Waste and Biomass Valorization (2022) 13:3821–3834

https://doi.org/10.1007/s12649-022-01708-x

ORIGINAL PAPER

Reuse of Waste Plastics in Developing Countries: Properties of Waste

Plastic‑Sand Composites
Alexander Kumi‑Larbi Jnr1 · Richard Galpin1 · Silva Manjula2 · Zoe Lenkiewicz3 · Christopher Cheeseman1

Received: 28 September 2021 / Accepted: 31 January 2022 / Published online: 14 February 2022
© The Author(s) 2022

Abstract
Waste plastics are a major problem in developing countries, where efficient collection and recycling systems often do not
exist. Plastic bonded sand composites provide a low-cost recycling alternative for selected waste plastics. This research has
investigated the production and properties of plastic bonded sand manufactured using low-density and high-density poly-
ethylene (LDPE and HDPE). Plastic bonded sand production in The Gambia was used as a case study to identify potential
barriers to the technology. Processing was done by oven moulding, or a heat-mixing technique, and the properties of the
LDPE and HDPE bonded sand samples formed have been determined. Processing at temperatures between 250 °C and
325 °C produced optimum compressive and flexural strengths. Higher processing temperatures reduced strength and lower
temperatures produced inhomogeneous samples. Thermal plastic degradation occurs at 400 °C in N ­ 2 and 250 °C in air.
Processing at temperatures below 250 °C in anoxic conditions is necessary to control off-gases. The optimum sand addition
to produce the highest compressive strength was between 65 and 80%, depending on the sand particle size. HDPE produced
higher maximum compressive strengths (37.1 MPa) compared to LDPE (27.2 MPa). Plastic bonded sand has increased
strength, toughness, ductility, and thermal conductivity compared to C20/25 concrete and sandcrete and it can be used for
wall construction blocks and paving tiles. The potential applications and implications of processing waste plastics in devel-
oping countries into plastic bonded sand are discussed.
Graphical Abstract

CIRCULAR ECONOMY FOR WASTE

PLASTICS IN DEVELOPING

Low-cost recycling of Waste plas cs

into Plas c bonded Sand construc on
products:

Ocean Plas cs • Material proper es behaviour

Applica ons and
• Cri cal produc on parameters limita ons of Plas c
• Health and Safety bonded sand

Keywords Circular economy · Ocean plastics · Sustainable development · Waste plastic recycling

* Alexander Kumi‑Larbi Jnr

alexander.kumi-larbi-jnr16@imperial.ac.uk Statement of Novelty
1
UKCRIC Advanced Infrastructure Materials Laboratory,
Department of Civil and Environmental Engineering, Low-cost recycling solutions are needed to address ocean
Imperial College London, London SW7 2BU, UK plastics from developing countries (DCs). Waste plastics are
2
Department of Materials, Imperial College London, being used as a binder to produce lightweight construction
London SW7 2AZ, UK products including paving and roof tiles in DCs to enhance
3
Waste Aid UK, Wye, Kent, UK recycling efforts. However, there is no comprehensive data

13
Vol.:(0123456789)
3822 Waste and Biomass Valorization (2022) 13:3821–3834

or clear understanding on the material behaviour of plastic wt.% of aggregate in low-density asphalt [18]. Flexible
bonded sand. This research gap has hindered its wide scale pavements have been produced with 30 wt.% of the aggre-
application despite its huge potential to enhance recycling gate replaced by LDPE [19].
efforts and provide jobs in DCs. The comprehensive primary A toolkit has been developed by WasteAid UK to train
data on the properties and health and safety concerns of plas- local communities in DCs on how to produce plastic
tic bonded sand provided in this work reveals the suitable bonded sand using ubiquitous single-use packaging plas-
applications and limitations of the novel technology. The tics [20]. The innovative use of plastic bonded sand in
findings underpin the development of a new value chain for paving blocks, tiles, advertisement poles and construction
waste plastics in DCs. panels for affordable housing is ongoing in many coun-
tries including Kenya, Columbia, Cameroon, Uganda, The
Gambia and Ghana. These types of recycling applications
Introduction provide local employment opportunities and ensure a cir-
cular economy for selected waste plastics [21–24]. How-
Despite significant improvements in global recycling rates, ever, there is limited research on the use of waste plastics,
less than 20% of waste plastics are properly managed [1]. particularly LDPE and HDPE, to produce plastic bonded
Improper plastic disposal causes environmental problems sand. Further understanding of the effect of temperature
with ocean plastics now recognised as a major global issue on the processing and properties of different compositions
[2, 3]. It is estimated that 80% of ocean plastics come from of plastic bonded sand is required. This research reports
coastal developing countries (DCs) where inadequate plastic on how processing conditions and composition influence
waste management systems result in indiscriminate dump- the gases generated during processing and the properties
ing of plastics into water bodies [4]. A typical example from of plastic bonded sand. The influence of processing tem-
Ghana is shown in Fig. 1. Improperly managed waste plas- perature, sand addition level and sand particle size on the
tics block drainage systems and waterways, increasing the compressive strength and flexural strength of LDPE and
risk of flooding and stagnant water and this can have a seri- HDPE based plastic bonded sand are reported. The poten-
ous deleterious effect on public health. tial for these materials to provide a solution to the plastic
Low-cost plastic recycling technologies that produce waste problem in DCs is discussed.
aggregates and fibres for use in lightweight concrete and
flexible pavements exists [5–9]. Concrete containing poly-
ethylene aggregate have improved flexibility and tough-
ness [10]. High performance flexible pavements have
been produced from waste plastic modified asphalt with
reduced costs [11–17]. PET has been used to replace 20

Fig. 1  Indiscriminate dumping of plastic waste into the Korle Lagoon in Accra, Ghana

