Construction and Building Materials 302 (2021) 124190

Contents lists available at ScienceDirect

Construction and Building Materials

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

Mechanical, physical and microstructural properties of a mortar with

melted plastic waste binder
Moussa Thiam , Mamadou Fall *
Department of Civil Engineering, University of Ottawa, Ottawa, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: The feasibility of using plastic waste as a binding material to develop a new construction material (mortar with
Plastic waste plastic waste binder (MPB)) is presented in this paper. Two types of plastic: high density polyethylene (HDPE)
Recycling and low density polyethylene (LDPE), are combined in three blend (weight) ratios (HDPE/ LDPE of 40/60, 50/
Ecological mortar
50, 60/40) and mixed with sand in four plastic contents (45%, 50%, 60%, and 65%). First, the plastic material is
Green materials
Cement
melted in an oven. Then, sand and the melted plastic are thoroughly mixed under a controlled temperature into
an homogenous mixture. The mixture is then poured into standardized molds and cured for different periods of
time. Mechanical, physical and water absorption tests are performed on the MPB samples after different curing
times to assess the influence of the change in the plastic content and the blend ratio (HDPE/LDPE or H/L) on
their mechanical and physical properties. Moreover, microstructural analyses and tests are performed on some of
the samples for a better understanding of the behavior of this MPB at its microstructure level. The results show
that samples with plastic contents of 50% and 60% and a blend ratio of 50/50 have the highest unconfined
compressive strength. The same range of plastic content is found to provide the MPB with the highest splitting
tensile strength when the H/L blend ratio is equal to 40/60 or 60/40. A relationship between the splitting tensile
strength (ft) and the compressive strength is proposed. It is also found that the MPB is more ductile with greater
plasticity than conventional mortar materials that use Portland cement. The plastic content and H/L ratio have
significant effects on the water absorption, density and microstructure of the mortar. The microstructure of the
MPB plays an important role in the development of its strength. This new material has good potential for use as a
construction material to recycle plastic waste and reduce the impact of plastic on the environment.

1. Introduction diseases [4]. Incineration of plastic waste leads to air pollution, in­
creases the risk of the expansion of carcinogenic diseases, and contrib­
The current material consumption habits globally have led to utes to the deterioration of air quality close to the incineration plants
massive waste production in local regions [1]. According to the esti­ [5]. Moreover, discarding plastic waste in aquatic environments greatly
mations from the World Bank, 174 and 289 million tons of solid waste damages fauna and flora [6] with harmful consequences on biodiversity
were produced in Sub-Saharan Africa and North America in 2016, and fishery and potential risk to food security and human health [7].
respectively, which are projected to increase by 54.6% and 18.3% About four to twelve million tons of plastic are disposed in the oceans
respectively by 2030 [2]. annually [8,9] with an estimate of 150 million tons of accumulated
Plastic, an invention of the 20th century that is widely used in con­ waste [10–12].
sumer goods today, is contained in many waste products [1]. After its Considering the above issues and the need to find an ecological and
useful lifetime, plastic can remain as intact waste for over 100 years sustainable solution to manage plastic waste and its related problems,
without any degradation, thus causing substantial environmental, so­ some studies have been conducted in civil engineering to recycle plastic
cial, and health related problems [3]. For instance, disposal of plastic into a cementitious material, much like cement and mortar. Numerous
waste in landfills can lead to soil and water pollution as well as blockage experimental studies [e.g., 3,13–16] have been carried out to evaluate
of the city’s drainage systems, which in turn, can result in proliferation the properties of fresh and/or hardened cementitious materials (con­
of mosquitoes and other insects, thus contributing to the spread of crete and mortar) that contain different types of plastic wastes for use as

* Corresponding author at: Department of Civil Engineering, University of Ottawa, 161 Colonel By, Ottawa, Ontario K1N 6N5, Canada.
E-mail address: [email protected] (M. Fall).

