-
ually normalized to a range of 0–1.0. It is evident that the shapes of
the normalized P-CMOD curves of fly ash based GPC keep almost
unchanged regardless of the variation of notch depth. Making use
of the test results of beam specimens with same dimension but dif-
ferent notch depths, it is revealed that the notch depth imposes no
significant effects on the deformation behavior of fly ash based
GPC. A closer examination of the development of COD of fly ash
based GPC with loading will be presented later in terms of P-
CTOD curve. Herein, the accuracy of DIC for COD measurement is
examined by making use of the P-CMOD curve. To this end, the
CMODs measured by clip gauge are also shown in Fig. 6(a) as solid
lines. As it is seen, the CMODs obtained by DIC and those measured
using clip gage are almost overlapped with each other, demon-
strating a very good agreement between them. Therefore, the accu-
racy of DIC for COD measurement is comparable to that of clip
gauge. However, it is worth noting that uncertainties may be intro-
duced into the measurement results of DIC during the initial phase
of cracking as the magnitude of CMOD then is equivalent to or even
smaller than the resolution of DIC. In addition, the crack openings
may not be transferred on to the surface at the onset of crack,
which also increases the difficulties for DIC measurements as they
are done on the surface of beam specimen. Nevertheless, with the
fulfillment of sub-pixel resolution of DIC, the usefulness of DIC for
COD measurement does not suffer big losses. Furthermore, the
capability to measure COD at any position along the crack makes
DIC more attractive as it is very difficult to obtain them through
other conventional measurement techniques.
The crack extension length Da as shown in Fig. 5, which is
another most important parameter in fracture mechanics analysis,
can also be achieved by making use of the horizontal displacement

crack tip should be determined. Recalling that the opening of crack

3 was manifested by the jump in horizontal displacement, attempts
were therefore made to use the magnitude of horizontal displace-
2 ment jump for the location of crack tip. To have a closer examina-
tion of the horizontal displacement jump, Fig. 7 plots the
1 horizontal displacements on two vertical lines aligned symmetri-
cally to the center of precut notch (with a distance of 5 mm in
0
0 0.2 0.4 0.6 0.8 1.0
CMOD (mm)
0.03
(b) 1.1 x = 5 mm
a 0 /d = 0.2
1.0 x = -5 mm
a 0 /d = 0.3
0.9 0.02
a 0 /d = 0.4
0.8 a 0 /d = 0.5

0.7 0.01
u (mm)