13
Waste and Biomass Valorization (2022) 13:3821–3834 3823

Case Study of Plastic Bonded Sand recover ~ 80% of valuable recyclables from municipal solid
Production in the Gambia waste using simple sorting techniques [25]. LDPE was
separated from the waste using the recycling number on
Plastic bonded sand production in Gunjur, The Gambia containers (4 for LDPE) and simple flame and stretching
was studied in June 2019 to understand how these mate- tests, as shown in Fig. 2a. LDPE produces a blue flame
rials are produced in the field and identify the barriers when ignited and has greater flexibility and ductility than
to more widespread use of the production process. The other plastics. Local construction sand was used as the
study was conducted in collaboration with WasteAid UK filler. The sorted LDPE and sand were sun-dried prior to
and involved 90 Gunjur residents collecting, sorting, and processing and firewood was used as fuel. Although fire-
processing plastic waste into plastic bonded sand products. wood is ubiquitous in Gunjur, a more efficient fuel source
800 kg of LDPE was processed each week into roofing and is recommended for the process.
paving tiles and these were used to pave local businesses’ The required quantity of LDPE was heated in a locally
premises. fabricated metallic steel drum preheated to about 150 °C.
Waste plastics were collected from households The LDPE was then further heated until it melted at tem-
and waste dumps. The informal waste collectors can peratures between 220 to 420 °C as shown in Fig. 2b. Sand
was then mixed into the molten LDPE, until a homogenous

Sorting process and Melting, sand Casting Process

Heating of melting addition and mixing

(a) (b) (c)

Fig. 2  Plastic bonded sand production in The Gambia. The process involves sorting and melting LDPE waste plastics, mixing of sand into
molten plastics and casting into steel moulds to form plastic bonded sand paving tiles

13
3824 Waste and Biomass Valorization (2022) 13:3821–3834

dough-like paste was formed. This was then cast into steel observations that formed the basis for the experiments in
moulds, pre-coated with palm oil, to form plastic bonded this research included:
sand paving and roofing tiles as shown in Fig. 2c. The tem-
perature of the paste before casting was between 200 and • The release of smoke and off-gases due to processing at
210 °C. All temperature measurements were recorded using uncontrolled temperatures
an infrared thermometer. • The effect of varying processing temperature on the plas-
Those involved in the processing wore gas masks, fire tic bonded sand properties
retardant overalls, safety boots, heat-proof gloves, and safety • The effect of sand addition and particle size on plastic
glasses to minimise the risk of exposure to off-gases. The bonded sand properties.
plastics often ignited during melting and produced smoke
as shown in Fig. 2b. The smoke may contain hazardous
compounds that pose a health risk to the workers [26–28].
Potential health impacts of the compounds may include
sensory effects, target organ damage, nervous system and Materials and Methods
respiratory effects [29]. Work exposure limits for selected
compounds produced during the thermal processing of pol- Polyethylene ­(C2H4)n is a non-polar polymer with no func-
yethylene at > 325 °C are shown in Table 1 [30–32]. The tional groups attached to the carbon backbone and is catego-
resin type, residual monomer content, degree of stabilization rised into low-density polyethylene (LDPE) and high-density
and the presence of additives influence the characteristics polyethylene (HDPE). LDPE has high ductility with densi-
of the VOCs emitted. Additives and UV stabilizers present ties between 0.91 and 0.94 g ­cm−3 and crystallinity between
in processed plastics such as dibutyl hydroxytoluene vola- 35 and 55%. HDPE has a density greater than 0.94 g ­cm−3
tilize at processing temperatures > 265 °C [33]. Key field with higher crystallinity, strength and stiffness compared to

Table 1  EH40/2005 workplace exposure limits and GHS hazard classes and categories for compounds identified in off-gases at LDPE process-
ing workshops [30, 34]
Compound Work exposure limit Hazard class and category (H)
Long-term Short-term 224 300 311 314 317 319 330 335 340 350 400
(8 h) /ppm (15 min) /ppm 225 301 315 331 336 341 410
226 302 332 372 411
304 373 412

Formaldehyde 2 2 ✔ ✔ ✔ ✔ ✔ ✔ ✔
Acetaldehyde 20 50 ✔ ✔ ✔ ✔ ✔
Acetone 500 1500 ✔ ✔ ✔
Acrolein 0.02 0.05 ✔ ✔ ✔ ✔ ✔ ✔
Methyl ethyl ketone 200 300 ✔ ✔ ✔
Methyl propyl ketone 200 250 ✔ ✔ ✔ ✔
Heptan-2-one 50 100 ✔ ✔ ✔
Heptan-3-one 35 100 ✔ ✔ ✔
propionic acid 10 15 ✔
n-Pentane 600 – ✔ ✔ ✔ ✔
dibutyl hydroxytoluene – – ✔
Acetic acid – – ✔ ✔
Butanoic acid – – ✔
Pentanoic acid – – ✔ ✔
Propanone – – ✔ ✔ ✔
1,2-Dichloroethane 5 – ✔ ✔ ✔ ✔ ✔ ✔
Styrene 100 250 ✔ ✔ ✔ ✔ ✔
Ethylbenzene 100 125 ✔ ✔ ✔ ✔
Benzene 1 – ✔ ✔ ✔ ✔ ✔ ✔ ✔

H224/H225/H226 flammable liquid and vapours, H300/H301/H302/H304 oral toxicity, H311 dermal toxicity, H314/H315 skin irritation/corro-
sion, H317 skin sensitization, H319 eye damage/eye irritation, H330/331/332 inhalation toxicity, H335/H336/H372/H373 specific target organ
toxicity, H340/H341 germ cell mutagenicity, H350 carcinogen, H400/H410/H411/H412 hazardous to the aquatic environment