https://doi.org/10.1016/j.conbuildmat.2021.124190

Available online 14 July 2021

0950-0618/© 2021 Elsevier Ltd. All rights reserved.
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

aggregate, filler or fiber with some success. For example, Bayasi and In this paper, the experimental program and results of a multiple year
Zeng [13] reported that the air entrainment in fresh concrete increases research project on recycling plastic for use in construction material are
with the inclusion of polypropylene fibers with a fiber volume over presented with the following objectives: (i) to create a new construction
0.3%. Furthermore, independent of the size, type (polystyrene, polyvi­ material called mortar with melted plastic waste binder (MPB) by
nyl chloride (PVC), polyethylene terephthalate (PET), polycarbonate mixing sand with melted plastic as the only binding material; (ii) to
(PC), fine plastic aggregates, etc.) and quantity of plastic as a substitute assess the strength of MPB made with two types of plastics (low density
aggregate, several authors have arrived at the conclusion that the polyethylene (LDPE) and high density polyethylene (HDPE)) with
introduction of plastic aggregates in concrete would generally reduce its different blend (weight) ratios (H/L) and different amounts of plastic;
fresh and dry densities [15,16,17–20]. The density of the concrete is less (iii) to study the deformation (stress–strain) behavior of the MPB; and
than 2000 kg/m3, which is considered to be the minimum dry density (iv) to investigate the microstructural and physical properties of MPB.
required for structural lightweight concrete according to the RILEM LC2
classification [21]. Other studies [e.g., 16,19,22,23] show that the 2. Experimental program
slump of fresh concrete that contains several types of plastic aggregates
is lower than that of conventional concrete mix. Moreover, most of the 2.1. Materials used
studies on the strength of concrete and mortar with plastic aggregates
[1,15,16,19,22,24–27] find a reduction in the compressive strength due 2.1.1. Sand aggregates
to the addition of plastic waste as a partial substitute of fine aggregates. Natural sand (Fig. 1) from Niger River collected from the banks of
For example, Ismail and Al-Hashmi [16] reported that the compressive Kalaban-coro in Bamako, Mali, is used in this study. A sieve analysis was
strength of the concrete with 10%, 15% and 20% of fine natural ag­ carried out on the sand in accordance with ASTM C136 / C136M stan­
gregates substituted with PET aggregates are lower than the compres­ dards [29] to determine the grain size distribution. The results of the
sive strength of concrete that contains only the natural aggregates but sieve analysis indicate that the grain size distribution of the sand can be
higher than the minimum compressive strength required for structural described as follows: d10 (effective diameter) = 0.25 mm, d30 = 0.32
concrete, which is 17.24 MPa [16]. However, with 5% and 20% of mm, d60 = 0.85 mm, Cu (coefficient of uniformity) = 3.4 and Cc (coef­
plastic aggregates, the compressive strength shows a sharp decrease of ficient of curvature) = 0.5.
23% and up to 72% of the original strength respectively [22]. Similar
reduction, but with a smaller effect, is found in both the splitting tensile 2.1.2. Plastic materials
and flexural strengths of plastic aggregate samples [22]. However, the Two types of plastic wastes were used, HDPE and LDPE. The plastic
splitting failure of the concrete samples is more ductile with more plastic was collected from the Medina-Coura waste transit depot in Bamako,
aggregates and does not exhibit the typical brittle failure observed in Mali. Fig. 2 shows samples of the plastic waste used. HDPE and LDPE
conventional concrete with natural aggregates. Choi et al. [17] inves­ represent the most frequent types of polyethylene. The latter is one of
tigated the effects of PET bottles as waste plastic lightweight aggregate the most extensively used thermoplastics in the world. This type of
(WPLA) on the modulus of elasticity of concrete. Water–cement ratios of plastic can be found in almost everything from grocery bags to shampoo
45%, 49%, and 53% with WPLA contents of 0%, 25%, 50%, and 75% by bottles.
volume of fine aggregates were studied. The modulus of elasticity of the
concrete is found to decrease with an increase in the amount of PET 2.2. Sample preparation and curing
aggregates. In addition, Marzouk et al. [20] observed that substituting
conventional aggregates with plastic aggregate decreases the rate of 2.2.1. Mortar with melted plastic
water absorption. The sorptivity, that is, the ability of concrete to absorb Natural river sand, as the only aggregate, was mixed with different
water, is found to further decrease with 50% of plastic aggregate. In proportions of molten HDPE and LDPE plastic for the preparation of the
conclusion, they showed that concrete with plastic aggregates is more MPB. The sand was oven dried at 105 ◦ C and passed through a 2 mm
durable when exposed to aggressive solutions. sieve. The plastic was cut in small pieces (less than 6 cm in length) to fit
From this literature review, it is obvious that there is a lack of studies into the oven. Different proportions (by weight ratio) of HDPE and LDPE
that use plastic wastes as the only binding agent. In addition, most of the plastics were melted in the oven at an average temperature of approx­
previous research studies have not explored the use of low-density imately 250 ◦ C; see Table 1. The melting time took 20 to 45 min
polyethylene bags or plastic films because they are difficult to recycle. depending on the type and amount of plastic. Then, the melted plastic,
It is also challenging to find an equivalent grain size that corresponds to which constitutes the plastic paste binding material, was thoroughly
the grain sizes of sand or gravel [28]. This research work will address the mixed into a homogenous liquid paste and the required amount of sand
knowledge gap in this area. was added to the melted plastic in a concrete mixer at a temperature of

