Journal of Cleaner Production 149 (2017) 925e935

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

A sustainable application of recycled tyre crumbs as insulator in

lightweight cellular concrete
Alireza Kashani a, *, Tuan Duc Ngo a, Priyan Mendis a, Jay R. Black b,
Ailar Hajimohammadi a
a
Department of Infrastructure Engineering, University of Melbourne, Victoria, 3010, Australia
b
School of Earth Sciences, University of Melbourne, Victoria, 3010, Australia

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

Article history: End-of-life waste tyres are known to have negative economic and environmental impacts due to the
Received 27 October 2016 difficulty associated with their disposal and recycling. The use of recycled tyre crumb (RTC) as a
Received in revised form component of construction materials has emerged as a potentially sustainable solution to this envi-
20 February 2017
ronmental issue. There exists extensive research covering the use of RTC as a filler in conventional
Accepted 21 February 2017
Available online 22 February 2017
concrete. However, there are very limited studies focused on insulation properties of RTC in lightweight
cellular concrete (LCC) as a sustainable application that can add value to the final product. This study
investigates the effects of different RTC contents on compressive strength, porosity, thermal conductivity,
Keywords:
Recycled tyre crumbs
sound insulation and water permeability of LCC as an insulator. A rather homogeneous distribution of
Lightweight cellular concrete tyre crumbs within LCC structure was achieved. It has been found that LCC samples containing RTC are
Sound insulation suitable as insulators because the sound and thermal insulation are improved as rubber content is
Thermal insulation increased compared to the sample with similar density but without RTC. The insulation properties of this
Permeability product also come with the same total porosity and a significant reduction in the rate of water
Porosity permeability compared to the sample without RTC. 3D restructuring of micro-CT scan images of this
composite revealed the effects of tyre crumb on the total and local porosity of this composite. Despite the
fact that addition of tyre crumb reduced the strength, surface treatment of crumb rubber with sodium
hydroxide solution exhibited significant improvement in the compressive strength of LCC as the result of
better physical bonding with hydrated cement proven by scanning electron microscopy (SEM) images
and energy dispersive spectroscopy (EDS) analysis.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction alternative fuels and raw materials were reviewed and tyre-derived
scrape was listed as an energy-efficient alternative to fossil fuel in
Waste material production is rising at an alarming rate world- cement manufacturing (Mikul ci
c et al., 2016). However, after
wide. As a result, it is increasingly important to find sustainable reduction in the oil price in recent years, the demand for tyre-
methods to manage and recycle this substantial waste. Recycling derived fuel has been decreased substantially. On the other hand,
and reusing of the waste tyre are challenging due to the high-cost a new trend of recycling process by application of tyre-derived
recycling processes and a lack of substantial demand. Approxi- aggregates in hollow concrete has emerged as a sustainable solu-
mately 51 M equivalent passenger tyres reach their end-life each tion with increased sound insulation and good thermal resistivity
year in Australia. Only 5% of this waste is recycled locally, and the properties (Mohammed et al., 2012). Therefore, the application of
rest are exported to be used as a high-emission fuel or disposed to the recycled tyre in construction materials e.g. concrete can result
landfill which generates environmental issues from facilitating in less energy consumption of the buildings. With the recent de-
mosquito breeding grounds to being potential toxic fire hazards velopments in material technologies, an opportunity exists to
(TSA, 2017). Cleaner and sustainable cement production by using improve the uptake of recycled tyre crumb in concrete by providing
improved elasticity, ductility, damping, toughness, high sound ab-
sorption and freeze and thaw properties in spite of reduction in
* Corresponding author. stiffness, compressive and tensile strength (Segre and Joekes, 2000;
E-mail address: [email protected] (A. Kashani).

http://dx.doi.org/10.1016/j.jclepro.2017.02.154
0959-6526/© 2017 Elsevier Ltd. All rights reserved.
926 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935

