Theoretical and Applied Fracture Mechanics 90 (2017) 133–141

Contents lists available at ScienceDirect

Theoretical and Applied Fracture Mechanics

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

Fractal analysis of crack paths into innovative carbon-based

cementitious composites
L. Restuccia a,⇑, A. Reggio a, G.A. Ferro a, R. Kamranirad b
a
Department of Structural, Geotechnical and Building Engineering, Politecnico di Torino, 10129 Turin, Italy
b
Department of Structural Engineering, Tarbiat Modares University, Tehran, Iran

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

Article history: Fractal geometry has been widely used in literature to characterize the mechanical behaviour of quasi-
Received 3 March 2017 brittle materials. In this work, innovative cementitious composites with carbon-based pyrolyzed
Revised 21 March 2017 micro-aggregates were tested until complete fracture and their fracture behaviour was studied in the
Accepted 21 March 2017
light of fractal geometry. Images of the crack paths across the tested specimens were acquired by
Available online 22 March 2017
Scanning Electron Microscopy (SEM) and their fractal dimension was calculated via the box counting
method. Results show that the pyrolyzed micro-aggregates, characterized by high strength and stiffness
Keywords:
due to their significant carbon content, are able to alter the crack path by increasing its tortuosity, thus
Cementitious composites
Crack path
inducing toughening mechanisms in the cementitious composites. This favourable behaviour is explained
Fractal dimension by means of fractal geometry: it is found that, the greater the fractal dimension of the crack path, the
Box counting higher the fracture energy.
Fracture energy Ó 2017 Elsevier Ltd. All rights reserved.
Fracture toughness

1. Introduction ture toughness, reporting a fractal dimension of 2.087 for cement

The word fractal, from the Latin adjective fractus, meaning bro-
ken, was coined by Mandelbrot in his fundamental essay [1] to
describe objects that were too irregular to fit into a traditional
Euclidean geometrical setting, where points, lines, surfaces, and
volumes have integer topological dimensions of 0, 1, 2, and 3,
respectively. Falconer [2] developed the fractal theory into the
mathematical field, giving prominence to the concept of fractal
dimension, a measure of self-similarity and irregularity across
scales, which can assume fractional values. Since then, fractal
geometry has been widely used in material science and engineer-
ing to quantitatively describe complex non-smooth objects, such
as rough surfaces [3] or grain boundaries [4], whose fractal dimen-
sion is usually greater than their topological dimension. Numerous
fractographic studies have also reported that fracture surfaces [5]
and crack paths [6] exhibit self-similar fractal nature and can be
analyzed by fractal geometry in order to characterize their rough-
ness and correlate it with mechanical properties [7]. This kind of
approach has been successfully applied to cement-based materials
[8]: the surface of hydrated Portland cement paste has been proved
to be fractal in character [9,10]; Lange et al. [11] demonstrated a
positive correlation between fracture surface roughness and frac-
⇑ Corresponding author.
E-mail address: luciana.restuccia@polito.it (L. Restuccia).
paste and of 2.117 for fine mortar; Saouma and Barton [12] showed
that cracked concrete surfaces are fractal and determined the frac-
tal dimension of their one-dimensional fracture profiles in the nar-
row range from 1.06 to 1.12. Fractal geometry has also been used
for explaining the size effects on the mechanical properties of
cementitious materials. In particular, a multifractal scaling law
(MFSL) has been defined, according to which the decrease in the
nominal tensile strength of concrete with the increasing size of
specimen is due to the influence of the microstructure disorder
[13–15]. Moreover, a similar approach has been used together with
fragmentation theory to account for size effects on dissipated
energy density in compression [16]. Issa et al. [17] analyzed the
fracture surfaces of normal strength concretes with four different
maximum aggregate sizes, highlighting their fractal characteristics
and deriving a linear correlation between the fractal dimension of
fracture surface, with values between 2.15 and 2.29, and the frac-
ture toughness of concrete, represented by fracture energy; a sim-
ilar result was reported in [18], although a simple relation between
fractal dimension and fracture energy could not be established, and
fractal dimension of crack paths was found to range from 1.03 to
1.25 depending on the nature of aggregates and the resolution of
the processed images.
Following up previous studies concerning the mechanical beha-
viour and the manufacturing optimization of cementitious com-
posites with carbon-based inerts [19–22], the aim of the present