Fig. 1. (a) Sampling site of the sand in Kalaban - Coro on the banks of the Niger River in Bamako, Mali and (b) image of the sand sample used in the experiments.

2
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

Fig. 2. Samples of (a) HDPE and (b) LDPE plastics wastes used in testing program.

about 100 ◦ C. Mixing continued until a homogeneous mixture was ob­

Table 1
tained. The average mixing time was 15 min. The temperature change of
Mix proportion and curing time of the MPB with melted HDPE and LDPE
the hot MPB during mixing was insignificant. Subsequently, the hot
plastics.
liquid MPB was poured into metallic cylinders of 10 cm in diameter and
Sample Plastic Plastic HDPE / Granular Curing
20 cm in height, or 5 cm in diameter and 10 cm in height. Note that prior
Designation Binder Content LDPE material time
(%) (days) to pouring the liquid (fresh) mortar, the inner surface of the cylindrical
molds was coated with lubricant to facilitate extraction of the sample.
45% P – H/L HDPE - 45 50/50 Sand 1, 3, 7, 28
50/50 LDPE
When the molds were filled, a manual hand press was used to compact
50% P – H/L HDPE - 50 60/40 Sand 1, 3, 7, 28 the fresh mortar in the mold and keep it under pressure to limit sagging
60/40 LDPE of the sample. Finally, the samples were extracted after half an hour.
50% P – H/L HDPE - 50 50/50 Sand 1, 3, 7, 28 Fig. 3 shows the ecological mortar samples before and after extraction
50/50 LDPE
from the mold. The samples were then cured at room temperature for
50% P – H/L HDPE - 50 40/60 Sand 1, 3, 7, 28
40/60 LDPE periods of 1, 3, 7, and 28 days.
60% P – H/L HDPE - 50 60/40 Sand 1, 3, 7, 28 The mix proportions of the MPB and curing times are shown in
60/40 LDPE Table 1. The rationale for the recipe or mix proportions considered is
60% P – H/L HDPE - 50 50/50 Sand 1, 3, 7, 28 based on preliminary tests conducted to assess the MPB performance.
50/50 LDPE
The preliminary tests used equal amounts of HDPE and LDPE plastics.
60% P – H/L HDPE - 50 40/60 Sand 1, 3, 7, 28
40/60 LDPE The percentage of the plastic varied from 10% to 100% of the mass of
65% P – H/L HDPE - 65 50/50 Sand 1, 3, 7, 28 dry sand. In the preliminary test, some procedures in making the sam­
50/50 LDPE ples are optimized, including the timing to introduce different types of
P: plastic content in percentage (%) of dry mass of sand, H/L mass ratio of HDPE plastics and sand. Another important parameter that is established in the
to LDPE. preliminary test is the melting and casting temperature. At an ambient

Fig. 3. Samples of MPB before and after extraction from the mold.