13
Waste and Biomass Valorization (2022) 13:3821–3834 3825

LDPE. Water sachets made of LDPE and HDPE bottle caps plastic bonded sand has been investigated using the OMT
were used in these experiments. preparation method. The effect of thermoplastic binder
Commercially available silica sand with a particle density type on the compressive strength was also investigated,
of 2.65 g ­cm−3 was used as the filler. This was dried and although the different plastics required different melt times
sieved into three different size fractions with particle sizes to produce optimum samples. LDPE samples were heated
(d) in mm of d < 0.30, 0.30 < d < 0.60 and 0.60 < d < 1.35. for 70 min while HDPE samples were heated for 90 min.
The three sizes were assumed to have average particles sizes The effect of sand proportions and sand particle size on the
of 0.15 mm, 0.45 mm, and 0.98 mm. plastic bonded sand properties using the HMT production
Test samples were produced using two processes. The method has also been investigated.
oven moulding technique (OMT) involved heating layers of A Zwick Roell 1474 test machine was used to deter-
a mix of plastics and sand to the required processing tem- mine compressive strengths, plasticity, and elastic modu-
perature (Wild Barefield M253 oven with a temperature lus. Compressive loads were applied at a strain rate of
sensitivity of ± 5 °C). The mix was stirred after 7, 10 and 10 mm ­min−1 until the samples exhibited 10% deformation
15 min at the temperature required to achieve a homogene- or the applied stress had reduced after peak load to 10% of
ous mix. The mix was then cast into 50 mm-cubed steel the peak strength. Vernier callipers were used to measure
moulds, pre-heated to at least 150 °C to ensure slow cooling, sample dimensions for stress–strain computations and data
effective compaction and prevent deformation due to rapid outputs were analysed using the TestXpert 3 version 1.11
temperature change [35, 36]. The mix was then compacted software to produce compressive stress–strain curves.
and cooled to room temperature to form the plastic bonded Three-point bending tests were conducted in accord-
sand samples. The test samples were processed at a tem- ance with ASTM C580-02 (Zwick Roell Z010). At least
perature range between 250 and 375 °C at 25 °C intervals. five samples were tested, and the results averaged to give
In the heat-mixing technique (HMT) the plastic was each data point. Tests were completed at a crosshead speed
melted in a covered saucepan on a hotplate. The required of 1 mm ­m in−1 using a support distance (L) of 30 mm.
quantity of sand was then mixed into the molten thermo- The sample bar depth (d) and width (b) were measured
plastic. Mixing was continued until a homogenous mix was using Vernier callipers. TestXpert 3 software measured
formed which was then compacted into pre-heated 50 mm- deflections (D) at the centre of the beam to generate
cubed steel moulds. The samples were formed after cooling force–displacement curves. The total work of fracture was
to room temperature [35, 36]. computed from the area under the stress–strain curves.
Table 2 shows the different mix designs of samples pre- Flexural stress and strain were calculated using the fol-
pared and tested in this work. The effect of processing lowing equations:
temperature on the compressive and flexural strength of

Table 2  Systemic variation in mix design of test samples to investigate the effect of processing temperature, sand particle sizes, thermoplastics
binder type and proportions on mechanical properties of plastic bonded sand
Experiment Plastic type Sand to plas- Sand particle size Processing parameters Number
tic propor- of sam-
tions ples
(wt.%) (d in mm) Temp Time
(°C) (min)

Effect of processing temperatures on com- LDPE 75.0 d < 0.30 250–375 (at 25 °C intervals) 32 18
pressive strength (OMT)
Effect of processing temperatures on flex- LDPE 75.0 d < 0.30 250–375 (at 25 °C intervals) 17 and 27 60
ural strength (OMT)
Effect of thermoplastic binder type on LDPE 50–83.3 d < 0.30 200–300 70 15
compressive strength (HMT) HDPE 50–83.3 d < 0.30 250–300 90 15
Effect of sand proportions and particle size LDPE 50–83.3 d < 0.30 200–300 70 15
on plastic bonded sand properties (HMT) 50–83.3 0.30 < d < 0.60 200–300 70 15
50–83.3 0.60 < d < 1.35 200–300 70 15
HDPE 50–83.3 d < 0.30 250–300 90 15
50–83.3 0.30 < d < 0.60 200–300 90 15
50–83.3 0.60 < d < 1.35 200–300 90 15

OMT oven moulding technique, HMT heat mixing technique

13
3826 Waste and Biomass Valorization (2022) 13:3821–3834

( ) 3FL 6Dd 30
Flexural stress 𝜎f = ; Flexural strain(𝜀f ) = 2
2bd2 L
Thermal conductivities of plastic bonded sand samples 25

were measured using a non-destructive test based on the

Compressive Stress (MPa)

transient plane source technique in accordance with ISO 20
22007–2:2015 (Hot Disk M1 analyser, Thermal 18 Instru-
ments Ltd). The hot disk sensor was placed between the sur-
faces of two block samples at room temperature (23.3 °C). 15
Heat was applied from the sensor at an output power of
0.25mW for 40 s and temperature was measured as a func- 10
tion of time [37].
Images of fractured surfaces of samples tested in three-
point bending were obtained using scanning electron micros- 5
copy (SEM, Hitachi TM4000Plus machine).
Thermogravimetric Analysis (TGA) was used to study 0
the effects of temperature and time on plastic degradation 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
(Mettler Toledo TGA 2 with a 70 µl alumina crucible). 6 mg Strain
of LDPE was heated from 30 °C to 600 °C at a constant rate (a)
of 10 K ­min−1 in air and ­N2 under a purge gas flow rate of
50 ml ­min−1. Test samples were also heated at 260 °C and 30
310 °C for 20 min in isothermal experiments to observe the
effects of processing time on thermal degradation in air. 25
Compressive Strength (MPa)

20

Results
15

The effect of processing temperature on the stress–strain

10
behaviour and compressive strength of LDPE plastic bonded
sand samples containing 75 wt.% sand with a particle size
5
d ≤ 300 µm is shown in Fig. 3. Increasing the temperature
from 185 to 240 °C increases the compressive strength by
0
30–65% depending on the sand content. Processing at tem- 220 240 260 280 300 320 340 360 380 400
peratures between 250 and 350 °C produced no significant
Processing Temperature (°C)
variation in compressive strength. Higher temperatures
decreased compressive strengths and reduced the work of (b)
fracture due to plastic degradation. Samples prepared at
375 °C could support approximately 20% of the maximum Fig. 3  The effect of varying processing temperatures on: a stress–
load at a strain of 0.075 while samples prepared at 250 °C strain behaviour of LDPE plastic bonded sand with 75wt.% sand
retained at least 74% of ultimate strength at a strain of 0.024. and d ≤ 300 µm. Increasing brittleness is observed with increasing
temperature. b the compressive strength of LDPE bonded sand with
The compressive modulus of plastic bonded sand determined
75wt.% sand and d ≤ 300 µm
between 0 and 0.012 strain varied between 0.6 and 0.8 GPa.
The failure modes of the samples tested in compression
are shown in Fig. 4. Samples processed above 325 °C failed 250 and 325 °C. Processing above 325 °C caused a sig-
along a 45° shear plane and disintegrated into smaller pieces nificant reduction in flexural strength and a reduced work
[35]. Below 325 °C samples failed along the compaction of fracture. Increasing processing time produces a more
layers without disintegrating due to the heterogeneity of the homogenous sample with improved sand packing that
samples processed at lower temperatures. enhances the compressive and flexural strengths. However,
The effects of processing temperatures on the flexural Fig. 5b shows that LDPE plastic bonded sand samples pro-
properties of plastic bonded sand samples are shown in cessed for 17 min had a greater flexural strength than sam-
Fig. 5. The flexural modulus of samples processed above ples processed for 27 min.
275 °C ranged between 0.5 to 0.7 GPa. Flexural strengths The work of fracture of plastic bonded sand samples
of samples processed for 27 min remained constant between decreased with increasing processing temperature as shown