http://dx.doi.org/10.1016/j.tafmec.2017.03.016
0167-8442/Ó 2017 Elsevier Ltd. All rights reserved.
134 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141

work is to apply fractal geometry to innovative composites with

pyrolyzed hazelnut shells added as micro-aggregates, in order to
explain the toughening mechanisms observed in their experimen-
tal fracture behaviour [23,24]. Three Points Bending (TPB) tests
until complete fracture were carried out on prismatic specimens
made of the innovative cementitious composites and of a plain
cement paste. The addition of a small amount of carbonaceous
micro-aggregates strongly influenced the mechanical perfor-
mances of the cement paste, resulting in a significant increase in
strength together with an increment in fracture energy, and then
in ductility. Images of the crack paths across the tested specimens
were acquired by Scanning Electron Microscopy (SEM), subse-
quently adjusted and binarized via image processing techniques
and later examined via fractal geometry. The fractal dimension of
the crack paths was calculated by implementing an algorithm
based on the box counting method and a relationship with the val-
ues of fracture energy computed from the experimental loading
curves was finally established. The analysis of the results shows
that the increment of the fracture energy due to the addition of
the carbonaceous micro-aggregates is strongly related to an incre-
ment of fracture surface tortuosity.

2. Materials and methods

2.1. Manufacturing of materials and specimens

Cementitious composites were manufactured by using ordinary

Portland cement (Type I, CEM I 52, 5R, light grey color), deionized
water and by adding to the cement paste carbon-based micro-
aggregates produced from the pyrolysis of hazelnut shells.
Pyrolysis was carried out into a hermetically sealed reactor,
where an inert environment was ensured by a continuous flowing
of nitrogen. A heating ramp of 6 °C/min was held for 1 h, reaching a
final temperature of 800 °C. After the pyrolysis process, the mate-
rial was collected from the reactor, grinded for 10 min into a plan-
etary mill to crush the bigger particles, and then milled for 1 h
using ethanol and alumina balls of 2 mm diameter. At the end of
the grinding cycle, the particle size was analyzed by laser granu-
lometry using a Malvern Particle Sizer, obtaining an average value
in the range of a few nanometers to 10 lm.
Mixing procedure is crucial for the performance of cementitious
composites [25], hence the same procedure described here was fol-
lowed to manufacture all the specimens. The pyrolyzed particles
were dispersed in a solution of water and superplasticizer (Mapei
Dynamon SP1), the latter intended for limiting as much as possible
the re-agglomeration problem of the particles [26,27] and for keep-
ing the water to cement (w/c) ratio equal to 0.35. The solution was
sonicated in an ultrasonic bath for 15 min and then transferred
into the mixing bowl of a homogenizer. The cement was gradually
added to the solution while operating the homogenizer at 440 rpm
for 2 min. Afterwards, the homogenizer speed was increased to
630 rpm for further 2 min, up to a total mixing time of 4 min. Var-
ious percentages of addition of pyrolyzed particles with respect to
the cement weight were considered, in agreement with previous
studies [28–30], and four different mixtures were obtained. The
weights of the materials used for each mixture are reported in
Table 1.
Fig. 1. TPB test setup.
A total of 32 prismatic specimens of dimensions 20 mm

20 mm
75 mm were manufactured by casting the mixtures into
acrylic molds and letting them mature for 24 h at room tempera-
ture in airtight plastic containers with 90% humidity level. After
maturation, the specimens were removed from the molds and
immersed in tap water for curing. At the end of a curing time of
either 7 or 28 days, the specimens were notched by means of a
Remet type TR100S abrasive cutter with 2 mm thick diamond
cut-off wheel, realizing a 6 mm deep U-shaped notch.
2.2. Experimental tests
Each notched specimen was subjected to a crack mouth opening
displacement-control Three Points Bending (TPB) test. TPB tests
were performed using a single column Zwick Line-Z010 testing
machine equipped with a 1 kN load cell and a clip-on strain gauge
to measure the Crack Mouth Opening Displacement (CMOD), as
shown in Fig. 1; a span length of 65 mm and a displacement rate
of 0.005 mm/min were adopted.
Besides the mechanical tests, the chemical characterization of
pyrolyzed hazelnut shells was carried out using the X-Rays Fluo-
rescence (XRF) technique. The chemical composition reported in
Table 2 shows that the pyrolyzed particles are almost exclusively
made of carbon and present very low percentages of impurities.

Table 1
Weights of the materials used for each mixture.