3
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

air temperature between 24 ◦ C and 30 ◦ C, the average melting tem­ repeatability of the results.
perature in the oven is found to be 250 ◦ C. The results of the preliminary
test indicate that, for a plastic content less than 40%, the mortar will be 2.3.3. Absorption test
very heterogeneous and some of the sand particles are not covered by Water absorption by immersion and capillarity tests were conducted
the plastic binder. On the other hand, the mortar is found to be too fluid on the MPB and conventional mortar samples. The immersion test was
or liquid with more than 70% plastic, thus allowing rapid settlement of conducted in accordance with ASTM C97 / C97M – 18 [32]. The cy­
the sand grains, and leading to segregation and a heterogeneous sample lindrical samples (5 cm in diameter, 10 cm in height) were first dried in a
during casting. In addition, flowing of the MPB in a liquid or fluid state ventilated oven at 60 ◦ C for 2 days. After drying, the samples were
does not allow proper casting of the sample which affects hardening of cooled in a temperature-controlled room (23 ± 2 ◦ C) for around 30 min.
the material. Immediately after cooling, the samples were completely immersed in
Considering the results from the preliminary tests, the plastic con­ distilled water in a temperature-controlled room (23 ± 2 ◦ C) for 48 hrs.
tents vary between 45% and 65% of the initial mass of dry sand. Four At the end of the immersion, water on the surface of the sample was
proportions of plastic contents are chosen (45%, 50%, 60% and 65%) in wiped off with a cloth, and the samples were weighed to calculate the
this study with three HDPE/LDPE ratios (40/60, 50/50 and 60/40); see water absorption by immersion. The absorption of water, expressed in
Table 1. The combination of parameters here will cover the practical percentage (%), is equal to the difference in weight before and after
range of values in making a workable mortar. immersion for 48 h over the initial weight. Each test was repeated two
times and the average was calculated for a given plastic content and H/L
2.2.2. Conventional mortar with cement paste ratio.
Since “conventional” or “normal” mortar made of sand, cement and Capillary tests were carried out in accordance with ASTM C1585 – 13
water is one of the most widely used construction materials in civil [33] on the MPB and conventional mortars. Before performing the
engineering worldwide, some conventional mortar samples were pre­ capillary tests, the samples were oven-dried at 50 ◦ C until they reached a
pared to compare and identify or understand how the MPB differs from constant mass. Then, the sides of the sample were covered with plastic
conventional mortar. It should be noted that the goal of this exercise is tape for a unidirectional flow of water. The specimen was then placed in
not to compare the performance of these materials, but rather to gain a a pan with water and a cover was used to avoid direct evaporation of the
better understanding of the properties or behavior of the MPB. There­ water from the pan. The water level was kept at 3 mm from the bottom
fore, several samples of conventional mortar with a water to cement of the specimens throughout the experiment. The experiment was con­
ratio (W/C) of 0.5, and sand to cement (S/C) ratio of 3 were prepared. ducted under a controlled temperature of 23 ± 2 ◦ C and tap water was
Type I Portland cement and tap water were used to prepare the samples. used for the test. The weight of the sample was determined at about 5-
The ingredients were mixed in a concrete mixer for about 7 min, and minute time intervals at the beginning of the test and increased
then poured into metallic cylinders of 5 cm in diameter and 10 cm in accordingly as the test progressed. The capillary absorption (I) is
height. The cylinders were then sealed (to avoid evaporation) and cured calculated using the following formula:
at room temperature. Moreover, cement paste (CP) samples for a
Δm
microstructural analysis were also prepared by using the same proced­ I= (1)
A*d
ures. The W/C ratio of all of the CP specimens was constant and equal to
0.5. After the required curing times, the samples were subjected to where Δm is the difference between the mass after immersion at time t
various tests and analyses as described below. from the initial mass of the sample, A is the exposed area of the spec­
imen, and d is the density of water, which is assumed to be 1 g/cm3.
2.3. Tests By plotting the capillary absorption against the square root of time,
the graph has two parts with different slopes, which are called the rate of
2.3.1. Uniaxial compression strength and splitting tensile tests absorption or sorptivity. The initial sorptivity or initial rate of absorp­
Mechanical properties, such as (compressive, tensile) strength and tion is considered to be the slope of the curve in the first 6 h, while
stress–strain behavior, are very important for mortar with regard to its secondary sorptivity is the slope of the curve between 1 and 7 days, as
capacity to withstand load and cracking potential when forces are described in ASTM standard C1585-13 [33]. However, for the purpose of
applied to the mortar. Consequently, these properties enable to establish this research, the change in the slope is observed after 10 h. Therefore,
the service life that can be expected for a given construction material. the initial sorptivity will be calculated based on the slope of the capillary
Thus, compression and splitting tests were performed on prepared the absorption curve up to 10 h.
MPB samples. It should be noted the determination of the water vapour perme­
Compression tests were conducted in accordance with ASTM C39 / ability of the proposed material was out of the scope of this study. Future
C39M – 18 to determine the uniaxial compressive strength of the MPB investigations should address it since this parameter is important in
[30]. In these tests, a computer-controlled mechanical press was used. assessing potential of water condensation.
The load was applied at a slow rate (1 mm/min). The axial deformations
were recorded with a data acquisition system. Each uniaxial compres­ 2.3.4. Microstructural analyses
sion strength (UCS) test was repeated at least 3 times to ensure the To understand the microstructure of the MPB and the conventional
repeatability of the results. mortar and how it affects the mechanical properties of the material,
Splitting tensile tests, which are commonly known as the “Brazilian several microstructural testing techniques were used to evaluate the
test”, were conducted on cylindrical samples that were 5 cm in diameter microstructure of some of the samples [34]. These tests or analyses
and 10 cm in height to determine the splitting tensile strength of the include mercury intrusion porosimeter (MIP) tests, scanning electron
material. The splitting tensile tests were performed in accordance with microscopy (SEM) and thermogravimetric analyses (TGA). Samples for
ASTM C496 / C496M – 17 [31]. Each test was repeated at least three the microstructural analyses were oven dried at 45 ◦ C. Drying at this
times. temperature does not cause cracking of the specimens [34,35].
For examination under a scanning electron microscope, approximate
2.3.2. Hardened density test square samples were cut and carbon coated for the SEM by using the Joel
The density of a hardened cemented material is a parameter that can JSM-6610LV scanning electron microscope. SEM examinations with
affect its strength and permeability. Therefore, the density of the hard­ different magnifications were used to analyze the adhesion/cohesion
ened MPB was determined by following ASTM C 138 / C138M − 17a between various elements of the MPB, its morphology and texture, for a
[32]. Each density test was performed at least three times to verify the