13
Waste and Biomass Valorization (2022) 13:3821–3834 3827

Compressive Force
30
250℃
25 325℃
350℃

Flexural Stress (MPa)

20 375℃

15

10

5
50mm
- ∆L
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
Strain
(a)

25 17 mins

27 mins
50mm
20
(a)
Flexural Strength (MPa)

Compressive Force 15

10

0
240 260 280 300 320 340 360 380
50mm
Processing Temperature (°C)
- ∆L
(b)

Fig. 5  The effect of varying processing temperatures on: a Flexural

stress–strain behaviour of LDPE plastic bonded sand with 75wt.%
sand and d ≤ 300 µm. Increasing brittleness is observed with increas-
ing temperature. b Flexural strength of LDPE bonded sand with
75wt.% sand and d ≤ 300 µm, processed for 17 and 27 min
50mm+∆Lh

(b) in Fig. 6. Work of fracture data for samples processed at

250 °C for 17 min was inconsistent due to sample hetero-
Fig. 4  Failure modes observed in LDPE plastic bonded sand samples geneity. The work of fracture reduced rapidly at 275 °C.
with 75wt.% sand and d ≤ 300 µm: a approximate 45° shear plane
observed for samples processed at > 325 °C. b shear failure along
Increasing processing temperature from 325 to 375 °C
compaction layers in samples processed at ≤ 325 °C. ∆L refers to the reduced the average work of fracture of LDPE samples pro-
deformation in direction of the applied compressive force and ∆Lh is cessed for 17 min from 68 to 10% of the value at 275 °C. For
the deformation perpendicular to the applied compressive force samples processed for 27 min, the work of fracture reduced
from 64 to 13% of the value at 275 °C. This was due to the
thermal degradation of the thermoplastic binder resulting
from longer heat exposure at elevated temperatures.

13
3828 Waste and Biomass Valorization (2022) 13:3821–3834

3.0 40 d<300µm
27 mins
300µm<d<600µm
17 mins 35
2.5
600µm<d<1.35mm

Compressive Strength (MPa)

30
Work of Fracture (MJ/m3 )

2.0

LDPE Bonded Sand

25

1.5
20

1.0 15

10
0.5

0.0
240 260 280 300 320 340 360 380 0
40 45 50 55 60 65 70 75 80 85 90
Processing Temperature (°C )
Sand proportions (wt.%)

Fig. 6  The effect of processing temperatures on the work of fracture (a)

of LDPE plastic bonded sand processed for 17 and 27 min. Error bars
are indicative of the standard deviation of 5 samples
40

Figure 7 shows the effect of sand proportions and 35

sand particle sizes on the compressive strength of LDPE

30
and HDPE plastic bonded sand samples. The optimum
Compressive Strength (MPa)

sand proportions that gave the highest compressive

HDPE Bonded Sand

25
strengths ranged between 65 to 75 wt.% for LDPE and
65 to 80wt.% for HDPE, depending on sand particle 20
size (d). For the LDPE samples, sand with d < 300 µm,
300 µm < d < 600 µm, and 600 µm < d < 1.35 mm produced 15 d<300µm
maximum compressive strengths of 27.2 MPa, 21.8 MPa,
300µm<d<600µm
and 20.1 MPa, respectively. Optimum sand proportions 10

for compressive strength decreased with decreasing sand 600µm<d<1.35mm

particle sizes. Increasing sand content above the optimum 5

proportions caused a more rapid decline in the compres-

0
sive strength for LDPE samples with the same sand par- 40 45 50 55 60 65 70 75 80 85 90
ticle size. HDPE plastic bonded sand samples had maxi- Sand proportions (wt.%)
mum compressive strengths between 31.4 and 37.1 MPa
depending on the sand particle size. The maximum com- (b)
pressive strengths decreased with increasing sand parti-
cle size. However, the relationship between compressive Fig. 7  The effect of varying sand proportions and particle sizes on the
strength and sand particle size was dependent on sand compressive strengths of: a LDPE plastic bonded sand and b HDPE
plastic bonded sand. Error bars indicate the standard deviation of the
proportions. The compressive strength of LDPE samples
compressive strength values for three test samples
with ≤ 75 wt.% sand decreased with increasing sand parti-
cle sizes. Increasing sand particle sizes increased the com-
pressive strength of LDPE samples with > 75 wt.% sand The effects of varying sand addition on the stress–strain
and d < 500 µm. Further increase in sand particle sizes behaviour of samples are shown in Fig. 8. LDPE samples are
above 500 µm decreased LDPE compressive strength. In more ductile with lower compressive strengths and tough-
contrast, for HDPE samples, the compressive strength ness than HDPE samples. The work of fracture of samples
decreased with increasing sand particle sizes irrespective ranged between 0.6 to 2.3 MJ ­m−3 for LDPE samples and
of sand proportions. Optimised plastic bonded sand com- 1.0 to 2.1 MJ ­m−3 for HDPE samples depending on sand
pares favourably with C20/25 concrete with a compressive addition and particle size. The work of fracture of LDPE
modulus of 31GPa after 28 curing days. However, it has samples increased with increasing sand proportions to 68
only 2–2.5% of the stiffness of concrete. wt.%. The sand particle size had no impact on the work of

13
Waste and Biomass Valorization (2022) 13:3821–3834 3829

40 50.0 wt.% 2.2

66.7 wt.% 2.0

35 75.0 wt.%
1.8

Thermal Conductivity (W mK-1)

80.0 wt.%
30 1.6
Compressive Stress (MPa)