Mixture ID Cement (g) Water (g) w/c (–) SP1 (g) Pyrolyzed hazelnut shells (g) Pyrolyzed hazelnut shells (%)a
PLAIN CEM 214 74.9 0.35 3.21 – –
PY-HS_0.5% 214 74.9 0.35 3.21 1.07 0.5
PY-HS_0.8% 214 74.9 0.35 3.21 1.71 0.8
PY-HS_1% 214 74.9 0.35 3.21 2.14 1.0
a
Percentage with respect to the cement weight.
 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141 135

Table 2
Chemical composition of pyrolyzed hazelnut shells by XRF.

Components C Si Mg K Ca P Zn S Cr Cu Fe
Weight% 97.9 0.11 – 1.01 0.44 0.05 – 0.02 0.02 0.19 0.25

2.3. Analysis of fractal geometry 2.3.2. The box counting method

2.3.1. Image processing
A prerequisite for the application of the box counting method is
the detection of the crack path edge by means of image processing
techniques.
SEM images are in grayscale, with each pixel color specified by a
number ranging from 0 (black) to 255 (white), while binary
images, including pixel values equal to either 0 (black) or 1 (white),
are needed for the edge detection. To transform a grayscale image
into a binary one, the image quality must be first enhanced by
adjusting contrast and sharpness and filtering the noise; after-
wards, a threshold level is set so that pixels with values smaller
than the threshold are assigned to be equal to zero, while pixels
with values larger than the threshold are assigned to be equal to
unity. At the end of a number of iteration, the binarization process
leads to detect the ideally one pixel wide edge of the crack path,
identified by the elements equal to 0 in the data matrix obtained
from the image matrix (Fig. 2).
Box counting is one of the most effective and useful method for
obtaining fractal dimensions. The procedure is here applied by cov-
ering the image of the crack path by a mesh grid of characteristic
length rk and counting the number Nk of grid boxes needed to over-
lap the crack path. The count is repeated by increasing the box size
until the largest box fits the whole image and trivially returns 1 for
the box number. If the crack path exhibits fractal geometry, the
relationship between Nk(rk) and rk obeys a power law [2]:
Nk ðr k Þ  cr D
k ð1Þ
where c is a constant and D is the fractal dimension of the crack
path. Taking logarithms of (1) and reporting the data points (log rk,
log Nk) on a log-log plot, fractal dimension D can be estimated by
linear least squares as minus the slope of the regression straight line
over a suitable range of rk.
In the present work, the box size is parameterized as rk = kd
where k = 1, 2, 4, . . . , 2n is a multiplying factor given in geometric
progression of ratio 2 and d is the size of a pixel point. In Fig. 3,

Fig. 2. Example of crack path detection in a binary image and the corresponding data matrix.

Fig. 3. Example of application of the box counting method on a binary matrix, three subsequent iterations: (a) k = 2, Nk = 14; (b) k = 4, Nk = 7 and (c) k = 8, Nk = 4.
136 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141

of cementitious composite with the 0.8% addition of pyrolyzed

Fig. 4. Load – CMOD curves from the TPB tests for plain cement paste (specimen ID:
PLAIN CEM – #1) and cement paste with 0.8% addition of pyrolyzed micro-
aggregates (specimen ID: PY-HS_0.8% - #2).
an example of three subsequent iterations of box counting is
shown.
3. Results and discussion
3.1. Fracture energy
In Fig. 4, a comparison between the load-CMOD curves obtained
from the TPB tests for the cement paste without and with the pyr-
olyzed micro-aggregates after a curing time of 7 days is shown: the
thick line refers to the specimen of plain cement paste (specimen
ID: PLAIN CEM – #1), while the thin line refers to the specimen
hazelnut shells (specimen ID: PY-HS_0.8% – #2). Comparing the
plots, it is apparent that the addition of pyrolyzed micro-
aggregates improves the mechanical behaviour of the cementitious
composite by significantly increasing both its flexural strength and
its ductility with respect to the plain cement paste. The differences
in the post-cracking behaviour are particularly remarkable: while
the plain cement paste undergoes brittle fracture, the cementitious
composite with pyrolyzed micro-aggregates shows an ultimate
displacement that is much greater than the value corresponding
to the appearance of the first crack, thanks to a slow crack propa-
gation and to the absorption of a large amount of energy before
incurring in complete fracture. Such an increase in fracture tough-
ness can therefore be quantified by evaluating the fracture energy
GF of the cementitious composites.
In this work, fracture energy GF is determined from the load-
CMOD curves of notched specimens under TPB test by applying
the method provided by the Japan Concrete Institute Standard
JCI-S-001-2003 [31], whose recommendations, represented in
Fig. 5, are satisfied by the geometry of the tested specimens. Due
to the unconventional dimensions with which the tested speci-
mens were manufactured, it is not possible to apply either the
standard RILEM TC-50 [32], to determine the fracture energy GF,
or the standard RILEM TC-89 [33], to determine the critical stress
intensity factor KSIc and the critical crack tip opening displacement
CTODc. According to the adopted method, it is
0:75W 0 þ W 1
GF ¼ ð2Þ
Alig
where GF is the fracture energy; Alig is the area subjected to rupture
above the notch, referred to as the ‘‘ligament”; W0 is the area below
the load – CMOD curve up to the rupture of the specimen; W1 is the