Siddique and Naik, 2004; Khaloo et al., 2008; Ganjian et al., 2009; et al., 2013; Huang et al., 2013; Najim and Hall, 2013). NaOH is
Ismail and Hassan, 2016; Richardson et al., 2016). Concrete com- deemed to remove zinc stearat, an additive in rubber to prevent
posites which utilise recycled tyres as aggregates have the potential oxidation, from the surface of RTC which contributes to the poor
to reduce overall greenhouse gas emissions, provide potential adhesion between rubber and cement, thus leading to the greater
economic benefit and reduce the environmental issues associated compressive strength of rubber-treated concrete samples
with end-of-life tyres as compared to conventional concrete (Mohammadi et al., 2016).
(Meyer, 2009). The addition of RTC into conventional concrete also has the ef-
Lightweight cellular concrete offers thermal insulation and high fect of reducing the density because of the substantially lower
strength to weight ratio in addition to low cost and good freeze/ specific gravity of rubber particles than conventional fine or coarse
thaw stability (Ramamurthy et al., 2009). Low-density aggregates aggregates (Pierce and Blackwell, 2003). The significant lower
such as polystyrene and polyurethane foam granules have been density of RTC compared to conventional concrete has also been
used to replace high-density aggregates in lightweight concrete known to cause segregation during the casting and vibration of
(Laukaitis et al., 2005; Ben Fraj et al., 2010). Crumbed rubber RTC-concrete composite. This results in the movement of rubber
derived from tyre recycling have a high strength to weight ratio and particles towards the top surface of the concrete, creating a non-
can also be used as an aggregate. The densities of lightweight homogeneous compound with a significantly lower compressive
concrete differ considerably, depending largely on the void ratio strength, resulting in failure at lower stresses at the top surface of
and the density of its aggregates, but are typical between 300 kg/m3 the concrete (Ganjian et al., 2009). However, application of RTC in
and 1500 kg/m3. Lightweight concrete with lower density range has lightweight concrete has the advantage of the lower density of LCC
gained notability in the construction industry for its application in compared to the conventional concrete mixture. The density of LCC
non-load bearing structures where a desire for higher strength to can be very close to the density of rubber crumbs hence decreasing
weight ratio and easily installable products with excellent insu- the possibility of segregation.
lation properties exists. Lightweight concrete is known to provide a While there has been a wealth of research investigating the
higher degree of thermal insulation due to its porous structure properties of recycled tyre crumbs incorporated in conventional
containing air pockets with very low thermal conductivity. concrete, limited research is available with regard to the application
of RTC in lightweight cellular concrete and its insulation properties.
1.1. Literature review This paper will investigate the inclusion of RTC up to 30 weight
percentage of solid mass in LCC. Due to the suitability of RTC-LCC
The lower thermal conductivity of concrete composite with composite for a sustainable application as building insulators, an
rubber is mainly due to the lower thermal conductivity of rubber emphasis on to the insulation properties was taken into account.
(0.15e0.25 W/m.K) compared to other conventional aggregates e.g. The effect of tyre crumb on water permeability, total and local
sand (~1.5 W/m.K) (Abdel Kader et al., 2012; Mohammed et al., porosity, sound and thermal insulation properties, as well as
2012). Concrete-rubber composite is currently being used for its compressive strength, were investigated. Additional interest in the
sound insulation properties in various structures such as highway 3D analysis of porous structure using micro-CT scan imaging as well
sound barriers and sound proof flooring. Conventional concrete as the effects of chemical treatment of crumb rubber with NaOH
exhibits poor sound absorption capabilities, thus causing echoing solution to enhance its bonding properties with hydrated cement
within enclosed spaces. The addition of tyre crumb rubber into were followed.
concrete exhibits superior acoustic abilities; showing a higher
sound absorption coefficient relative to conventional concrete 2. Materials and test methods
(Holmes et al., 2014).
However, the introduction of tyre crumb into concrete mixtures 2.1. Materials
poses various challenges, most notably, the lower Young’s modulus,
stiffness and compressive strength of RTC relative to conventional The RTC was derived from recycled truck tyres with the metal
aggregates (Siddique and Naik, 2004; Khaloo et al., 2008; Najim and and thread removed and was supplied by Tyrecycle, Australia. The
Hall, 2012; Pacheco-Torgal et al., 2012). The reduction in tyre crumbs had an irregular shape with a rough surface and were a
compressive strength can be attributed to the poor adhesion be- random mixture of sizes ranging from 0.9 mm to 2.5 mm with an
tween rubber particles and cement paste in the interfacial zone estimated specific gravity of about 1000 kg/m3. The rubber content
(Pacheco-Torgal et al., 2012) and/or due to the increase in air con- in each sample was calculated as a 10, 20, 30 weight percentage of
tent in normal concrete as the rubber content proportion increases total solid mass. Ordinary Portland cement (general purpose grade
(Siddique and Naik, 2004). The latter is the result of the non-polar with the trademark of Baston, Australia) was used. The water to
nature of rubber particles and their tendency to entrap air in their cement of 0.5 was chosen, and a commercial foaming agent was
rough surfaces, thereby decreasing the unit weight of the normal used for making the pre-made foam.
concrete and hence the compressive strength. The greater the size Portland cement, water and rubber were thoroughly mixed in a
or proportion of RTC used to replace concrete aggregate, the greater bowl by hand for 30 s to form a slurry, then the bowl was placed in a
the drop in the compressive strength of the sample (Pierce and planetary mixer for a further 10 m mixing at low speed. Using a
Blackwell, 2003; Najim and Hall, 2012). Despite that, RTC can Dema foam generator machine, foam with a density of around
improve flexural toughness and damping properties up to a certain 100 kg/m3 was obtained. The generated foam was added to the
volume percentage of concrete (Khaloo et al., 2008; Najim and Hall, cement mixture and mixed by hand for 1 m. The initial amount of
2012). Improving the mechanical performance of concrete-rubber foam was about 10% of the dry cement. Extra foam was added to
composite has the potential to improve the practicality and achieve the wet density of around 600 kg/m3 for the samples with
viability of utilising of RTC in commercial applications. In order to higher contents of rubber. The dry density after curing was slightly
improve the properties of rubber in concrete, some methods have higher for some samples. After moulding, samples were not
been developed and tested primarily on conventional concrete. vibrated to avoid foam collapse but in order to fully fill the mould
Surface treatment of rubber with chemicals e.g. sodium hydroxide edges with concrete foam (especially for cubic moulds used in
(NaOH) and silane coupling agent was focused on, increasing the compression test), the moulds were gently tapped on the bench.
strength of the bonds between rubber and cement particles (Dong The tapping caused slight foam collapse and increase the density of
 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935 927

some samples. However, density was measured for individual 3.1.