83.3 wt.%

LDPE bonded Sand

LDPE Bonded Sand

1.4
25
1.2
20
1.0

15 0.8

0.6
d<300µm
10 300µm<d<600µm
0.4
600µm<d<1.35mm
5 0.2

0.0
0 40.00 50.00 60.00 70.00 80.00 90.00
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
Sand proportions (wt.%)
Strain (%)
(a)
(a)
2.2

40 50.0 wt.% 2.0

66.7 wt.%

Thermal Conductivitiy (W mK-1)

1.8
35 75.0 wt.%
80.0 wt.% HDPE bonded Sand 1.6
Compressive Stress (MPa)

30 83.3 wt.% 1.4

HDPE Bonded Sand

1.2
25
1.0
20 0.8

0.6 d<300µm
15
0.4
300µm<d<600µm
10
0.2
600µm<d<1.35mm

5 0.0
40.00 50.00 60.00 70.00 80.00 90.00
0 Sand Proportions (wt.%)
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
(b)
Strain (%)
(b) Fig. 9  The effect of varying sand proportions and particle sizes on the
thermal conductivities of: a LDPE bonded sand and b HDPE bonded
sand
Fig. 8  The effects of varying sand proportions on the stress–strain
behaviour of: a LDPE bonded Sand with d < 300 µm. b HDPE
bonded sand with d < 300 µm
the highest elastic modulus in LDPE samples. The elastic
modulus of LDPE samples with ≤ 75 wt.% sand decreased
fracture of LDPE samples with ≥ 75 wt.% sand. The work of with increasing sand particle size. For LDPE samples with
fracture of LDPE samples with ≤ 75wt% sand increased with more than 75 wt.% sand the elastic modulus increased with
decreasing sand particle size. A maximum work of fracture increasing sand particles size and 80 wt.% sand additions
of 2.1 MJ ­m−3, 1.9 MJ ­m−3 and 2.0 MJ ­m−3 was observed produced the highest elastic modulus in HDPE samples.
for HDPE samples with d < 300 µm, 300 µm < d < 600 µm, The thermal conductivities of LDPE and HDPE samples
and 600 µm < d < 1.35 mm, respectively. The work of frac- were between 0.8 to 1.6 W m ­ K−1 and 1.0 to 1.9 W m
­ K−1
ture required to fracture HDPE samples increased up to the respectively. Figure 9 shows that thermal conductivity of
optimum sand addition of 67wt.%. Further increases in sand plastic bonded sand increases with increasing sand content
content decreased HDPE sample work of fracture irrespec- and decreases with increasing sand particle sizes.
tive of sand particle size. Thermal degradation of the plastic binder changes the
The elastic modulus of LDPE samples was between 0.5 fracture properties of plastic bonded sand. SEM images in
and 1.0 MPa and HDPE samples had an elastic modulus Fig. 10 shows the higher ductility in samples processed at
between 1.2 and 1.3 MPa. A 75 wt.% sand addition produced 275 °C with long stretched-out plastic strands producing a

13
3830 Waste and Biomass Valorization (2022) 13:3821–3834

rough fracture surface. Increasing the processing tempera-

ture to above 325 °C resulted in a much smoother fracture
surface with minute cracks within the plastic-sand matrix.
These defects act as stress concentration points which caused
brittle failure in samples processed at higher temperatures.
Figure 11 shows the TGA curves of LDPE waste plastics.
Thermal degradation begins at 400 °C and 240 °C in N ­ 2
and air respectively. The degradation rate rapidly increased
with increasing temperatures > 450 °C in N ­ 2. Peak degrada-
tion occurred at 490 °C and 99.5% of LDPE mass was lost
by 507 °C. Thermal degradation in air occurred in three
distinct phases: low-rate thermal oxidation below 0.25%
°C−1, rapid degradation above 0.25% °C−1, and degradation
of residuals. The degradation rate increased gradually with
increasing temperature until 370 °C. Higher temperatures
sharply increased the degradation rate to a peak at 450 °C.
Degradation rate then slowed until 99.5% of the LDPE mass
was lost at 570 °C. The isothermal experiments revealed
that peak degradation of samples occurred at the onset of
the isothermal temperatures (310 °C and 260 °C) and then
declined due the accumulation of degraded products on the
sample surface. Peak degradation at 310 °C was 62% greater
than at 260 °C. After 6 min exposure to the isothermal tem-
peratures, both samples lost 50% of the total mass lost over
the 20 min isothermal period.

Discussion

Transforming waste LDPE and HDPE plastics into plastic

bonded sand composites is a low-cost recycling option. Plas-
Fig. 10  SEM images of fractured LDPE bonded sand samples show- tic bonded sand composites are durable lightweight mate-
ing: a rough surfaces with long plastic strands in samples processed
at 275 °C and b smooth surfaces of samples processed at ≥ 325 °C
rials produced with low energy requirements. LDPE and
HDPE plastic bonded sand have properties suitable for use
in selected construction products as shown in Table 3.
Understanding the effects of the processing temperature,
sand content and sand particle sizes can optimise plastic

Fig. 11  TGA curves of LDPE

waste plastic showing the mass
loss and rate of mass loss with
increasing temperature at a
­ in−1 in air
heating rate of 10 K m
and nitrogen