4
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

clear view on its porosity microscopically. plastic behaves differently when interacting with granular materials. At
The MIP tests were carried out to evaluate the pore size distribution high temperatures during the initial stage of cooling after casting, weak
or the total porosity of the specimens (MPB, conventional mortar). The spots develop at the interfacial transition zone (ITZ) between the plastic
tests were performed by using a Micrometrics Auto Pore III 9420 mer­ paste and the aggregates. With time, the interface (ITZ) becomes
cury porosimeter in accordance with ASTM D4404-10 [36]. stronger during the solidification phase, which binds the constituent
The thermogravimetric analysis (TG-DTG tests) was performed with elements of MPB together to act as one unit and contributes to higher
a thermal analyzer (TGA Q5000 V3.15 Build 263). Less than 40 mg from compactness and elasticity (due to increase in degree of crystallinity of
each material, plastic wastes, MPB, and conventional mortar, were the plastic as its solidification progresses) of the material [42]. Thus,
placed on an alumna pan and heated under a nitrogen atmosphere from part of the strength improvement can be linked to the high elastic
10 ◦ C to 1000 ◦ C at a heating rate of 10 ◦ C / min and a flow rate of 10 modulus and better adhesion in the MPB. From Fig. 4, it can be seen that
mL/min to assess the thermal degradation. For analysis purposes, the the UCS of the MPB specimens reached around 60% to 80% of their final
change in weight is recorded as a function of the temperature increment values just after one day of curing. This increase in the early age strength
for the samples. for MPB is much higher than conventional mortar, which typically
reaches less than 40% of its final strength after 1 day of curing [43].
3. Results and discussion Fig. 5 illustrates the effect of plastic content with an H/L ratio of 50/
50 on the compressive strength of the MPB for various curing times.
3.1. Compressive strength of the MPB From Figs. 4 and 5, it is obvious that not only the curing time, but also
the plastic content, significantly affects the UCS of the MPB. The samples
Fig. 4 shows the time-dependent development of the compressive with 50% and 60% of plastic and H/L of 50/50 show the highest rate of
strength of the MPB with different plastic (P) contents for a constant H/L gain in strength as well as the ultimate strength, whereas the samples
ratio (50/50). Regardless of the plastic content, the compressive with 45% plastic and H/L of 50/50 show the slowest rate of gain in
strength is found to increase with time. It is observed that several hours strength and the lowest ultimate strength. These results suggest an
after casting the sample, the sample does not take its final shape, thus optimal plastic content from 50% to 60% with respect to the highest
indicating a slow transition (solidification) of the material from the UCS. Lower or higher than this optimal plastic content means that the
liquid to solid state which has an impact on the compressive strength of strength curves for different ages tend to decrease. This behavior can be
the material. When the MPB is poured into the mold after melting the attributed to two main factors. First, at a low plastic content, or less than
plastics, the samples exposed to an ambient temperature (24 ◦ C to 30 ◦ C) 45%, there is not enough melted plastic paste to cover all of the sand
exhibit accelerated solidification which leads to a rapid gain in strength. grains to solidly bind all the particles together. That leads to the creation
The change in strength is dependent on the polymer and the type of of weak spots in the MPB with low inter-particle bonding which reduces
polymer which can even increase after recycling [37,38]. LDPE and the UCS. On the other hand, when the plastic content is higher than 65%,
HDPE are both semicrystalline polymers but the degree of crystallinity is plastic then constitutes a significant portion of the material with less
higher in HDPE, due to the different number of polymer branches [39]. sand. Although this contributes to an increase in the amount of plastic as
The changes in strength can be linked to the changes in thermal prop­ a binding material, the decrease in sand content will lead to a decrease in
erties, melting temperature, crystallization, etc., and the physical/sur­ dry density which affects the strength of the material. Therefore, the
face properties of the mix after processing [40]. Volk et al. [41] observed compressive strength of the MPB is decreased at a plastic content higher
a similar behaviour on the improvement in the tensile strength of linear than the optimal values [18,19].
low-density polyethylene with surface treated cellulose nano-fibers. Figs. 6 and 7, show the change in UCS with time for MPB with 50%
Therefore, further understanding of the strength of the MPB can be and 60% of plastic, respectively, and H/L ratio of 40/60, 50/50 and 60/
gained by examining the material microscopically. Furthermore, melted 40. The results show clear increase in UCS with more HDPE. A H/L ratio