samples for all the tests and is reported as an average of three Micro-CT scanning was performed with a Phoenix Nanotom m
specimens or individual samples based on the type of the tests. Six (GE Sensing & Inspection Technologies GmbH, Wunstorf, Ger-
types of LCC mixtures comprising 0 wt %, 10 wt %, 20 wt % (w/out many). Four cylindrical samples with dimensions detailed in
NaOH treatment) and 30 wt % (w/out NaOH treatment) RTC were Table 2 with various quantities of rubber (0, 10, 20 and 30 wt% of
prepared and poured into moulds and placed in sealed bags. The solids) were mounted on the micro-CT stage and positioned as
mixture proportions of samples are shown in Table 1. No fine ag- close to the x-ray source as possible to achieve 20 mm resolution on
gregates (e.g fine sand) were used in this study in order to only a small subsection (4.8
4.8 mm) at the centre of the cores.
focus on the effects of tyre rubber crumb aggregates in properties of Samples were scanned for 20 m (timing ¼ 500 ms, av ¼ 1, skip ¼ 0)
lightweight concrete. The samples were de-moulded after 48 h and at 80 kV and 375 mA, collecting 1500 x-ray projections of each
re-sealed to cure at room temperature for 7 d, 14 d and 28 d. For sample through 360 of rotation. A 0.25 mm Cu filter was placed in
NaOH treatment, the tyre crumbs were soaked in a 20 wt % NaOH the collimator on the x-ray source to prevent oversaturation of the
aqueous solution for 30 m before being removed by filtration, x-ray detector. Volume reconstruction of the micro-CT data was
washed with water, dried and added to the dry cement mixture. performed using Phoenix datos|x reconstruction software (GE
Sensing & Inspection Technologies) and data was exported as 32-
2.2. Test methods bit float volume files. This allowed for direct comparison of gray-
scale values in the histograms between data sets. A region of
In order to evaluate the compressive strength of the samples, 1,500
1,500
1,500 voxels in the centre of the sample was
the INSTRON 5569A was used in accordance with a treated version reconstructed. Volume data was processed using Avizo (FEI).
of the ASTM C170 standard. The compressive strength of LCC cubes Porosity was segmented using the same region of the 32-bit gray-
was evaluated after 7 d and 28 d of curing and the average of three scale histogram (1 to 0.0166) to produce binary images of the
samples per system was reported. The machine was set to a con- porosity in the sample. The volume fraction of segmented pore
stant displacement rate of 1 mm/m. The load bearing area of each space was then computed by x-y slice through every x-y plane
specimen was measured and recorded using digital callipers. The (1500 slices).
load was applied in order to obtain stress-strain curves for all of the The determination of water absorption of samples provides an
specimens, then the maximum load and the contact area was used understanding of the effects of RTC on the water permeability of
for calculation of compressive strength. Slices with 5 mm height LCC. The water absorption of the samples is obtained by using a
from the top and the bottom of cubic samples (slice size: treated version of the method listed in the ASTM C1754 standard.
50
50
5 mm) were cut using a band saw in order to analyse the Cylindrical samples with dimensions detailed in Table 2 were used
pores and the rubber distribution. Optical images of these slices for this test. The water absorption test was conducted on samples
were captured using a high-resolution camera. Cuboids with the after 14 d of curing. The samples were then placed in an oven at
size of 6
6
2 mm of these slices were also used for SEM images. 60  C and weighed every 2 h until there was no difference in weight
SEM images were captured using an FEI Quanta FEG 200 machine. between two consecutive weighs. The final weight of the sample
Micro-elemental analysis was conducted using an Oxford energy was recorded as the dry mass of the specimen. The sample was then
dispersive spectroscopy (EDS) instrument attached to the SEM submerged in water for 30 m, 24 h and 48 h. The samples were held
which was equipped with an X-max silicon drift detector (SDD). A submerged using a metal cage to avoid floating of the samples
standard-less quantitative EDS analysis was done only for differ- because their densities are lower than water. The samples were
entiating rubber from hydrated cement in high magnification SEM then removed from the water and weighed after wiping away
images. The samples cannot be polished because rubber would be excess water. The average results of three specimens for each type
easily removed from the surface during the polishing and small were reported.
cement dust would fill the gaps between the rubber and cement The effect of RTC content on thermal conductivity was analysed
(the area which is of great interest for this study). The samples were using a treated version of the IEEE 421-1981 standard test which
run at high vacuum to reduce the interference of carbon and oxy- utilises a data logger to record the heat dissipated from a thermal
gen elements in the air. Because of these limitations, the EDS results probe situated in the centre of a cylindrical sample. The thermal
were used only as an indication rather than accurate elemental probe features a thermocouple and a heating element for
analysis. The data were then analysed with Aztec software version measuring thermal conductivity.

Table 1
Mixture proportion of the samples.

Cement (kg) Water (kg) Foam (kg) Rubber (kg) Rubber washed with NaoH

Control Sample 0.3 0.15 0.03 0.000 No

10% RTC 0.3 0.15 0.032 0.033 No
20% RTC 0.3 0.15 0.034 0.075 No
30% RTC 0.3 0.15 0.036 0.129 No
20% RTC (NaOH treated) 0.3 0.15 0.034 0.075 Yes
30% RTC (NaOH treated) 0.3 0.15 0.036 0.129 Yes

Table 2
Shapes and dimensions of samples for different test methods.