0.6
P/Pmax

0.5
0
0.4

0.3
-0.01
0.2

0.1

0 -0.02
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 -40 -30 -20 -10 0 10 20 30 40 50 60
y (mm)
CMOD/CMODmax
Fig. 7. Horizontal displacements on two vertical lines aligned symmetrically to the
Fig. 6. P-CMOD curves of fly ash based GPC. center of precut notch for beam specimen with a0/d = 0.4 under peak load.
376 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380
between the vertical line and notch) for the beam specimen shown
in Fig. 4 (with a notch-depth ratio of 0.4) under peak load. The ori-
gin of y coordinate is located at the tip of precut notch. As it is seen,
the horizontal displacements are asymmetric to the centre of pre-
cut notch, implying that the crack path deviates from an upright
direction. In addition, it is difficult to ensure an exact alignment
between the action point of load and the center of precut notch.
As a result, a small deviation between them may also contribute
to the asymmetry in horizontal displacement. It is worth noting
that the horizontal displacement of the ligament is nonzero
throughout the depth of ligament. The nonzero horizontal dis-
placement of the uncracked ligament ahead of the crack tip is
attributed to the elastic deformation of fly ash based GPC. Conse-
quently, the horizontal displacement jump will be nonzero
throughout the section depth of beam specimen except for the
point on the neutral axis. Therefore, it is unworkable to use the
magnitude of horizontal displacement jump to locate the crack
tip. Recognizing that the displacement derivative əu/əx is sensitive
to the displacement jump, therefore, attempts have been made to
use the displacement derivative to locate the crack tip. As an illus-
tration, Fig. 8 presents the displacement derivatives along the
crack for the same beam specimen under a same load. The dis-
placement derivates within the ROI, which is 50 mm in width
and 100 mm in height, are all plotted in this Figure. It is seen that
the displacement derivatives fluctuate within a small range
throughout the notch depth, which is in line with expectations as
the fly ash based GPC on the two sides of precut notch are
uncracked and thereby the displacement derivatives are small in
magnitude. At the tip of precut notch (y = 0), drastic increases of
displacement derivatives occur abruptly, implying the opening of
crack at the tip of precut notch. After that, the displacement deriva-
tives decrease as the distance to the tip of precut notch increases.
At the point where y = 10.80 mm, the displacement derivatives
turn into small fluctuations again, indicating that the opening of
crack ends at this point which is namely the crack tip. By this
means, the crack extension length in Fig. 8 is determined to be
10.80 mm finally.
The vertical displacement field obtained by DIC is employed in
this study to achieve the mid-span deflection d of beam specimen,
which is indispensible for the calculation of macro-average frac-
ture energy. First, the measurement results of vertical displace-
ment are removed off the rigid body motions induced by the
deformations of supports. Then, the average of the vertical
-3
x 10
5 7
4
3
∂u/∂x
2 1
1 0
0 a 0 /d = 0.3
-1
-40 -30 -20 -10 0 10 20 30 40
y (mm)
Fig. 8. Horizontal displacement derivatives of beam specimen with a0/d = 0.4 under
peak load.
displacements over a 10  10 mm square that is 10 mm below
the action point of load is computed as the mid-span deflection
of beam specimen. Fig. 9 plots the P-d curves of fly ash based
GPC for the four batches of beam specimens. As it is seen, the P-d
curves of fly ash based GPC are of similar shape as that of OPC con-
crete. Furthermore, it is seen that the shapes of the P-d curves of
beam specimens with different notch-depth ratios are also similar.
Seemingly, the pre-peak part of the P-d curve of fly ash based GPC
exhibits a linear relationship. At the beginning of loading, a linear
relationship between load and deflection is in line with expecta-
tions as the fly ash based GPC is undergoing linear elastic deforma-
tion. As the load approaches the peak, however, the linear
relationship between load and deflection conflicts with expecta-
tions as the fly ash based GPC is a quasi-brittle material in which
crack will initiate and propagate in a steady state before the occur-
rence of peak load. As the crack opening will cause a reduction in
the flexural stiffness of beam specimen and thereby an increment
in the deflection, the P-d curve is therefore expected to deviate
from the linear relationship accompanied by a reduction in the
inclination. Nevertheless, the flexural stiffness of beam specimen
may be insensitive to the crack opening and the amount of stiffness
loss may be insignificant in the stage of steady-state crack propa-
gation. As a result, the nonlinearity between load and deflection
of fly ash based GPC before the occurrence of peak load may be
too slight to be detectable in the overall P-d curve. To evidence this,
a zoom-in P-d curve of the beam specimen with a notch-depth
ratio of 0.4 around the peak load is also presented in the upper
righter corner of Fig. 9. As it is seen, the P-d curve of fly ash based
GPC does exhibit slight nonlinearity before the occurrence of peak
load. Noting that the pre-peak part of the P-d curve of fly ash based
GPC generally exhibits a near-linear relationship, the determina-
tion of crack initiation in fly ash based GPC with the use of P-d
curve will be very challenging. As will be shown later, the onset
of crack in fly ash based GPC can be determined without much
effort with the use of P-CTOD or P-Da curve. In the post-peak part
of the P-d curve of fly ash based GPC, the load decreases as the
deflection increases. It is an outcome attributed to the drop of flex-
ural resistance of beam specimen induced by the continuous prop-
agation of crack in the fly ash based GPC. Furthermore, it is seen
that the reduction in load bearing capacity is slight while the incre-
ment in deflection is more significant right after the occurrence of
peak load, as shown in the zoom-in P-d curve, implying that the
crack propagation in the fly ash based GPC is in a semi-stable state
5
6 4
3
5
2
4
P (kN)
3
0
0.05 0.1 0.15 0.2
2 a 0 /d = 0.2
1 a 0 /d = 0.4
a 0 /d = 0.5
50 60 0
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
δ (mm)
Fig. 9. P-d curves of fly ash based GPC.
 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380 377