12

10

8
UCS, MPa

4
Plas�c content
45%P - H/L 50/50
2 50%P - H/L 50/50
60%P - H/L 50/50
65%P - H/L 50/50
0
0 5 10 15 20 25 30
Curing �me, days

Fig. 4. Compressive strength development vs time of the MPB with different percentages of plastic waste (sand particle size <2 mm, H/L of 50/50).

5
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

12

10

8
UCS, MPa

4
Curing �me
1 DAY
2 3 DAYS
7 DAYS
28 DAYS
0
40% 45% 50% 55% 60% 65% 70%
Plas�c content

Fig. 5. Effects of plastic content on the compressive strength for different ages of the MPB with H/L = 50/50.

20

18

16

14

12
UCS, MPa

10

6
Plas�c content
4 50%P - H/L 50/50

50%P - H/L 40/60

2
50%P - H/L 60/40
0
0 5 10 15 20 25 30
Curing �me, days

Fig. 6. Unconfined compressive strength vs time of the MPB for different H/L ratios with plastic content (P) of 50%.

of 50/50 gives the lowest strength, while a ratio of 60/40 provides the 3.2. Splitting tensile strength of the MPB
highest strength regardless of the curing time and plastic contents. This
behaviour can be related to the physico-chemical transformation of the Typical results of the splitting tensile tests on the MPB are presented
plastic during melting, casting and curing [38,40,41], as well as the in Figs. 8 and 9. Similar to the UCS test results, the splitting tensile
lower porosity of the MPB with H/L = 60/40. Further discussion on this strength of the MPB is found to increase with time for different H/L
point will be provided in the sections below related to the microanalysis ratios. The UCS lies between 1.5 and 4.6 MPa. Moreover, the splitting
and absorption tests. tensile strength is shown to be a function of the H/L ratio and plastic
content. It is believed that the mechanisms responsible for these changes

6
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

14

12

10

8
UCS, MPa

6
Plas�c content
4
60%P - H/L 50/50

2 60%P - H/L 40/60

60%P - H/L 60/40

0
0 5 10 15 20 25 30
Curing �me, days

Fig. 7. Compressive strength development vs time of MPB for various H/L ratio with plastic content (P) of 60%.

5.0
Spli�ng tensile strength, MPa

4.0

3.0

2.0
Plas�c content
50%P - H/L 40/60
1.0
50%P - H/L 50/50
50%P - H/L 60/40
0.0
0 5 10 15 20 25 30
Curing �me, days

Fig. 8. Splitting tensile strength vs time of the MPB with different H/L ratio and plastic content (P) of 50%.