Test Sample shape Dimensions (mm)

Compressive strength and UPV Cubes 50
50 x 50

Thermal conductivity and water absorption Cylinder (with 2 mm diameter hole in the center for thermal conductivity samples only) Radius: 25
Length: 110
928 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935

The ultrasonic pulse velocity (UPV) test provides insight into the treated RTC showed a significant improvement in compressive
effects of RTC content on sound insulation of LCC. The UPV strength relative to untreated specimens. As shown in Fig. 1, the
measured by Pundit PL-200 instrument (sourced by Proceq) for samples with 30 wt % NaOH-treated RTC have about 40% higher
evaluation of the velocity of sound transmission through the compressive strength compared to the samples containing non-
specimen. A water-based gel was applied to the transducers and treated tyre crumb. This method shows a simple and effective
then they were placed on the opposite sides of the cubes. An ul- treatment which leads to significant improvement in compressive
trasound pulse voltage of 400 V was applied. The velocity of ul- strength. The results of scanning electron microscopy and micro-
trasonic pulses was recorded once the reading stabilised. Table 2 elemental analysis prove the enhanced bonding between rubbers
provides descriptions of the shapes and dimensions of samples and cement hydration products, which will be explained in more
for each test methods. details in Section 3.2. Fig. 2 shows an increase in compressive
strength from 7 d to 28 d of curing for LCC containing 20 wt % RTC
with and without treatment by sodium hydroxide solution. As ex-
3. Results and findings
pected, longer curing time in a confined environment will increase
the strength because of the continuous process of cement
3.1. The effects of rubber contents and its surface treatment on
hydration.
compressive strength of LCC

The results of the 7 d compressive strength show a trend of 3.2. Distribution of pores and tyre crumbs within LCC
decreasing compressive strength as rubber content is increased.
Fig. 1 shows around 10% reduction in compressive strength A relatively homogenous distribution of tyre rubber crumbs was
resulting from the addition of 10 wt % of RTC, however, increasing observed inside all LCC samples as shown in Fig. 3. Agglomeration
the amount of tyre crumb to 30 wt % resulted in about half of the of tyre crumbs was very limited which shows effective mixing
compressive strength compared to the control sample. This is in resulted in the homogenous distribution of tyre crumbs. Settling of
agreement with the results from previous studies conducted using tyre crumbs before setting of the cement can result in a gradient
recycled tyre crumb in conventional concrete (Pierce and Blackwell, content of RTC in the samples, thus mechanical and thermal per-
2003; Siddique and Naik, 2004; Najim and Hall, 2012; Pacheco- formance can vary in different directions. However, a very limited
Torgal et al., 2012). However, for non-load bearing applications of variation in rubber content was observed when top and bottom
lightweight concrete e.g insulators, compressive strength is less cuts were compared in any of the samples in this study. Fig. 3b and c
important. A variation between densities of the LCC samples with shows the top and bottom cut of LCC with 10 wt % RTC with
different rubber content can directly affect the strength of the negligible difference in RTC content. It can be concluded that
samples as well as other microstructural and thermal properties. limited settlement has happened before setting due to an excellent
Higher density generally results in higher strength in lightweight containment of rubber particles within cement binder. The net
cellular concrete (Narayanan and Ramamurthy, 2000). The aim was forces on the tyre crumbs pushed against the gravity force and held
to control and eliminate the effect of density for better comparison the rubber crumbs in place until the setting of the lightweight
between different samples. The wet density was targeted at around concrete.
600 kg/m3 for all samples, however, the dry density of the samples Fig. 4 shows the 3D visualisation of samples with and without
after curing was higher and varied between different samples. The tyre crumbs based on micro-CT scan data. Black, white and grey
higher density compared to wet density is the result of foam colours were chosen to better represent tyre crumbs, pores and
instability (after tapping the cubic moulds in order to fill the edges) hydrated cement structure respectively (These are not the genuine
causing the release of the entrained air before it is encapsulation colour of these materials in the micro-CT scan images). A combi-
during setting of cement. The average densities of three samples nation of closed and interconnected pores was seen for LCC with
with their average strength is shown in Fig. 1. and without tyre crumbs. The calculated total porosity using micro-
Additionally, it was found that the specimens containing NaOH- CT scan analysis will be further discussed in Sections 3.3 and 4.
A wide range of pore size distribution from a few hundred
micrometres up to around 1 mm was shown in the SEM images of
LCC with and without tyre crumbs (Fig. 5). SEM images have been
effectively used in order to examine the bonding behaviour

Fig. 1. The effects of wt % RTC on compressive strength of LCC (the columns with
dotted pattern show the increase of strength after surface treatment of rubber with Fig. 2. Effect of curing age on compressive strength of LCC containing 20 wt % RTC w/o
NaOH solution). treatment by sodium hydroxide solution.
 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935 929

Fig. 3. Distribution of RTC within LCC sample. Inside the slices with the dimension of 50
50 mm are shown (a. control; b. 10 wt % (top cut); c. 10 wt % (bottom cut); d. 20 wt %; e.
20 wt % NaOH-treated; f. 30 wt %; g. 30 wt % NaOH-treated; all photos expect (c) are top cuts).