at that time. As the crack keeps propagating in the fly ash based
GPC, it will finally steps into an unstable state, as manifested in
the P-d curve by the rapid fall of load bearing capacity and fast
increase of deflection, which will cause the rupture of beam spec-
imen ultimately.
3.2. Crack tip opening displacement and crack extension length
The P-CTOD curves of fly ash based GPC are presented in Fig. 10
for the four batches of beam specimens. It is seen that the P-CTOD
curves for beam specimens with different notch-depth ratios are of
similar shape. Hereafter, mainly the P-CTOD curve for the beam
specimen with a notch-depth ratio of 0.4 is discussed in detail,
with its zoom-in curve shown in the upper right corner of
Fig. 10. As per the inclination, the P-CTOD curve of fly ash based
GPC can be clearly divided into four segments, i.e., the segment
oa, ab, bc, and ce, which correspond to four stages of crack growth.
In the first stage (segment oa), the CTOD increases proportionally
to load with an inclination of nearly 90°. As will be evidenced later
by the P-Da curves of fly ash based GPC, no crack is formed at the
tip of precut notch in this period. Therefore, the increment in CTOD
in this stage is mainly attributed to the small linear elastic defor-
mation of fly ash based GPC. At the end of first stage where the load
is about 76.1% of peak load, the P-CTOD curve begins to deviate
slightly from the linear relationship accompanied by a reduction
in the inclination, implying the occurrence of damage namely the
initiation of crack at the tip of precut notch. Recalling the P-d curve
of fly ash based GPC in Fig. 9 in which the crack initiation is almost
invisible, the crack initiation is much more evident in the P-CTOD
curve of fly ash based GPC. Until the occurrence of peak load (point
b) where the second stage ends, the nonlinearity in P-CTOD rela-
tionship gets more apparent and the inclination of P-CTOD curve
becomes smaller. As a whole, the magnitude of CTOD is small
and the load bearing capacity of beam specimen is ascending in
this stage, implying that the crack propagation in the fly ash based
rate. In the third stage, the CTOD increases about 3 times (from
11.3 mm to 34.3 mm). Meanwhile, the load decreases from peak
by 5.8% only. The substantial increases in CTOD and insignificant
losses in load bearing capacity, which is an outcome attributed
to the quasi-brittleness of fly ash based GPC as well, imply that
the crack propagation in the fly ash based GPC steps into a semi-
stable state. In the fourth stage, a faster drop of the load bearing
capacity of beam specimen will commence along with a rapid
increase in CTOD, showing that the crack propagation in the fly
ash based GPC enters an unstable state which finally causes the
rupture of beam specimen.
Fig. 11 presents the P-Da curves of fly ash based GPC for the four
batches of beam specimens, where both non-normalized and nor-
malized curves are plotted. It is obvious that the shapes of the P-Da
curves of beam specimens with different notch-depth ratios are
similar. Again, the P-Da curve for the beam specimen with a
notch-depth ratio of 0.4 is discussed in detail. Likewise, the P-Da
curve of fly ash based GPC can also be clearly divided into four seg-
ments as per the inclination, i.e., segment oa, ab, bc, and ce in
Fig. 11(a), which correspond to four stages of crack growth. In
the first stage (segment oa), the crack extension length is zero,
demonstrating without doubt that no crack is formed at the tip
of precut notch and the fly ash based GPC undergoes a linear elastic
deformation. In the second stage (segment ab), crack has initiated
at the tip of precut notch and propagates in the fly ash based GPC,
7
(a) a 0 /d = 0.2
6 a 0 /d = 0.3
a 0 /d = 0.4
5 a 0 /d = 0.5
b
c
4
P (kN)

GPC is in a steady state. As fly ash based GPC is quasi-brittle in nat-

ure, it is therefore inferred that fracture process zone (FPZ) is for- 3
a
mulated in this stage, in which the cohesive force along the crack
interface may account for the steady-state propagation of crack. 2
After the occurrence of peak load, the CTOD continues to increase
while the load bearing capacity of beam specimen turns to
1
decreasing. As per the loss rate of load bearing capacity, it is further
e
divided into two stages, i.e., the third stage (segment bc) with a
small loss rate and the fourth stage (segment ce) with a large loss 0
0 10 20 30 40 50 60 70 80
Δ a (mm)