Fig. 5. Recommendations on the geometry of the notched specimens in JCI-S-001-2003 [31].

Table 3
Determination of fracture energy GF in accordance with standard JCI-S-001-2003 [31], specimens after 7 curing days.

Specimen ID JCI-S-001-2003
Mixture # W0 (Nmm) 0.75 W0 (Nmm) CMODc (mm) W1 (Nmm) GF (N/mm) GF mean value
PLAIN CEM 1 1.38 1.03 0.03 1.26 0.008 0.016
PLAIN CEM 2 1.44 1.08 0.04 2.03 0.011
PLAIN CEM 3 2.55 1.91 0.06 3.15 0.018
PLAIN CEM 4 3.96 2.97 0.09 4.69 0.027
PY-HS_0.5% 1 4.45 3.34 0.10 5.15 0.030 0.030
PY-HS_0.5% 2 4.14 3.11 0.10 5.16 0.029
PY-HS_0.5% 3 4.34 3.26 0.10 5.03 0.029
PY-HS_0.5% 4 4.14 3.11 0.11 5.26 0.029
PY-HS_0.8% 1 5.00 3.75 0.10 5.17 0.031 0.032
PY-HS_0.8% 2 5.03 3.78 0.11 5.24 0.032
PY-HS_0.8% 3 4.97 3.72 0.11 5.34 0.032
PY-HS_0.8% 4 4.97 3.73 0.10 4.88 0.030
PY-HS_1% 1 3.05 2.29 0.06 2.76 0.018 0.022
PY-HS_1% 2 3.04 2.28 0.07 3.65 0.021
PY-HS_1% 3 3.84 2.88 0.08 3.86 0.024
PY-HS_1% 4 3.65 2.74 0.08 3.74 0.023
 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141 137

work done by the deadweight of specimen and loading, evaluated in which l is the loading span, L is the total length of the specimen,
as m1 is the mass of the notched specimen, m2 is the mass of the load-
  ing arrangement part not attached to testing machine but placed on
l
W 1 ¼ 0:75 m1 þ 2m2 g  CMOCc ð3Þ the specimen, g is the acceleration due to gravity and CMODc is the
L
crack mouth opening displacement at rupture.

Table 4
Determination of fracture energy GF in accordance with standard JCI-S-001-2003 [31], specimens after 28 curing days.

SPECIMEN ID JCI-S-001-2003
Mixture # W0 (Nmm) 0.75 W0 (Nmm) CMODc (mm) W1 (Nmm) GF (N/mm) GF mean value
PLAIN CEM 5 2.83 2.12 0.05 2.48 0.016 0.018
PLAIN CEM 6 3.39 2.54 0.06 3.05 0.020
PLAIN CEM 7 2.72 2.04 0.08 3.98 0.022
PLAIN CEM 8 1.23 0.92 0.06 3.05 0.014
PY-HS_0.5% 5 4.69 3.52 0.09 4.27 0.027 0.031
PY-HS_0.5% 6 4.07 3.05 0.08 4.08 0.025
PY-HS_0.5% 7 9.04 6.78 0.11 5.31 0.043
PY-HS_0.5% 8 5.35 4.01 0.08 4.13 0.029
PY-HS_0.8% 5 5.46 4.09 0.05 2.66 0.024 0.033
PY-HS_0.8% 6 5.91 4.43 0.15 7.45 0.042
PY-HS_0.8% 7 5.94 4.46 0.10 5.14 0.034
PY-HS_0.8% 8 6.16 4.62 0.07 3.71 0.029
PY-HS_1% 5 3.54 2.66 0.11 5.62 0.029 0.029
PY-HS_1% 6 3.95 2.96 0.11 5.39 0.029
PY-HS_1% 7 3.92 2.94 0.10 5.11 0.028
PY-HS_1% 8 3.76 2.82 0.11 5.29 0.028

Fig. 6. Mean values of fracture energy GF evaluated from TPB tests, 7 days cured specimens.