Control 10% RTC

20% RTC 30% RTC

Fig. 4. 3D visualisation micro-CT scan images of control LCC and with 10, 20 and 30 wt % rubber (dimension: 4.8
4.8
4.8 mm) - black spots: tyre crumbs, white spots: pores and
light grey spots: hydrated cement.
930 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935

Fig. 5. SEM images of LCC (a. control sample and b. with 30% RTC).

between rubber and cement (Segre and Joekes, 2000; Najim and analysis was done. The results indicate less gap between rubber and
Hall, 2013; Karakurt, 2015). Lack of bonding has been shown by hydrated cement in samples that rubber crumb was washed with
the gaps between rubber and cement, however, in the aerated NaOH solution before used in LCC. Five different sections of these
concrete system, it is hard to distinguish this gap between tyre images were analysed by EDS to distinguish the presence of major
crumbs and cement binder at lower magnification because the elements in each section. The areas with high carbon content
main components of the LCC are the air pores. Despite this fact, a represent rubber particles that are partially covered with cement
very thin gap between rubber and cement binder was observed in hydration products (sections a and d). A distinguishable gap be-
some areas e.g. the red circle in Fig. 5b shows this gap. tween rubber (section a) and cement binder (section b) are
Fig. 6 shows the SEM images of LCC with 30 wt % rubber but observed for LCC with untreated rubber (Fig. 6a), however, NaOH-
higher magnification. The image on the left has un-treated rubber treated rubber shows better bonding with hydrated cement as
and the image on the right has NaOH-treated rubber. In order to observed in Fig. 6b between hydrated cement (section c and e) and
distinguish the rubber from hydrated cement in the SEM images rubber (section d). The better bonding is deemed to be the main
with higher magnification, a standard-less quantitative EDS reason for the better compressive strength of NaOH-treated RTC as

(a) (b)

Section (d)
Section (a) Section (c)

Section (e)
Section (b)

Fig. 6. SEM images of LCC (a. 30% RTC and b. 30% RTC-NaOH) with micro-elemental analysis of the different sections.
 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935 931

was shown in Fig. 1. The results of SEM and EDS of tyre rubber of RTC. In these images, black, dark grey and grey spots represent
particles in conventional concrete by Pelisser et al. (2011) showed pores, tyre crumbs and hydrated cement, respectively. A combi-
that there is a gap between rubber tyre crumbs, identified with nation of closed and interconnected pores (black areas) with
high carbon content, and cement binder. This gap was reduced by various sizes is also observed in these images. The percentage of the
surface treatment of rubber with NaOH solution. The better total area of the black spots (pores) divided by the area of each slice
bonding between cement and NaOH-treated rubber evidenced by (4.8
4.8 mm) gives the porosity for each slice. Fig. 8 shows the
SEM images was also shown in a few other studies of rubber- porosity of each of 1500 slices for all the samples. There is only a
cement composites (Segre and Joekes, 2000; Najim and Hall, slight variation in the porosity along the height of all the samples,
2013; Karakurt, 2015). which is a sign of a homogenous distribution of pores (Fig. 8). Total
porosity is calculated as the average porosity of all the slices for
3.3. Total porosity calculation by 3D processing of micro-CT scan each sample. The results showed porosity of the samples with
images different rubber contents but the same density has not changed.
However, it must be noted that by the addition of tyre crumb
A cubic region of interest (4.8 mm) within four cylindrical rubber (considered as zero porosity sections of the sample), the
samples with dimensions detailed in Table 2 with various quanti- remaining area of the sample should have higher porosity in order
ties of rubber (0, 10, 20 and 30%) and low-density variation to maintain the constant total porosity as RTC increased. This
(600 ± 2 kg/m3) was used for the micro-CT scan analysis. The water means higher porosity for lightweight concrete which surrounds
absorption was tested on the same cylindrical specimens. Fig. 7 tyre crumbs compared to the control LCC by increasing RTC. The
shows micro-CT scan images (one x-y plane slice of the total of relationship between porosity, strength and insulation properties
1500 slices) for control LCC and the sample with 10, 20 and 30 wt % will be discussed further in Section 4.

Control 10% RTC

RTC

20% RTC 30% RTC

Fig. 7. Micro-CT scan images of control LCC and with 10, 20 and 30 wt % rubber - black spots: pores, dark grey spots: tyre crumbs and light grey spots: hydrated cement.
932 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935

50

40

Porosity (%)
30

20

10

Control 10% RTC 20% RTC 30% RTC

0
0 250 500 750 1,000 1,250 1,500
Slice number
Fig. 8. Porosity distribution through 1500 slices of micro-CT data (x-y planes in 20 mm steps) along z-axis (cylinder’s height) of LCC with and without tyre crumbs.

3.4. Water absorption route is diffusion through the microscopic porous structure by
capillary suction (sorptivity) within the hydrated cement binder
The density of the samples was about 600 kg/m3 and average of itself (Narayanan and Ramamurthy, 2000). However, it has also
three samples were tested. One cylinder from each type of samples been shown that crumb rubber decreases the sorptivity of cement
with the density of 600 ± 2 kg/m3 was used for measuring porosity binder matrix by reducing the capillary pressure (Bravo and de
using micro-CT scan prior to the water absorption test. As discussed Brito, 2012; Meddah et al., 2014; Onuaguluchi, 2015). The hydro-
in Section 3.3, the porosity of the samples with different rubber phobic nature of rubber can also be the reason for lower water
contents but the same density is the same. A decrease in water absorption of the mixtures with the high content of tyre crumbs.
absorption as rubber content is increased was observed as shown in
Fig. 9. About 55% water uptake per mass of control samples after 3.5. Thermal conductivity
immersion into the water was observed. This massive water-uptake
substantially reduces by increasing rubber content, especially for Variation in thermal conductivity with respect to the RTC con-
20 wt % and 30 wt % of RTC with about half of the increase in water tent is presented in Fig. 10. Samples with densities of 615 ± 10 kg/
absorption compared to the control sample in spite of having the m3 were used to reduce the effect of density on thermal conduc-
same porosity. Also, longer immersion time in the water increased tivity between different samples. The thermal conductivity of the
water absorption for samples with higher contents of rubber. There control samples is about 0.2 W/m.K compared to the range of
are two main routes for water transfer from the surface to the 0.15e0.25 W/m.K reported for recycled tyre crumb rubber (Abdel
centre of samples. One route is via diffusion of water through Kader et al., 2012). It can be concluded that tyre crumb and LCC
interconnected macroscopic pores (permeability) and the other with 615 ± 10 kg/m3 density has very similar thermal conductivity