13
Waste and Biomass Valorization (2022) 13:3821–3834 3831

Table 3  The suitability of LDPE bonded sand and HDPE bonded forms the bonding mechanism. For mechanical coupling,
sand for selected applications the sand surface irregularities are filled with the binder
Designation Minimum aver- Maximum LBS HBS to form an interlocking pattern [40]. Higher sand content
age compressive water absorp- increase sand surface irregularities. Lower sand content
strength (MPa) tion (%) reduces adhesion sites with increasing sand particle size.
Building Building brick (ASTM-C62) However, the low quantities of irregularly shaped sand par-
SW 20.7 13.26 √ √ ticles slightly increase the effect of mechanical interlocking
MW 17.2 19.36 √ √ to overcome internal friction and de-wetting. Sand parti-
NW 10.3 No limit √ √ cles impose relatively higher strain rates to induce matrix
Facing brick (ASTM-C216) strengthening in plastic bonded sand. The proportion and
SW 20.7 13.26 √ √ size of the sand particles influences the effects of these fac-
MW 17.2 19.36 √ √ tors on the mechanical properties of Plastic bonded sand
Thin veneer brick (ASTM-C1088) [35, 41].
Ext 17.2 13.26 √ √ Plastic bonded sand properties are controlled by the
Int 17.2 19.36 √ √ rigidity percolation theory [42]. The percolation point is a
Pavement Pedestrian and light traffic paving brick (ASTM-C902) continuous path of interconnected filler particles formed as
SX 27.6 8 x √ the filler content increases [43]. At the rigidity percolation
MX 20.7 14 √ √ point, there is just enough plastic binder to form a rigid inter-
NX 20.7 No limit √ √ particle network. Percolation of rigid regions is required to
Heavy vehicular paving brick (ASTM-C1272) cause significant property changes. Rigidity percolation is
F 69.0 6 x x achieved in plastic bonded sand with 20–25 vol% polymer
R 55.2 6 x x content [43]. Above rigidity percolation, porosity increases
Solid interlocking paving units (ASTM C936) because there is not enough binder volume to completely
50.0 7 x x encapsulate sand particles. The use of larger sand particles
in a given volume reduces the plastic bonding efficiency as
LBS LDPE plastic bonded sand, HBS HDPE plastic bonded sand, SW the binder thickness encapsulating the sand reduces [35].
severe weathering conditions, MW medium weathering conditions,
NW for use in negligible weathering conditions, Ext grade exterior: This is compensated for in normal Portland cement con-
for use where high resistance to cyclic freezing and thawing damage crete using a higher water/ admixture ratio [44]. In plastic
is desired, Int grade interior: for use where moderate resistance to bonded sand, porosity increases and reduces the compressive
cyclic freezing and thawing damage is permissible, SX for use where strength [35]. The rigidity percolation point is controlled by
brick may be frozen with saturated water, MX- for exterior use where
freezing resistance is not required, NX for exterior use but may be the plastic binding capacities. Using different blends of ther-
accepted for interior use where freezing resistance is required. F brick moplastics alters the binder volume capacities. Coating sand
to be set in an aggregate setting bed, R brick to be set in mortar or with LDPE and PET reduced the compressive strength of
bituminous setting beds the composites. Partial replacement of PET with PP reduced
composite strengths due to the difficulty in blending the two
bonded sand properties. The ideal processing temperature thermoplastics because of varying melting points [36]. Fur-
needs to achieve the ideal binder viscosities for effective ther studies to investigate the optimum melting temperature
incorporation of the sand into the plastic matrix to form the for thermoplastic blends is recommended.
plastic bonded sand. The optimal processing temperature for Optimised LDPE and HDPE plastic bonded sand sam-
plastic bonded sand produced using OMT and HMT ranged ples are suitable for applications where strength, toughness
between 250 and 325 °C and there was no significant vari- or ductility are required including pavement, roof tiles and
ation in compressive and flexural strength of samples pro- building partitions. Optimised plastic bonded sand has com-
cessed at such temperatures. These temperatures allow flow pressive strengths comparable to C20/25 concrete. The flex-
of the binder to reduce pores within the plastic-sand matrix. ural to compressive strength ratio of LDPE plastic bonded
Increased cooling times also increase the degree of crystal- sand is at least 5 times higher than concrete. PP bonded
linity [38, 39]. Higher temperatures adversely affect plastic sand samples recorded a flexural strength approximately 3
bonded sand properties due to thermal degradation of the times higher than concrete and 5 times higher than Asphalt
plastic. Lower temperatures form heterogenous samples with concrete [45]. The optimum sand proportions that produced
compromised properties due to inefficient mixing. the maximum compressive strengths in LDPE and HDPE
Plastic bonded sand properties are influenced by the samples ranged between 65 to 75 wt.% and 65 to 80% wt.%
plastic matrix strength and the interfacial bonding between respectively depending on sand particle size. This is con-
the plastic and sand. Sand encapsulation within the plastic sistent with values recorded for other plastic bonded sand
binder through mechanical coupling and molecular bonding materials including PET bonded laterite soil [36]. Plastics

13
3832 Waste and Biomass Valorization (2022) 13:3821–3834

compounded with particulate fillers have reduced ductility as arrangement of the thermoplastic binder and aggregate in
observed in polypropylene filled with paper disposable cups the composite matrix affects properties. The processing tem-
[46]. Plastic bonded sand exhibits a viscoelastic behaviour perature for achieving optimum plastic bonded sand proper-
similar to the stress–strain properties of asphalt concrete ties was in the range between 250 and 325 °C. Higher tem-
[35]. Initial compressive loading yields a linear stress–strain peratures reduce material toughness, plasticity, flexural and
curve. A shear plane is formed upon further loading due compressive strengths due to thermal degradation. Lower
to the inter-particle frictional forces generated as the sand temperatures produced heterogenous samples with reduced
particles slide against each other. The curve flattens out at strength due to inefficient mixing. Recycling waste plastics
the peak friction coefficient point and descends linearly as into plastic bonded sand is a resource efficient alternative
the coefficient of friction reduces [47]. with significant environmental and public health benefits
Plastic aggregates reduce the thermal conductivity of compared to the existing disposal options in DCs. However,
concrete due to the low thermal conductivity of plastics [48]. there is a risk of exposure to harmful off gases when plastic
However, plastic bonded sand samples recorded a higher bonded sand is produced at temperatures > 250 °C in air.
thermal conductivity than concrete or cement mortar. This Processing temperature and time need to be controlled. Pro-
was due to the increased heat transfer rate resulting from cessing temperatures at ≤ 250 °C under anoxic conditions
reduced sample porosity as observed in asphalt [49]. are recommended to control the off gases due to polymer
The existing heat-mixing process practiced in The Gam- degradation.
bia provides a simple inexpensive plastic recycling option
for job creation in low-income communities. However, an Acknowledgements The authors would like to thank Marcus Yio
(PhD) for the assistance in conducting some mechanical tests. Zoom-
improved design is needed to address the health and safety lion Ghana Limited is also acknowledged for sponsoring this project.
concerns and improve the plastic bonded sand product qual- Thanks to Pierre Kamsouloum, Africa Environmental Sanitation
ity. The primary mode of degradation during processing is (AFES) Consult and Waste AID UK for their support in the field work.
due to thermal oxidation and not pyrolytic degradation since
the typical processing temperatures are below the thermal Funding This research work was funded by Zoomlion Ghana Limited
through a PhD research at Imperial College London.
degradation temperature in the anoxic environment (400 °C).
Further research to evaluate the risk of exposure to the off Data Availability Not applicable.
gases produced during plastic bonded sand processing is rec-
ommended. The safety concerns and significant heat losses Code Availability Not applicable.
make it difficult to upscale the technology for mass plastic
bonded sand production. Extrusion and injection mould- Declarations
ing techniques could effectively maintain lower processing
temperatures to minimise off gases without compromising Conflict of interest The authors declare that they have no competing
the product quality. Optimisation of processing parameters interests.
including melt-mixing temperature, moulding temperature
Open Access This article is licensed under a Creative Commons Attri-
and pressure for the extrusion and injection moulding tech- bution 4.0 International License, which permits use, sharing, adapta-
niques need to be prioritised for superior plastic bonded sand tion, distribution and reproduction in any medium or format, as long
properties [39]. as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
Conclusions included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
Waste LDPE and HDPE plastics can be used to manufac- the article's Creative Commons licence and your intended use is not
ture plastic bonded sand materials using inexpensive pro- permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
cesses with low energy and water requirements. This can copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
provide an important reuse application for waste plastics
and particularly in DCs. The processing parameters critical
for optimising plastic bonded sand properties for selected
construction materials are reported. Plastic bonded sand References
made with LDPE and HDPE are durable and suitable for
use in pavement blocks, roof tiles and partition walls. Opti- 1. The Economist: The environment; The known unknowns of plas-
mised samples have maximum compressive strengths that tic pollution. Economist (United Kingdom) (2018). https://​www.​
are comparable to C20/25 concrete and significantly higher econo​mist.​com/​inter ​natio​nal/​2018/​03/​03/​the-​known-​unkno​wns-​
of-​plast​ic-​pollu​tion. Accessed 28 Jan 2021.
than sandcrete. They are tough and have higher ductility
and thermal conductivities than concrete. The structural