Fig. 7. Mean values of fracture energy GF evaluated from TPB tests, 28 days cured specimens.
138 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141

Results from the TPB tests reported in Tables 3 and 4 show an
increase of GF, with respect to the plain cement paste, for all the
percentages of pyrolyzed micro-aggregates added to the cementi-
tious composite, both after 7 and 28 curing days. In Figs. 6 and 7,
bar plots show the mean values of GF, calculated as average of four
specimens according to JCI-S-001-2003. It can be noticed that the
0.8% addition improves the most the fracture toughness of the
cementitious composites, with an increase, compared to plain
cement paste, of 69% and 75% in 7 and 28 curing days, respectively.
Since the difference between the results after 7 and 28 curing days
is not significant, the pyrolyzed micro-aggregates are proved to be
immediately able to confer a better mechanical behaviour to the
cementitious composites.
3.2. Toughening micromechanisms and fractal geometry of the crack
paths
The increase in fracture toughness and fracture energy due to
the pyrolyzed micro-aggregates can be explained by examining
the fractal characteristics of crack paths and fracture surfaces
[10,17].
Figs. 8 and 9 depict the crack paths and fracture surfaces across
the specimens of plain cement paste (PLAIN CEM – #1) and of
cementitious composite with 0.8% addition of pyrolyzed hazelnut
shells (PY-HS_0.8% – #2), respectively. It is apparent that the crack
in plain cement paste follows a linear path, as typical in brittle frac-
ture, while the crack in cementitious composite progresses with a

Fig. 8. Plain cement paste, specimen ID: PLAIN CEM – #1. (a) Crack path and (b) fracture surface [24].

Fig. 9. Cementitious composite with 0.8% addition of pyrolyzed hazelnut shells, specimen ID: PY-HS_0.8% – #2. (a) Crack path and (b) fracture surface [24].

Fig. 10. SEM image of the crack path across the cementitious composite with 0.8% addition of pyrolyzed hazelnut shells, specimen ID: PY-HS_0.8% – #2.
 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141 139

more articulated and tortuous path. By examining by SEM the

Fig. 11. Crack-front pinning as a toughening micromechanism in the cementitious
composite with the 0.8% addition of pyrolyzed hazelnut shells, specimen ID: PY-
HS_0.8% – #2.
paths at the microstructure level, it can be seen that the particles
of pyrolyzed hazelnut shells behave as attractors of fracture and
bifurcations (Fig. 10) and are responsible for toughening and
strengthening mechanisms [34,35]. In particular, the mechanism
known as crack-front spinning is observed (Fig. 11): the crack does
not possess enough energy to penetrate the high-strength carbon-
based particles, hence it tends to surround them, resulting in a
number of secondary crack paths.
SEM images of the crack paths were later treated by image pro-
cessing, in order to detect the crack edge and analyze its fractal
geometry in accordance with the methodology defined in Sec-
tion 2.3. As illustrated in Figs. 12 and 13, the linear least-squares
regression of the box counting data leads to values of the coeffi-
cient of determination R2 close to 1 for both the plain cement paste
(R2 = 0.9975) and the cementitious composite with 0.8% addition of
pyrolyzed hazelnut shells (R2 = 0.9959), proving a very good fit of
data points by linear regression and, hence, a distinctive fractal
geometry of the crack paths. The values obtained for the fractal
dimension D are 1.055 for the plain cement paste and 1.234 for
the cementitious composite with pyrolyzed hazelnut shells. Since

(a) (b)

(c) (d)

Fig. 12. Image processing and box counting method for plain cement paste (specimen ID: PLAIN CEM – #1). (a) SEM image, (b) binary image, (c) edge detection of the crack
path and (d) box counting results and linear least-squares regression.
140 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141

(a) (b)

(c) (d)