60
0.5 h 24 h 48 h

50
Water absorption (%)

40

30

20

10

0
control 10%RTC 20%RTC 30%RTC
Fig. 9. Water absorption (%) after immersion into water for 0.5, 24 and 48 h versus rubber content.
 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935 933

0.25 rubber content is able to show a promising reduction in velocity of

sound transmittance.
Thermal conductivity (W/mK)

0.2
4. Discussion on the effects of RTC on the relationship
0.15 between porosity, water permeability, strength and insulation
properties
0.1
For compressive strength results, there has been some variation
in densities because of limitations in sample making as described in
0.05 Section 2.1. Generally, the compressive strength of porous concrete
increases by lowering the porosity (or increasing density) (Akthar
0 and Evans, 2010). In spite of the variation between densities of
Control 10%RTC 20%RTC 30%RTC the samples for compressive strength test, the reduction of strength
by addition of rubber was less affected by density. The samples with
Fig. 10. Thermal conductivity versus rubber content for lightweight cellular concrete
samples. 10% and 20% RTC have slightly higher densities compared to the
control sample but lower strength. The lack of good bonding be-
tween cement and tyre crumb is one reason for strength reduction.
properties. The slight decrease in thermal conductivity as rubber It was shown in Fig. 1 that by improving the bonding after washing
contents increase can be attributed to the fact that the porosity of the tyre crumb with NaOH solution, the strength increased sub-
the lightweight concrete composite with RTC in the sections sur- stantially. In addition, it must be noted that by the addition of tyre
rounding tyre crumbs is higher than the control samples (as dis- crumb rubber (considered as zero porosity sections of the sample),
cussed in Section 3.3). In other words, the porosity of the the remaining area of the sample should have higher porosity in
lightweight concrete in the sections around the tyre crumbs is order to maintain the total porosity constant. This means higher
higher than the control sample (it must be noted that the total porosity for lightweight concrete surrounding tyre crumbs
porosity including tyre crumbs is the same as the control sample as compared to the control LCC. Fig. 12 shows the porosity of the
shown in Fig. 8). Higher porosity means the existence of many air composite which tyre crumbs considered as areas of zero porosity
pockets with very low thermal conductivity within the structure in the calculation of total porosity. It shows unchanged porosity by
and around the tyre crumbs which itself has low thermal increasing the RTC. However, calculation of the porosity after
conductivity. exclusion of the tyre crumb volume shows increased porosity of
LCC surrounding the tyre crumbs. Therefore the combination of
increasing porosity of LCC around tyre crumbs and lack of good
3.6. Sound transmittance binding between rubber and hydrated cement resulted in sub-
stantial strength reduction for the samples with 30 wt% rubber
As shown in Fig. 11, the velocity of the ultrasonic sound pulse compared to the control sample.
massively decreases by 65% as RTC content is increased from zero to With regards to water absorption, longer immersion time in the
30 wt % of the system. Although the frequency of ultrasound is water increased water absorption for samples with higher contents
higher than audible sound, the concept of sound transmission of rubber. This implies that for the sample without rubber or less
through the material is the same. The UPV of conventional concrete rubber, water saturation in the pores happen quickly. On the other
is between 3000 and 5000 m/s, the lower values come with using of hand, a higher amount of rubber aggregates slows down water
lightweight aggregates in the concrete mixture (Bogas et al., 2013). penetration into the porous structure because of their water
The UPV value of LCC with the density of 615 ± 10 kg/m3 and repellency effects. Therefore the longer the period of immersion
without rubber is less than half of conventional concrete and is caused higher water absorption. It is possible that samples with
around 1340 m/s (Fig. 11). The UPV of LCC with 30 wt % rubber is as higher contents of rubber and the same porosity as the control
low as 400 m/s. The UPV test showed that LCC with increased sample can reach the same level of water absorption in very long

1,600

1,200
UPV (m/s)

800

400

0
Control 10%RTC 20%RTC (w/o NaOH) 30%RTC (w/o NaOH)
Fig. 11. Ultrasonic pulse velocity of LCC samples with different contents of RTC.
934 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935

50

45

40

35
Porosity (%)
30

25

20

15

10

0
Control 10% RTC 20% RTC 30% RTC

Porosity of LCC (RTC included) Porosity of LCC (RTC excluded)

Fig. 12. Porosity of the LCC samples with different rubber contents (black columns: rubber crumbs are included in total volume as areas of zero porosity; columns with dotted
pattern: rubber crumbs are excluded from volume and the porosity of the remaining is calculated).