13
Waste and Biomass Valorization (2022) 13:3821–3834 3833

2. Gondal, M.A., Siddiqui, M.N.: Identification of different kinds of 23. The Times of India: Indian homes and pavements made from
plastics using laser-induced breakdown spectroscopy for waste plastic—BBC News—YouTube (2019). https://​www.​youtu​be.​
management. J. Environ. Sci. Heal. 42(13), 1989–1997 (2007) com/​watch?v=​icgze​SWo7C4. Accessed 24 Feb 2021.
3. Gall, S.C., Thompson, R.C.: The impact of debris on marine life. 24. CGTN Africa: Entrepreneur transforms plastic waste into con-
Mar. Pollut. Bull. 92(1–2), 170–179 (2015) struction materials—YouTube (2019). https://​w ww.​youtu​b e.​
4. Derraik, J.G.B.: The pollution of the marine environment by plas- com/​watch?v=​yA0nG​BSy6_I. Accessed 24 Feb 2021.
tic debris: a review. Mar. Pollut. Bull. 44(9), 842–852 (2002) 25. D. C. Wilson et al.: Global Waste Management Outlook. Inter-
5. Saikia, N., De Brito, J.: Use of plastic waste as aggregate in national Solid Waste Association (2015).
cement mortar and concrete preparation: a review. Constr. Build. 26. An, T., Huang, Y., Li, G., He, Z., Chen, J., Zhang, C.: Pollution
Mater. 34, 385–401 (2012) profiles and health risk assessment of VOCs emitted during
6. Siddique, R., Khatib, J., Kaur, I.: Use of recycled plastic in con- e-waste dismantling processes associated with different disman-
crete: a review. Waste Manag. 28(10), 1835–1852 (2008) tling methods. Environ. Int. 73, 186–194 (2014)
7. Huang, Y., Bird, R.N., Heidrich, O.: A review of the use of recy- 27. Lerner, J.E.C., Sanchez, E.Y., Sambeth, J.E., Porta, A.A.: Char-
cled solid waste materials in asphalt pavements. Resour. Conserv. acterization and health risk assessment of VOCs in occupational
Recycl. 52(1), 58–73 (2007) environments in Buenos Aires, Argentina. Atmos. Environ. 55,
8. Ahmadinia, E., Zargar, M., Karim, M.R., Abdelaziz, M., Ahma- 440–447 (2012)
dinia, E.: Performance evaluation of utilization of waste poly- 28. Sax, S.N., Bennett, D.H., Chillrud, S.N., Ross, J., Kinney, P.L.,
ethylene terephthalate (PET) in stone mastic asphalt. Constr. Spengler, J.D.: A cancer risk assessment of inner-city teenag-
Build. Mater. 36, 984–989 (2012) ers living in New York City and Los Angeles. Environ. Health
9. Moghaddam, T.B., Karim, M.R., Syammaun, T.: Dynamic prop- Perspect. 114(10), 1558–1566 (2006)
erties of stone mastic asphalt mixtures containing waste plastic 29. He, Z., Li, G., Chen, J., Huang, Y., An, T., Zhang, C.: Pollution
bottles. Constr. Build. Mater. 34, 236–242 (2012) characteristics and health risk assessment of volatile organic
10. Binici, H., Gemci, R., Kaplan, H.: Physical and mechanical compounds emitted from different plastic solid waste recycling
properties of mortars without cement. Constr. Build. Mater. workshops. Environ. Int. 77, 85–94 (2015)
28(1), 357–361 (2012) 30. Health and Safety Executive: EH40/2005 workplace exposure
11. Yildirim, Y.: Polymer modified asphalt binders. Constr. Build. limits limits for use with the Control of Substances (Fourth
Mater. 21(1), 66–72 (2007) Edition 2020) (2020).
12. Casey, D., McNally, C., Gibney, A., Gilchrist, M.D.: Develop- 31. Mitera, J., Michal, J., Kubát, J., Kubelka, V.: Analysis of
ment of a recycled polymer modified binder for use in stone thermo-oxidation products of polypropylene and polyethyl-
mastic asphalt. Resour. Conserv. Recycl. 52(10), 1167–1174 ene by Gas chromatography/mass spectrometry. Fresenius’
(2008) Zeitschrift für Anal. Chemie 281(1), 23–27 (1976)
13. Yildirim, Y., Hazlett, D., Davio, R.: Toner-modified asphalt 32. Andersson, T., Wesslén, B., Sandström, J.: Degradation of low
demonstration projects. Resour. Conserv. Recycl. 42(3), 295– density polyethylene during extrusion. I. Volatile compounds in
308 (2004) smoke from extruded films. J. Appl. Polym. Sci. 86(7), 1580–
14. Fang, C., Yu, R., Zhang, Y., Hu, J., Zhang, M., Mi, X.: Com- 1586 (2002)
bined modification of asphalt with polyethylene packaging 33. Patel, S.H., Xanthos, M.: Volatile emissions during thermoplas-
waste and organophilic montmorillonite. Polym. Test. 31(2), tics processing—a review. New Jerset (1995)
276–281 (2012) 34. U. National Library of Medicine: PubChem (2021). https://​