7 5
b (b) 1.1
c
4 1.0
6
a 0.9
3
5 0.8

2 0.7
4
e 0.6
P (kN)

P/Pmax

1
0.5
3
0 0.4
0 0.05 0.1 0.15
2 a 0 /d = 0.2 0.3 a 0 /d = 0.2

a 0 /d = 0.3 a 0 /d = 0.3
0.2
1 a 0 /d = 0.4
a 0 /d = 0.4
0.1
a 0 /d = 0.5
a 0 /d = 0.5
0
0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0 0.2 0.4 0.6 0.8 1.0
CTOD (mm) Δ a /( d -a 0 )

Fig. 10. P-CTOD curves of fly ash based GPC. Fig. 11. P-Da curves of fly ash based GPC.
378 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380
as reflected by the non-zero values of crack extension length.
Therefore, the onset of crack in the fly ash based GPC can be evi-
dently identified by making use of the P-Da curve. Furthermore,
the increment of crack extension length in this stage is slight as
a whole, which indicates that the crack propagation in the fly ash
based GPC keeps in a steady state. Again, it is attributed to the
quasi-brittleness of fly ash based GPC which leads to the appear-
ance of cohesive force along the crack interface and thereby the
formation of FPZ. In the third stage (segment bc), the crack propa-
gation in the fly ash based GPC steps into a semi-stable state, man-
ifested by large increases in the crack extension length while
insignificant losses in the load bearing capacity of beam specimen.
It again validates the quasi-brittleness of fly ash based GPC. In the
final stage (segment ce), not only significant increases of crack
extension length but also substantial losses of load bearing capac-
ity of beam specimen is observed, implying that the crack propaga-
tion in the fly ash based GPC enters an unstable state which leads
to the rupture of beam specimen ultimately. From Fig. 11(b), where
the crack extension length is normalized in terms of the depth of
ligament, it is interesting to see that the normalized crack exten-
sion lengths of the inflection points in the normalized P-Da curves,
where the load bearing capacity of beam specimen turns to rapid
decreasing, almost keep a constant value of 0.4 regardless of the
depth of precut notch. Recalling that this inflection point corre-
sponds to the onset of unstable crack propagation, it is therefore
speculated that a specific value of normalized crack extension
length, which is determined as 0.4 according to the test results
achieved in this study, may be found and employed as a fracture
criterion to pronounce the onset of unstable crack propagation in
fly ash based GPC regardless of the depth of precut notch.
3.3. Fracture energy and critical stress intensity factor
Fracture energy GF is the amount of energy necessary to create a
crack on unit surface area projected in a plane parallel to the direc-
tion of propagation of crack. In this study, the fracture energy was
calculated by the work of fracture method as given in the RILEM
guidelines [40], i.e.,
W 0 þ mgd0
GF ¼
Alig
where W0 is the area under the P-d curve of fly ash based GPC, m is
the mass of the beam specimen between the supports, g is the
acceleration of gravity, d0 is the deflection at the final failure of
beam specimen, and Alig is the area of fracture ligament of the beam
specimen.
The critical stress intensity factor KIC, also known as fracture
toughness, is used to indicate the magnitude of the stress concen-
tration that exists in front of the crack tip when the crack starts to
propagate. For three-point bending test of notched beam, the stress
intensity factor K calculated according to the theory of linear elas-
tic fracture mechanics is given by [43]
3Pl pffiffiffiffiffiffiffi
K¼ 3=2
paFðaÞ
2bd
1 1:99  a  ð1  aÞ  ð2:15  3:93a þ 2:7a2 Þ
FðaÞ ¼ pffiffiffiffi 
p ð1 þ 2aÞ  ð1  aÞ3=2
where P is the load, l is the span length of beam specimen, b is the
width of beam specimen, d is the depth of beam specimen, and
a = a/d, in which a is the elastic crack length. The critical stress
intensity factor can be obtained using Eqs. (2) and (3) with the
use of peak load and critical elastic crack length that corresponds
to the peak load. In the case of ideal brittle materials, the critical
elastic crack length is namely the notch depth because the onset
of crack in the ideal brittle material will immediately induce the
rupture of the material. In terms of fly ash based GPC, the critical
crack length consists of two parts, i.e., notch depth and crack exten-