time periods. However, after 48 h the water absorption of the analysis showed that in comparison with available lightweight
sample with 30 wt % of RTC was considerably lower compared to insulating panels, the proposed rubberized lightweight concrete
the control sample. with 30 wt % rubber crumbs performs extremely well. This product
Increasing the content of recycled tyre crumbs is found to has shown excellent thermal and sound insulation properties, as
improve the thermal insulation properties compared to the light- well as lower water-permeability for non-load bearing building
weight concrete with similar densities but without RTC. This result insulators.
has implications for the practical application of the product, sup-
porting its use where there is a desire for low thermal conductivity 5. Conclusions
but not high total porosity and permeability. These applications
include, but are not limited to, auditoriums, intertenancy walls in Excellent sound and thermal insulation with very low water
residential buildings, plant rooms, offices and hospital partitions. absorption and total porosity have been achieved by inclusion of
Decreasing heat transfer within these types of infrastructure may recycled tyre crumb in lightweight cellular concrete. Implications
produce significant energy and subsequent cost reductions. for a sustainable application of recycled tyre crumb in insulating
The main reason for excellent sound insulation properties of LCC LCC panels to be used in auditoriums and intertenancy walls in
with tyre crumbs is because of the lower velocity of sound in rubber residential buildings, plant rooms, offices, warehouses and hospi-
compared to hydrated cement (Holmes et al., 2014) and also an tals are suggested. Therefore, this new application of recycled tyre
increase in porosity of lightweight concrete around tyre crumbs as not only can reduce the stockpiles of wasted tyres worldwide but
RTC increased. The transmittance of sound in air is considerably also can help with regards to better insulation of buildings, which
slower compared to many solid materials. However, sound velocity finally results in less energy usage, less greenhouse gas emission
in vulcanised rubber which constitutes the main part of the tyre and cleaner production.
crumbs can be lower than air because of its elasticity properties Treating the RTC with NaOH solution causes improvement in
(Long, 2014). The results indicate that this product is potentially compressive strength via better bonding between cement and RTC
suitable for structures requiring sound insulation such as in inter- as evidenced by SEM, EDS and micro-CT scan images. However, a
nal walls within auditorium and warehouses. better understanding of surface chemistry and morphology in or-
The results from this study provide strong evidence that the use der to explain the causes of improvement in bonding strength
of waste tyre in the form of crumb rubber in lightweight concrete is should be targeted in the future studies. The distribution of rubber
an effective solution for manufacturing building insulators. The particles was rather homogenous in every direction of samples
results showed enhanced water, thermal and sound insulation, at a with limited agglomeration or settling. The porosity of samples
significant reduction in weight. In Australia, the market for light- with and without tyre crumbs was visualised and measured by the
weight building insulation panels mainly consists of unsustainable 3D restructuring of micro-CT scan images. The total porosity of
polymer-based foams such as expanded polystyrene (EPS), poly- samples with tyre crumbs is the same as the control sample with
urethane (PUR) and polyisocynature (PIR). The use of polymer- the same density. However, the porosity of lightweight concrete in
based foam, in spite of being very light and an excellent insulator, local areas around the tyre crumbs is slightly higher than the
has rapidly declined worldwide because of sustainability, control sample which causes better insulation but lower mechan-
combustibility and recyclability issues. There are various grades of ical strength. An LCC composite with 30 wt % crumb rubber can
polymer-based foams, some manufactured with special additives to offer excellent sound and thermal insulation properties with less
add fire-retardant properties. However, these additives would only water permeability compared to the lightweight concrete without
postpone the fire, and hazardous smoke is still the main problem. tyre crumbs but with the same density and total porosity.
The rubberized lightweight concrete investigated in this study as a The current study contributes to a better understanding of the
filler for insulating panels is still at its infancy stage. However, the effects of recycled tyre crumbs on insulation properties of
 A. Kashani et al. / Journal of Cleaner Production 149 (2017) 925e935 935

lightweight cellular concrete. This application is deemed as a sus- Karakurt, C., 2015. Microstructure properties of waste tire rubber composites: an
overview. J. Material Cycles Waste Manag. 17 (3), 422e433.
tainable solution in order to ultimately increase the uptake of
Khaloo, A.R., Dehestani, M., Rahmatabadi, P., 2008. Mechanical properties of con-
wasted tyre crumb in building components as an excellent insu- crete containing a high volume of tireerubber particles. Waste Manag. 28 (12),
lator. The results of this research can be used as a benchmark for 2472e2482.