are the same as those for the UCS. Moreover, the splitting tensile 3.3. Stress–strain behaviour of the MPB
strength appears to be very sensitive to the cracking of the material [35],
since different plastic contents and manufacturing processes can affect Fig. 11 shows typical stress–strain curves of the MPB under
the porosity of the mortar, cohesion between particles, and the adhesion compressive and tensile stresses. The stress–strain behaviour is similar
between plastic and sand particles in the mortar. These parameters will under compression and tension. In the first part of the test, the increase
affect the tensile strength of the materials. in stress leads to a small increase in strain under compressive or tensile
In practice, only the compressive strength is usually measured for stresses until reaching the peak values. At the peak, failure occurs, and
quality control and assessment of the performance of many construction the material continues to deform with a reduction in stress and continual
materials, particularly cemented materials. When needed, the tensile increase in strain. This mechanism continues until full disintegration of
strength can be determined by using empirical correlation or pre­ the material.
determined relationships. Measurements of the UCS and splitting tensile The MPB appears to have a similar stress–strain behaviour as the
strength make it possible to determine if there is a relationship between conventional mortar up to the peak strength. After the peak stress, the
them as shown in Fig. 10. Using a regression analysis, a linear rela­ conventional mortar tends to show brittle failure/fractures, see Fig. 12,
tionship is found between the splitting tensile strength (ft) and the which is not shown in the MPB. In other words, the MPB undergoes more
compressive strength (fc) which can be expressed by: plastic deformation, whereas with the conventional mortar is more
brittle with sudden failure [18].
ft = 0.1432*fc + 1.8129(inMPa) (2)
Figs. 13 and 14 present the photographs that show typical failure
modes of the MPB and the conventional mortar under compression.

7
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

Spli�ng tensile strength, MPa

Plas�c content
1 60%P - H/L 40/60
60%P - H/L 50/50
60%P - H/L 60/40
0
0 5 10 15 20 25 30
Curing �me, days

Fig. 9. Splitting tensile strength vs time of the MPB with different H/L ratio and plastic content (P) of 60%.

4
Spli�ng tensile strength, MPa

3 50%P - H/L 60/40

60%P - H/L 40/60

2
Correla�on
2
Linear (Correla�on)

1
5.00 7.00 9.00 11.00 13.00 15.00 17.00 19.00 21.00
UCS, MPa

Fig. 10. Splitting tensile strength vs UCS of the MPB with 50% plastic and H/L of 60/40 and with 60% plastic and H/L of 40/60.

Under compression, different types of failure modes are observed for the component [18,44,45]. This finding will be of great interest for future
MPB. Some samples split on the longitudinal side while others show applications especially when evaluating cracking, durability and loading
localized fractures in specific parts of the samples or failure into the capacity during the lifetime of pavements that use the MPB.
formation of a cone, see Fig. 13(a). Despite the fractures, as shown in
Fig. 11, the MPB tends to always maintain a connection between its 3.4. Water absorption of the MPB
different broken parts, which allow it to continue to support the applied
loads even after failure without total disintegration under both The purpose of the water absorption tests is to acquire a good un­
compression or tension. This new material undergoes extended straining derstanding of immersion and the capillary absorption behavior of the
before its peak with a change in slope of the deformation curve (Fig. 11). MPB to gain more insight into its porosity and pore structure, and sub­
After the peak, different parts of the sample tend to connect together for sequently link this pore structure to some of the key engineering prop­
a short period of time before completely disintegrating, thus indicating erties (e.g., strength, density, permeability, and durability) of the MPB,
that this material is more ductile than conventional mortar. Several similar to previous studies [47–49]. Fig. 15 presents the rate of ab­
researchers have found similar properties for construction materials that sorption by immersion vs time for up to 48 h for conventional mortar
contain plastic aggregates [44–46]. They all observe that concrete that and the MPB samples. The MPB samples with the highest and lowest UCS
contains plastic aggregates continue to deform after the peak load before were selected for the test. A higher rate of absorption, which ranges from
complete disintegration due to the ductile behaviour of the plastic 4% to 5%, can be observed in the conventional mortar with Portland

8
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

25

20

Stress, MPA
15

10

0
0 2 4 6 8 10 12
Strain, %
Fig. 11. Compressive and split tensile stress–strain curves of the MPB with 50% plastic and H/L of 60/40.