sion length. The crack extension length of fly ash based GPC com-
prises not only the length of traction-free crack but also the
length of FPZ, which may form ahead of the traction-free crack as
the fly ash based GPC is a quasi-brittle material. As a result, the crit-
ical crack length of fly ash based GPC may not be directly applied to
Eqs. (2) and (3) for the calculation of critical stress intensity factor
since it may not be a elastic crack length as required. Instead, the
effective critical elastic crack length, which is equivalent to the crit-
ical crack length, should be used if Eqs. (2) and (3) are employed to
calculate the critical stress intensity factor of fly ash based GPC.
According to the theory of linear elastic fracture mechanics, the
load, CMOD, and elastic crack length conform to the relationship
given by [43]
6Pla
CMOD ¼ 2
V 1 ðaÞ ð4Þ
Ebd
0:66
V 1 ðaÞ ¼ 0:76  2:28a þ 3:87a2  2:04a3 þ ð5Þ
ð1  aÞ2
where E is the elastic modulus of fly ash based GPC. In this study,
the CMOD of fly ash based GPC has been obtained with the help
of DIC (Fig. 6). Therefore, Eqs. (4) and (5) can be exploited to calcu-
late the effective critical elastic crack length of fly ash based GPC.
First, the elastic modulus of fly ash based GPC can be obtained by
making use of the linear part of P-CMOD curve. The effective critical
elastic crack length of fly ash based GPC can then be achieved by
substituting the peak load and the CMOD that corresponds to the
peak load into Eqs. (4) and (5).
Tables 2 and 3 summarize the fracture properties of the fly ash
based GPC obtained with the use of initial crack length and effec-
tive elastic crack length, respectively. As mentioned earlier, three
beam specimens were tested for each case of notch-depth ratio.
The mean m and standard deviation r of the test results of the
three beam specimens are also shown in Tables 2 and 3. It is seen
that the fracture energy and critical stress intensity factor (regard-
less of which crack length is used) of fly ash based GPC do not keep
ð1Þ
constant as the notch depth changes. Nevertheless, the discrete-
ness of a degree observed in this study does not lead to a conclu-
sion that the fracture energy and critical stress intensity factor of
fly ash based GPC have size effect. To evidence this, the standard
deviation is employed to quantify the discreteness of fracture
energy. In terms of beam specimens with different notch depths,
the standard deviation of fracture energy is computed to be 11.9.
In terms of beam specimens with same notch depth, the standard
deviation of fracture energy is 13.3, 8.6, 12.3, and 17.4 for the
notch-depth ratio of 0.2, 0.3, 0.4, and 0.5, respectively, which is
an outcome mainly attributed to the heterogeneous meso-
structure of fly ash based GPC. Therefore, it is obvious that the dis-
creteness of fracture energy induced by the variation of notch
depth is comparable to or smaller than the counterpart caused
ð2Þ
by the heterogeneity of the meso-structure of fly ash based GPC.
In the light of the test results achieved from the beam specimens
with same dimension but different notch depths, it seems that
ð3Þ no significant size effect appears in the fracture energy and critical
stress intensity factor of fly ash based GPC. Nevertheless, it is worth
noting that previous studies on size effect of OPC concrete have
been mainly carried out by making use of geometrically similar
specimens of different sizes, it therefore may not be rigorous
enough to study the size effect of fly ash based GPC by means of
single size variable notch test method. Furthermore, it may also
be insufficient to make a conclusion on the size effect of fly ash
based GPC as per the test results of beam specimens with same
 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380 379

Table 2
Fracture properties of fly ash based GPC obtained with initial crack length.

a0/d Specimen Pmax mP rP GF mG rG KIC mK rK

1/2
(kN) (N/m) (MPa mm )
1 7.94 191.1 37.3
0.2 2 7.93 7.46 0.81 167.6 175.9 13.3 37.2 35.0 3.8
3 6.52 168.9 30.6
1 4.95 173.5 30.1
0.3 2 5.89 5.42 0.47 188.6 178.7 8.6 35.9 33.0 2.9
3 5.40 173.9 32.9
1 4.23 158.7 33.5
0.4 2 4.52 4.44 0.19 159.7 152.1 12.3 35.8 35.2 1.5
3 4.57 138.0 36.3
1 2.97 147.6 31.6
0.5 2 3.37 3.03 0.31 177.4 167.7 17.4 35.9 32.3 3.3
3 2.77 178.0 29.5