Laukaitis, A., Zurauskas, R., Kerien, J., 2005. The effect of foam polystyrene granules
future studies in this area to accommodate a cleaner and more
on cement composite properties. Cem. Concr. Compos. 27 (1), 41e47.
sustainable production of building insulators. Long, M., 2014. Architectural Acoustics, second ed. Academic Press, Waltham, MA,
p. 2014.
Acknowledgements Meddah, A., Beddar, M., Bali, A., 2014. Use of shredded rubber tire aggregates for
roller compacted concrete pavement. J. Clean. Prod. 72, 187e192.
Meyer, C., 2009. The greening of the concrete industry. Cem. Concr. Compos. 31 (8),
This work was supported by Tyre Stewardship Australia (TSA) as 601e605.
part of an R&D program to develop new end-uses for recycled tyre Mikul ci
c, H., Klemes, J.J., Vujanovi c, M., Urbaniec, K., Dui c, N., 2016. Reducing
greenhouse gasses emissions by fostering the deployment of alternative raw
and in part by Surrey Engineering Pty Ltd as the industry partner. materials and energy sources in the cleaner cement manufacturing process.
The authors would like to thank Tyrecycle for the provision of J. Clean. Prod. 136 (Part B), 119e132.
recycled tyre crumbs. The authors acknowledge the support of the Mohammadi, I., Khabbaz, H., Vessalas, K., 2016. Enhancing mechanical performance
of rubberised concrete pavements with sodium hydroxide treatment. Mater.
Trace Analysis for Chemical, Earth and Environmental Sciences Struct. 49 (3), 813e827.
(TrACEES) platform from the Melbourne Collaborative Infrastruc- Mohammed, B.S., Anwar Hossain, K.M., Eng Swee, J.T., Wong, G., Abdullahi, M., 2012.
ture Research Program at the University of Melbourne. The authors Properties of crumb rubber hollow concrete block. J. Clean. Prod. 23 (1), 57e67.
Najim, K.B., Hall, M.R., 2012. Mechanical and dynamic properties of self-compacting
also acknowledge Derek Yong, Xiayang Cui, Thomas Alemis and
crumb rubber modified concrete. Constr. Build. Mater. 27 (1), 521e530.
Liam Ryan for their contribution in the preparation of samples and Najim, K.B., Hall, M.R., 2013. Crumb rubber aggregate coatings/pre-treatments and
execution of some of the tests. their effects on interfacial bonding, air entrapment and fracture toughness in
self-compacting rubberised concrete (SCRC). Mater. Struct. 46 (12), 2029e2043.
Narayanan, N., Ramamurthy, K., 2000. Structure and properties of aerated concrete:
References a review. Cem. Concr. Compos. 22 (5), 321e329.
Onuaguluchi, O., 2015. Effects of surface pre-coating and silica fume on crumb
Abdel Kader, M.M., Abdel-wehab, S.M., Helal, M.A., Hassan, H.H., 2012. Evaluation of rubber-cement matrix interface and cement mortar properties. J. Clean. Prod.
thermal insulation and mechanical properties of waste rubber/natural rubber 104, 339e345.
composite. HBRC J. 8 (1), 69e74. Pacheco-Torgal, F., Ding, Y., Jalali, S., 2012. Properties and durability of concrete
Akthar, F.K., Evans, J.R.G., 2010. High porosity (>90%) cementitious foams. Cem. containing polymeric wastes (tyre rubber and polyethylene terephthalate bot-
Concr. Res. 40 (2), 352e358. tles): an overview. Constr. Build. Mater. 30, 714e724.
Ben Fraj, A., Kismi, M., Mounanga, P., 2010. Valorization of coarse rigid polyurethane Pelisser, F., Zavarise, N., Longo, T.A., Bernardin, A.M., 2011. Concrete made with
foam waste in lightweight aggregate concrete. Constr. Build. Mater. 24 (6), recycled tire rubber: effect of alkaline activation and silica fume addition. J.
1069e1077. Clean. Prod. 19 (6e7), 757e763.
Bogas, J.A., Gomes, M.G., Gomes, A., 2013. Compressive strength evaluation of Pierce, C.E., Blackwell, M.C., 2003. Potential of scrap tire rubber as lightweight
structural lightweight concrete by non-destructive ultrasonic pulse velocity aggregate in flowable fill. Waste Manag. 23 (3), 197e208.
method. Ultrasonics 53 (5), 962e972. Ramamurthy, K., Kunhanandan Nambiar, E.K., Indu Siva Ranjani, G., 2009.
Bravo, M., de Brito, J., 2012. Concrete made with used tyre aggregate: durability- A classification of studies on properties of foam concrete. Cem. Concr. Compos.
related performance. J. Clean. Prod. 25, 42e50. 31 (6), 388e396.
Dong, Q., Huang, B., Shu, X., 2013. Rubber modified concrete improved by chemi- Richardson, A., Coventry, K., Edmondson, V., Dias, E., 2016. Crumb rubber used in
cally active coating and silane coupling agent. Constr. Build. Mater. 48, 116e123. concrete to provide freezeethaw protection (optimal particle size). J. Clean.
Ganjian, E., Khorami, M., Maghsoudi, A.A., 2009. Scrap-tyre-rubber replacement for Prod. 112 (Part 1), 599e606.
aggregate and filler in concrete. Constr. Build. Mater. 23 (5), 1828e1836. Segre, N., Joekes, I., 2000. Use of tire rubber particles as addition to cement paste.
Holmes, N., Browne, A., Montague, C., 2014. Acoustic properties of concrete panels Cem. Concr. Res. 30 (9), 1421e1425.
with crumb rubber as a fine aggregate replacement. Constr. Build. Mater. 73, Siddique, R., Naik, T.R., 2004. Properties of concrete containing scrap-tire rubber e
195e204. an overview. Waste Manag. 24 (6), 563e569.
Huang, B., Shu, X., Cao, J., 2013. A two-staged surface treatment to improve prop- TSA, 2017. Tyre Product Stewardship Scheme, Tyre Stewardship Australia. Retrieved
erties of rubber modified cement composites. Constr. Build. Mater. 40, 270e274. 20.02.2017, from. www.tyrestewardship.org.au/about-tsa/tyre-product-
Ismail, M.K., Hassan, A.A.A., 2016. Use of metakaolin on enhancing the mechanical stewardship-scheme.
properties of self-consolidating concrete containing high percentages of crumb
rubber. J. Clean. Prod. 125, 282e295.