Fig. 13. Image processing and box counting method for the cementitious composite with the 0.8% addition of pyrolyzed hazelnut shells (specimen ID: PY-HS_0.8% – #2). (a)
SEM image, (b) binary image, (c) edge detection of the crack path and (d) box counting results and linear least-squares regression.

the fractal dimension is a measure of the complexity of an object
across different scales, the morphology and irregularity of a crack
path determine its fractal dimension: the crack path following a
more linear pattern, in the plain cement paste, has indeed a fractal
dimension closer to 1, or the dimension of a straight line, while the
more irregular and tortuous crack path, in the cementitious com-
posite with pyrolyzed micro-aggregates, shows a higher fractal
dimension. The fractal dimension of the crack path is therefore
related to the fracture toughness of the material: with increasing
fractal dimension, the crack consumes more energy to propagate
along a more complicated pattern, thus resulting in boosting up
the fracture toughness.
sion of the crack path showed an increment consistent with the
fracture energy, indicating an increment in tortuosity and, there-
fore, an increment of the fracture surface. The presence of the
micro-aggregates, characterized by high strength and stiffness
due to their significant carbon content, forced the crack path to
deviate its trajectory and surround them. Even if the quantity of
these micro-aggregates was very small, the specific surface area
was impressive, due to their micro-dimension, and the increase
of fracture energy was significant.
References