Table 3
Fracture properties of fly ash based GPC obtained with effective elastic crack length.

a0/d Specimen E mE rE aeff ma ra KIC mK rK

(GPa) (mm) (MPa mm1/2)
1 35.6 34.0 53.6
0.2 2 31.6 32.6 2.6 30.2 30.2 3.8 48.4 46.1 8.9
3 30.7 26.5 36.3
1 32.1 42.9 42.6
0.3 2 33.0 32.9 0.7 43.5 40.2 5.3 51.5 43.5 7.5
3 33.5 34.1 36.5
1 65.1 45.7 39.4
0.4 2 35.1 35.2 0.1 45.9 45.5 0.5 42.4 41.2 1.6
3 35.3 44.9 41.7
1 33.6 59.7 44.2
0.5 2 36.1 35.0 1.3 59.4 61.4 3.2 49.6 48.4 3.7
3 35.2 65.0 51.3

dimension but different notch depth only. Instead, the size effect of propagation. Especially, the onset of crack can be effectively
fly ash based GPC deserves a separate investigation in which geo- identified from the P-CTOD and P-Da curves, while it is
metrically similar beam specimens of different sizes should be barely manifested in the P-d curve.
used. However, it lies beyond the scope of this study and it will (3) The normalized crack extension length almost keeps con-
be carried out in the future. stant for different depths of precut notch at the inflection
point where the load bearing capacity of beam specimen
4. Conclusions turns to rapid decreasing. It is therefore speculated that
the normalized crack extension length may be employed
The fracture behavior, especially the crack growth behavior, of as a fracture criterion to pronounce the onset of unstable
fly ash based GPC was investigated in this study with the help of crack propagation in fly ash based GPC.
DIC. The experimental work consisted of three-point bending tests
on single edge notched beam specimens with same dimension but
different notch depths. The displacement field of beam specimen
Acknowledgements
was measured by DIC throughout the test. The mid-span deflec-
tion, CMOD, CTOD, and crack extension length were obtained by
The work described in this paper was supported in part by
making use of the measured displacement field and the crack
grants from the National Natural Science Foundation of China (Pro-
growth in fly ash based GPC were elaborated upon them. The frac-
ject No. 51578424, 51208384), grants from the Science Technology
ture energy and fracture toughness were calculated according to
Department of Zhejiang Province, China (Project No. 2017C33242,
the theory of linear elastic fracture mechanics. The following con-
2012R10071), and grants from Wenzhou Science and Technology
clusions are drawn from this study:
Bureau (Project No. S20150018, S20140010).