14

12

10
Stress, MPA

0
0 0.1 0.2 0.3 0.4 0.5
Strain, %
Fig. 12. Compressive stress–strain curve of conventional mortar after 28 days (sand particle size <2 mm, W/C of 0.5 and S/C ratio of 3).

cement. For the MPB, as expected, the absorption is higher for the conventional mortar is 7.00 mm, which is higher than that of the MPB
sample with a plastic content of 45% and H/L of 50/50, which has the samples. This difference clearly reveals the effect of the pore structure
lowest UCS in comparison to the samples with 50% plastic and H/L of because water used in making the cement paste is the major source of
50/50 and 60% plastic and H/L of 50/50 which correspond to a higher creating porosity in the cement mortar after the hydration process [51].
UCS. This means that the sample with 45% plastic has more inter­ In addition, the MPB contains only melted plastic paste as the binding
connected pores and higher void ratio or porosity than samples with material, which if compacted enough, will contribute to limit water
50% and 60% plastic. This larger porosity will obviously result in a less absorption.
dense structure and less resistant material [53–55]. This finding agrees According to previous studies, initial and secondary sorptivities can
with the UCS results. be used to evaluate the connectivity of the pore network [51,52]. No
Fig. 16 shows the initial and secondary absorption rates of the con­ study has been carried out on the water absorption of mortar with HDPE
ventional mortar and different MPB samples with the highest UCS. The and LDPE plastics. However, there are a number of relevant works on the
capillary absorption is found to increase with time for all of the samples water absorption of concrete/mortar with plastic aggregates. In general,
and reaches a constant value after 7 days. The capillary absorption these studies have shown that the increase in plastic aggregates in
graphs for both the MPB and the conventional mortar show similar concrete or mortar contributes to the increase of the capillary absorption
behavior as predicted by Halls’ function [50] with two stages as [53–55]. This increase is due to the creation of more pores during
described in ASTM standard C1585 – 20. In the first hour of immersion, pouring, casting and curing. The presence of pores can be caused by the
there is not much difference between the absorption values for all the particle configuration at the ITZs between natural aggregates (sand,
samples. After two hours of immersion, a clear difference can be gravel) and plastic. However, MPB made with melted plastic is very
observed between the two materials. The maximum absorption values different from cement paste. In this case, the porosity and strength will
are 5.02 mm and 2.88 mm for the 50%P - H/L 50/50 and 60%P – H/L not depend on the shapes of the plastic aggregates but tied to the
40/60 samples, respectively. However, the peak absorption value of the compactness when pouring liquid plastic in forming the MPB and the

9
M. Thiam and M. Fall Construction and Building Materials 302 (2021) 124190

a) 50%P – H/L 50/50 samples after UCS tests

b) 50%P – H/L 60/40 specimen after compression

Fig. 13. Images of typical failure behaviors observed in UCS tests on various MPB samples.

interfacial behaviour of the transition zone between plastic and sand. In 3.5. Density of the MPB
addition, since the MPB samples are prepared and poured at a high
temperature, over 100 ◦ C, there is greater chance of forming micro Fig. 17 presents the density of the MPB with time for different plastic
cracks and weak compaction points in the material. These cracks are contents with an H/L ratio of 50/50. Depending on the composition of
formed due to thermal gradients which cause expansion and/or the MPB, the density is found to vary with time. The test results show
contraction [56] at certain depths while the MPB is in a liquid state or that the average density of the MPB lies between 1.7 and 2.3 g / cm3. An
during manual compaction. This will affect the various properties of the average density of 2 g/cm3 is considered by the RILEM LC2 classification
MPB such as the initial and secondary rates of capillary absorption. On to be the minimum density for lightweight concrete [1,21]. The 28-day
the other hand, the absorption values (both immersion and capillarity) density of the conventional mortar is found to be 2.24 g / cm3. There­
of the MPB with optimal plastic contents are lower than the ones with fore, the MPB listed in Table 1 is lighter than conventional mortar. This
45% plastic and H/L of 50/50 which appears to have higher porosity lightness can be explained by the fact that the density of plastic is low,
than the other samples. Thus, the variation in plastic content and H/L 0.915–0.940 g cm− 3 for LDPE and 0.940–0.965 g cm− 3 for HDPE.
has an impact on the absorption values of the MPB which leads to [15,16,18,57,58]. For the conventional mortar, the binding component
changes in strength due to the configuration at the ITZ and the degree of is cement paste and the cement itself is denser (2.8 g.cm− 3) than the
compaction. Moreover, loss of surface material due to flaking of con­ plastics. Therefore, the cement paste that bonds the granular materials
ventional mortar is observed. This is not observed in the MPB during the will have a higher density than the virgin cement, which supports the
absorption tests which is advantageous for practical use of the material fact that MPB is lighter than conventional mortar [59].
in situations when the material will be in contact with running or Fig. 18 shows a slight decrease of the density of the MPB samples
flowing water or under high concentrations of aggressive chemicals. with increased plastic content. This is related to the fact that MPB with

10