4. Conclusions
In the present work, a fractal analysis of the crack paths in inno-
vative composites with pyrolyzed hazelnut shells as micro-
aggregates has been presented. The principal goal of this approach
was to give a theoretical justification to the significant increment
of fracture energy experimentally measured in the specimens
added with the carbonaceous micro-aggregates. The fractal dimen-
[1] B.B. Mandelbrot, The Fractal Geometry of Nature, Freeman, San Francisco,
1982.
[2] K. Falconer, Fractal Geometry: Mathematical Foundations and Applications,
Wiley, Chichester, 1990.
[3] T. Ai, R. Zhang, H.W. Zhou, J.L. Pei, Box-counting methods to directly estimate
the fractal dimension of a rock surface, Appl. Surf. Sci. 314 (2014) 610–621.
[4] H. Khanbareh, X. Wu, S. van der Zwaag, Analysis of the fractal dimension of
grain boundaries of AA7050 aluminum alloys and its relationship to fracture
toughness, J. Mater. Sci. 47 (2012) 6246–6253.
[5] G.P. Cherepanov, A.S. Balankin, V.S. Ivanova, Fractal fracture mechanics – a
review, Eng. Fract. Mech. 51 (1995) 997–1033.
 L. Restuccia et al. / Theoretical and Applied Fracture Mechanics 90 (2017) 133–141
[6] G. Zheng, J. Zhao, L. Li, X. Cheng, M. Wan, A fractal analysis of the crack
extension paths in a Si3N4 ceramic tool composite, Int. J. Refract. Metals Hard
Mater. 51 (2015) 160–168.
[7] E. Charkaluk, M. Bigerelle, A. Iost, Fractals and fracture, Eng. Fract. Mech. 61
(1998) 119–139.
[8] S.P. Shah, C. Ouyang, D. Lange, Fracture behavior of cement based materials,
MRS Bull. 18 (1993) 55–59.
[9] D.N. Winslow, The fractal nature of the surface of cement paste, Cem. Concr.
Res. 15 (1985) 817–824.
[10] D.N. Winslow, J.M. Bukowsky, J.F. Young, The fractal arrangement of hydrated
cement pastes, Cem. Concr. Res. 25 (1995) 147–156.
[11] D. Lange, H. Jennings, S.P. Shah, Relationship between fracture surface
roughness and fracture behavior of cement paste and mortar, J. Am. Ceram.
Soc. 76 (1993) 589–597.
[12] V.E. Saouma, C.C. Barton, Fractals, fractures and size effects in concrete, J. Eng.
Mech. 120 (1994) 835–854.
[13] G.A. Ferro, Effetti di scala sulla resistenza a trazione dei materiali Ph.D. Thesis,
Politecnico di Torino, Torino, 1994.
[14] A. Carpinteri, G.A. Ferro, Size effects on tensile fracture properties: a unified
explanation based on disorder and fractality of concrete microstructure,
Mater. Struct. 27 (1994) 563–571.
[15] A. Carpinteri, B.M. Chiaia, G.A. Ferro, Scale dependence of tensile strength of
concrete specimens: a multifractal approach, Mag. Concr. Res. 50 (1998) 237–
246.
[16] G.A. Ferro, On dissipated energy density in compression for concrete, Eng.
Fract. Mech. 73 (2006) 1510–1530.
[17] Mohsen A. Issa, Mahmoud A. Issa, S. Islam, A. Chudnovsky, Fractal dimension –
a measure of fracture roughness and toughness of concrete, Eng. Fract. Mech.
70 (2003) 125–137.
[18] B. Chiaia, J.G.M. Van Mier, A. Vervuit, Crack growth mechanisms in four
different concretes: microscopic observations and fractal analysis, Cem. Concr.
Res. 28 (1998) 103–114.
[19] G.A. Ferro, J.M. Tulliani, P. Jagdale, L. Restuccia, New concepts for next
generation of high performance concretes, Proc. Mater. Sci. 3 (2014) 1760–
1766.
[20] G.A. Ferro, S. Ahmad, R.A. Khushnood, L. Restuccia, J.M. Tulliani, Improvements
in self-consolidating cementitious composites by using micro carbonized
aggregates, F&IS 30 (2014) 75–83.
[21] G. Ferro, J.M. Tulliani, A. Lopez, P. Jagdale, New cementitious composite
building material with enhanced toughness, Theor. Appl. Fract. Mech. 76
(2015) 67–74.
141
[22] R.A. Khushnood, L. Restuccia, G.A. Ferro, S. Ahmad, J.M. Tulliani, P. Jagdale,
Modified fracture properties of cement composites with nano/micro
carbonized bagasse fibers, F&IS 34 (2015) 534–542.
[23] L. Restuccia, G.A. Ferro, Nanoparticles from food waste: a ‘‘green” future for
traditional building materials, in: 9th International Conference on Fracture
Mechanics of Concrete and Concrete Structures (FraMCoS-9), Berkeley,
California, USA, 2016, Paper N. 276, 2016, http://dx.doi.org/10.21012/FC9.276.
[24] L. Restuccia, G.A. Ferro, Promising low cost carbon-based materials to improve
strength and toughness in cement composites, Constr. Build. Mater. 126
(2016) 1034–1043.
[25] J. Liu, Q. Li, S. Xu, Influence of nanoparticles on fluidity and mechanical
properties of cement mortar, Constr. Build. Mater. 101 (2015) 892–901.
[26] S. Montenegro Camacho, S. Bensaid, B. Ruggeri, L. Restuccia, G.A. Ferro, G.
Mancini, D. Fino, Valorisation of by-products/waste of agro-food industry by
the pyrolysis process, J. Adv. Catal. Sci. Technol. 3 (2016) 1–11.
[27] D. Wei, R. Dave, R. Pfeffer, Mixing and characterization of nanosized powders:
an assessment of different techniques, J. Nanopart. Res. 4 (2002) 21–41.
[28] L. Restuccia, Re-think, Re-use: Agro-food and C&D Waste for High-
performance Sustainable Cementitious Composites (Ph.D. Thesis), Politecnico
di Torino, Torino, 2016.
[29] S. Ahmad, R.A. Khushnood, P. Jagdale, J.M. Tulliani, G.A. Ferro, High
performance self-consolidating cementitious composites by using micro
carbonized bamboo particles, Mater. Des. 76 (2015) 223–229.
[30] S. Ahmad, G.A. Ferro, R.A. Khushnood, L. Restuccia, J.M. Tulliani, P. Jagdale,
Crack path and fracture surface modifications in cement composite, F&IS 34
(2015) 524–533.
[31] Japan Concrete Institute, Method of Test for Fracture Energy of Concrete by
Use of Notched Beam (JCI-S-001-2003), Tokyo, Japan, 2003.
[32] RILEM TC 50-FMC, Determination of the fracture energy of mortar and
concrete by means of three-point bend test on notched beams, Mater. Struct.
18 (1985) 285–290.
[33] RILEM TC 89-FMT, Determination of fracture parameters of plain concrete
using three-point bend tests, Mater. Struct. 23 (1990) 457–460.
[34] V.C. Li, M. Maalej, Toughening in cement based composites. Part I: cement,
mortar, and concrete, Cem. Concr. Compos. 18 (1996) 223–237.
[35] Q. Qin, Introduction to the composite and its toughening mechanism, in: Q.
Qin, J. Ye (Eds.), Toughening Mechanisms in Composite Materials, Woodhead
Publishing, 2015, pp. 1–32.