(1) DIC can be a very useful substitute for conventional mea-

surement technique in the monitoring of crack tip field. It
is featured with the capability to measure COD at any posi-
tion along the crack. The accuracy of DIC for COD measure-
ment is comparable to that of clip gauge. The crack
extension length, which is not readily available through
other methods, can also be obtained by DIC without much
effort.
(2) The crack growth of fly ash based GPC can be divided into
four stages, i.e., uncracked state, steady-state crack propaga-
tion, semi-stable crack propagation, and unstable crack
Compliance with ethical standards
Conflict of interest: The authors declare that they have no con-
flict of interest.
References
[1] J. Davidovits, Geopolymer Chemistry and Application, fourth ed., Institut
Géopolymère, Saint-Quentin, 2015.
[2] F.U.A. Shaikh, Review of mechanical properties of short fibre reinforced
geopolymer composites, Constr. Build. Mater. 43 (2) (2013) 37–49.
380 Z.L. Xie et al. / Construction and Building Materials 155 (2017) 371–380
[3] P.W. Ken, M. Ramli, C.C. Ban, An overview on the influence of various factors on
the properties of geopolymer concrete derived from industrial by-products,
Constr. Build. Mater. 77 (2015) 370–395.
[4] B. Singh, G. Ishwarya, M. Gupta, S.K. Bhattacharyya, Geopolymer concrete: a
review of some recent developments, Constr. Build. Mater. 85 (2015) 78–90.
[5] P. Duxson, J.L. Provis, G.C. Lukey, J.S.J. van Deventer, The role of inorganic
polymer technology in the development of ‘green concrete’, Cem. Concr. Res.
37 (12) (2007) 1590–1597.
[6] M. Weil, K. Dombrowski, A. Buchawald, Life-cycle analysis of geopolymers
ISBN-13:978 1 845691 449 4, in: J.L. Provis, J.S.J. van Deventer (Eds.),
Geopolymers, Structure, Processing, Properties and Applications, Woodhead
Publishing Limited, Abington Hall, Cambridge, UK, 2009, pp. 194–212.
[7] E.U. Von Weizsäker, K.C. Hargroves, M.H. Smith, C. Desha, P. Stasinopoulos,
Factor Five: Transforming the Global Economy Through 80% Improvements in
Resource Productivity, Earthscan, London, 2009.
[8] B.C. McLellan, R.P. William’s, J. Lay, A. van Riessen, G.D. Corder, Costs and
carbon emissions for geopolymer pastes in comparison to ordinary Portland
cement, J. Clean. Prod. 19 (9–10) (2011) 1080–1090.
[9] J.S.J. van Deventer, J.L. Provis, P. Duxson, Technical and commercial progress in
the adoption of geopolymer cement, Miner. Eng. 29 (3) (2012) 89–104.
[10] J.L. Provis, P. Duxson, J.S.J. Van Deventer, The role of particle technology in
developing sustainable construction materials, Adv. Powder Technol. 21 (1)
(2010) 2–7.
[11] P.N. Lemougna, K.T. Wang, Q. Tang, U. Chinje Melo, X.M. Cui, Recent
developments on inorganic polymers synthesis and applications, Ceram. Int.
42 (14) (2016) 15142–15159.
[12] D. Hardjito, S.E. Wallah, D.M. Sumajouw, B.V. Rangan, On the development of
fly ash-based geopolymer concrete, ACI Mater. J. 101 (6) (2004) 467–472.
[13] M. Ahmaruzzaman, A review on the utilization of fly ash, Prog. Energy
Combust. Sci. 36 (3) (2010) 327–363.
[14] A.M. Rashad, A brief on high-volume class F fly ash as cement replacement – a
guide for civil engineer, Int. J. Sustain. Built Environ. 4 (2) (2015) 278–306.
[15] V.M. Malhotra, P.K. Mehta, High-performance, high-volume fly ash concrete:
materials, mixture, proportioning, properties, construction practice, and case
histories, second ed., 2005 Supplementary Cementing Materials for
Sustainable Development Incorporated Ottawa.
[16] A.M. Rashad, A comprehensive overview about the influence of different
admixtures and additives on the properties of alkali-activated fly ash, Mater.
Des. 53 (1) (2014) 1005–1025.
[17] X.Y. Zhuang, L. Chen, S. Komarneni, C.H. Zhou, D.S. Tong, H.M. Yang, W.H. Yu,
H. Wang, Fly ash-based geopolymer: clean production, properties and
applications, J. Clean. Prod. 125 (2016) 253–267.
[18] VicRoads, Standard Specification Section 703, General Concrete Paving, 2010.