15. Ho, S., Church, R., Klassen, K., Law, B., MacLeod, D., Zan- pubch​em.​ncbi.​nlm.​nih.​gov/. Accessed 7 May 2021.
zotto, L.: Study of recycled polyethylene materials as asphalt 35. Kumi-Larbi, A., Yunana, D., Kamsouloum, P., Webster, M.,
modifiers. Can. J. Civ. Eng. 33(8), 968–981 (2006) Wilson, D.C., Cheeseman, C.: Recycling waste plastics in devel-
16. Dalhat, M.A., Wahhab, H.I.A.A.: Performance of recycled plas- oping countries: use of low-density polyethylene water sachets
tic waste modified asphalt binder in Saudi Arabia. Int. J. Pave- to form plastic bonded sand blocks. Waste Manag. 80, 112–118
ment Eng. 18(4), 349–357 (2017) (2018)
17. Murphy, M., O’Mahony, M., Lycett, C., Jamieson, I.: Bitumens 36. Maneeth, P.D., Pramod, K., Kishor, K., Shanmukha, S.: Utiliza-
modified with recycled polymers. Mater. Struct. Constr. 33(7), tion of waste plastic in manufacturing of plastic-soil bricks. Int.
438–444 (2000) J. Eng. Res. Technol. 2(4), 102–107 (2014)
18. Hassani, A., Ganjidoust, H., Maghanaki, A.A.: Use of plastic 37. Ashraf, A.: Thermal conductivity measurement by hot disk ana-
waste (poly-ethylene terephthalate) in asphalt concrete mixture lyser (2016)
as aggregate replacement. Waste Manag. Res. 23(4), 322–327 38. Uthaipan, N., Jarnthong, M., Peng, Z., Junhasavasdikul, B.,
(2005) Nakason, C., Thitithammawong, A.: Effects of cooling rates on
19. Zoorob, S.E., Suparma, L.B.: Laboratory design and investiga- crystallization behavior and melting characteristics of isotactic
tion of the properties of continuously graded Asphaltic con- polypropylene as neat and in the TPVs EPDM/PP and EOC/PP.
crete containing recycled plastics aggregate replacement (Plas- Polym. Test. 44, 101–111 (2015)
tiphalt). Cem. Concr. Compos. 22(4), 233–242 (2000) 39. Goli, V.S.N.S., Mohammad, A., Singh, D.N.: Application of
20. Lenkiewicz, Z., Webster, M.: Making waste work : a toolkit how municipal plastic waste as a manmade neo-construction mate-
to measure your waste part of making waste work : a toolkit rial: issues & wayforward. Resour. Conserv. Recycl. 161,
income countries (2017). 105008 (2020)
21. Tech Insider: Building houses out of plastic waste—YouTube 40. Awaja, F., Gilbert, M., Kelly, G., Fox, B., Pigram, P.J.: Adhe-
(2016). https://​www.​youtu​be.​com/​watch?v=​MXbrK​QE-​SOE. sion of polymers. Prog. Polym. Sci. 34(9), 948–968 (2009)
Accessed 24 Feb 2021. 41. Zahran, R.R.: Effect of sand addition on the tensile properties
22. UN Environment Programme: Plastic bricks in Kenya— of compression moulded sand/polyethylene composite system.
NZAMBI MATEE—Young Champion of the Earth 2020 for Mater. Lett. 34(3–6), 161–167 (1998)
Africa – YouTube (2020). https://​www.​youtu​be.​com/​watch?v=​ 42. Verbeek, C.J.R., Pickering, K.L.: Recent developments in poly-
QbZKP​4UAtL8. Accessed 24 Feb 2021. mer consolidated composites. J. Reinf. Plast. Compos. 26(16),
1607–1624 (2007)

13
3834 Waste and Biomass Valorization (2022) 13:3821–3834

43. Verbeek, C.J.R.: Effect of percolation on the mechanical prop- 47. Leon, L., Charles, R., Simpson, N.: Stress-strain behaviour of
erties of sand-filled polyethylene composites. J. Thermoplast. asphalt concrete in compression. Procedia Struct. Integr. 2,
Compos. Mater. 20(2), 137–149 (2007) 2913–2920 (2016)
44. Mehdipour, I., Khayat, K.H.: Effect of particle-size distribution 48. Iucolano, F., Liguori, B., Caputo, D., Colangelo, F., Cioffi, R.:
and specific surface area of different binder systems on pack- Recycled plastic aggregate in mortars composition: effect on
ing density and flow characteristics of cement paste, vol. 78. physical and mechanical properties. Mater. Des. 52, 916–922
Elsevier, New York (2017) (2013)
45. Dalhat, M.A., Wahhab, H.I.A.-A.: Cement-less and asphalt-less 49. Hassn, A., Aboufoul, M., Wu, Y., Dawson, A., Garcia, A.: Effect
concrete bounded by recycled plastic. Constr. Build. Mater. 119, of air voids content on thermal properties of asphalt mixtures.
206–214 (2016) Constr. Build. Mater. 115, 327–335 (2016)
46. Mitchell, J., Vandeperre, L., Dvorak, R., Kosior, E., Tarverdi,
K., Cheeseman, C.: Recycling disposable cups into paper plastic Publisher's Note Springer Nature remains neutral with regard to
composites. Waste Manag. 34(11), 2113–2119 (2014) jurisdictional claims in published maps and institutional affiliations.

13