[19] P.K. Sarker, Analysis of geopolymer concrete columns, Mater. Struct. 42 (6)
(2009) 715–724.
[20] K.M. Mo, U.J. Alengaram, M.Z. Jumaat, Structural performance of reinforced
geopolymer concrete members: a review, Constr. Build. Mater. 120 (2016)
251–264.
[21] J. Aldred, J. Day, Is geopolymer concrete a suitable alternative to traditional
concrete? In: 37th Conference on Our World in Concrete & Structures, 29–31
August, Singapore, 2012.
[22] Z. Pan, J.G. Sanjayan, B.V. Rangan, Fracture properties of geopolymer paste and
concrete, Mag. Concr. Res. 63 (10) (2011) 763–771.
[23] P.K. Sarker, R. Haque, K.V. Ramgolam, Fracture behaviour of heat cured fly ash
based geopolymer concrete, Mater. Des. 44 (2013) 580–586.
[24] P. Nath, P.K. Sarker, Fracture properties of GGBFS-blended fly ash geopolymer
concrete cured in ambient temperature, Mater. Struct. 50 (1) (2017) 1–12.
[25] T. Phoo-ngernkham, V. Sata, S. Hanjitsuwan, C. Ridtirud, S. Hatanaka, P.
Chindaprasirt, Compressive strength, bending and fracture characteristics of
high calcium fly ash geopolymer mortar containing portland cement cured at
ambient temperature, Arabian J. Sci. Eng. 41 (4) (2016) 1263–1271.
[26] D.P. Dias, C. Thaumaturgo, Fracture toughness of geopolymeric concretes
reinforced with basalt fibers, Cem. Concr. Compos. 27 (1) (2005) 49–54.
[27] D.S. Raj, R. Abraham, N. Ganesan, D. Sasi, Fracture properties of fibre reinforced
geopolymer concrete, Int. J. Sci. Eng. Res. 4 (5) (2013) 75–80.
[28] J.G. Sanjayan, A. Nazari, H. Pouraliakbar, FEA modeling of fracture toughness of
steel fibre-reinforced geopolymer composites, Mater. Des. 76 (2015) 215–222.
[29] J.W. Dally, W.F. Riley, Experimental Stress Analysis, fourth ed., McGraw Hill,
Columbus, 2005.
[30] Sharpe Jr bWN, Handbook of Experimental Solid Mechanics, Springer, US, New
York, 2008.
[31] I. Yamaguchi, A laser-speckle strain gauge, J. Phys. E Sci. Instrum. 14 (11)
(1981) 1270–1273.
[32] W.H. Peters, W.F. Ranson, Digital image techniques in experimental stress
analysis, Opt. Eng. 21 (3) (1982) 427–431.
[33] S. Roux, J. Réthoré, F. Hild, Digital image correlation and fracture: an advanced
technique for estimating stress intensity factors of 2D and 3D cracks, J. Phys. D
Appl. Phys. 42 (2) (2009). Article ID 214004.
[34] J.R. Yates, M. Zanganeh, Y.H. Tai, Quantifying crack tip displacement fields with
DIC, Eng. Fract. Mech. 77 (11) (2010) 2063–2076.
[35] S.G. Shah, J.M. Chandra Kishen, Fracture properties of concrete-concrete
interfaces using digital image correlation, Exp. Mech. 51 (3) (2010) 303–313.
[36] L.C.S. Nunes, J.M.L. Reis, Estimation of crack-tip-opening displacement and
crack extension of glass fiber reinforced polymer mortars using digital image
correlation method, Mater. Des. 33 (1) (2012) 248–253.
[37] L.C.S. Nunes, Crack-craze opening profiles near a crack tip in a
polytetrafluoroethylene, Polym. Testing 31 (3) (2012) 375–383.
[38] T.X. Tang, Z.P. Bažant, S.C. Yang, D. Zollinger, Variable-notch one-size test
method for fracture energy and process zone length, Eng. Fract. Mech. 55 (3)
(1996) 383–404.
[39] B.H. Bharatkumar, B.K. Raghuprasad, D.S. Ramachandramurthy, R. Narayanan,
S. Gopalakrishnan, Effect of fly ash and slag on the fracture characteristics of
high performance concrete, Mater. Struct. 38 (2005) 63–72.
[40] RILEM TC 50-FMC., Determination of the fracture energy of mortar and
concrete by means of three-point bend tests on notched beams, Mater. Struct.
18 (4) (1985) 287–290.
[41] P. Nath, P.K. Sarker, Effect of GGBFS on setting, workability and early strength
properties of fly ash geopolymer concrete cured in ambient condition, Constr.
Build. Mater. 66 (1) (2014) 163–171.
[42] ASTM C191-13, Standard Test Methods for Time of Setting of Hydraulic
Cement by Vicat Needle, ASTM International, West Conshohocken, PA, USA,
2013.
[43] H. Tada, P.C. Paris, G.R. Irwin, The Stress Analysis of Cracks Handbook, Del
Research Corporation, Hellertown, 1976.