Energy 189 (2019) 116353

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Transportation fuel from plastic wastes: Production, purification and

SI engine tests

a, *, Zso
Zsolt Dobo
fia Jakab a, Ga

bor Nagy a, Tama

s Koo

s a, Katalin Szemmelveisz a,

bor Mura
Ga
nszky b
a
ros, 3515, Miskolc, Hungary
University of Miskolc, Department of Combustion Technology and Thermal Energy, Egyetemva
b
ros, 3515, Miskolc, Hungary
University of Miskolc, Institute of Chemistry, Egyetemva

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

Article history: Thermal pyrolysis of HDPE, LDPE, PP, PS, PET and PUR plastic wastes were performed in a batch reactor
Received 20 August 2019 equipped with a temperature controlled reflux and the yields of pyrolysis oils and liquid fuels suitable for
Received in revised form transportation were determined. The gasoline fractions were investigated by GC-MS and tested in a
10 October 2019
traditional spark-ignition engine without any modifications or fuel blending. Fuel consumption and
Accepted 12 October 2019
Available online 14 October 2019
exhaust gas emission (NOx, CO) were measured and compared to a commercial fuel (gasoline, RON ¼ 95).
PS generated 70.5% gasoline range hydrocarbons from the solid waste, followed by PP with 42.1%, LDPE
with 40.8% and HDPE with 37.3%. Liquid product was not observed during PET pyrolysis. The engine was
Keywords:
Plastic
easily running with neat gasoline fractions obtained from HDPE, LDPE, PP and PS distillation; however, it
Waste was not able to start the engine with the neat dehydrated PUR distillate. The volumetric fuel con-
Pyrolysis sumption was reduced by 9.1e9.4% in case of PS compared to commercial gasoline (RON ¼ 95) but the
NOx emission was increased by 82e147%. Fuel consumption reduction was noticeable at HDPE, LDPE and
PP as well. PS gasoline decreased by 91e96%, while HDPE, LDPE and PP more likely increased the CO
emission of the engine compared to commercial gasoline.
© 2019 Elsevier Ltd. All rights reserved.

1. Introduction suitable for transportation as it can reduce the carbon footprint of

Plastic waste generation and accumulation became a serious
problem in the world. The global plastic waste generation reached
302 Mt in 2015, while the total plastic waste ever generated from
primary plastics had reached 5800 Mt [1]. By projecting current
global waste management trends to 2050, it is assumed that the
primary plastic waste generated will be around 26,000 Mt [1].
25 Mt of plastic ended up in waste stream in the EU during the year
of 2012 [2]. About 38% of this plastic waste was landfilled, 26% was
recycled while 36% was utilized for energy recovery [2]. The
amount of plastic waste slightly increased to 27.1 Mt in 2016, while
the landfilled amount dropped to 27.3%, the recycled portion
increased to 31.1% and the energy recovery also increased to 41.6%
[3]. As the amount of plastic waste continuously increases, some
alternative methods are being developed including pyrolysis,
which is a promising method to generate value-added liquid fuel
* Corresponding author.

).
E-mail address: zsoltdobo@gmail.com (Z. Dobo
transportation industry and optimize waste management towards
zero landfilling [4].
The pyrolytic products can be divided into a gas, liquid, and solid
fraction. Based on the process parameters the gas/liquid ratio can
be changed. The influencing factors of plastic waste pyrolysis were
investigated by several researchers such as temperature [5e7],
pressure [8,9], time [6,10], type of reactor [2,11], catalyst [12e14]
and plastic waste material used [2,15e22]. Among the pyrolysis
product description and properties, the studies utilizing the py-
rolysis oil for transportation were performed mostly in diesel en-
gines. Damodharan et al. [23] investigated the combined influence
of EGR and injection timing on the combustion, performance and
emission characteristics of a DI diesel engine fueled with neat WPO
under different EGR and timing rates. It was concluded that under
certain conditions (10% EGR and 25o CA bTDC) the WPO could be
effectively used in a compression-ignition engine. Mani et al. [24]
also investigated the influence of EGR when using plastic oil. 20%
exhaust gas recirculation was found as an optimum parameter
when using 100% plastic oil. In another study, Mani et al. [25]
concluded that WPO results in better performance than a blend of

https://doi.org/10.1016/j.energy.2019.116353
0360-5442/© 2019 Elsevier Ltd. All rights reserved.
2
et al. / Energy 189 (2019) 116353
Z. Dobo

WPO and diesel. Although WPO properties are summarized, compared to a commercial fuel with RON ¼ 95.
preparation of WPO nor plastic waste type were not reported.
Kaimal et al. [26] performed an experimental investigation to 2. Materials and methods
assess the effects of using plastic oil in a DI diesel engine and it was
found, that 25% plastic oil blended with diesel can be considered as The pyrolysis tests were performed using a 2.3 dm3 batch
an effective replacement for diesel fuel. Achyut et al. [27] investi- reactor equipped with an air cooled reflux, where both the tem-
gated a diesel engine performance using different blends of diesel perature and flow rate of the cooling air can be adjusted. The reflux
and PP pyrolysis oil. It was concluded, that the engine could operate was connected to a water-cooled heat exchanger in which the
with a maximum of 50% waste oil blended with diesel and the best molecules with a boiling temperature higher than ambient were
performance was observed up to 30%. condensed. The two-stage gas cooling enables better control of the
Although several diesel engine tests can be found in the litera- overall system and enhances the yields of gasoline range hydro-
ture, there is lack of information about spark ignition engine tests carbons. The liquid pyrolysis oil was collected in a product
using gasoline extracted from plastic pyrolysis oils and most of the container, while the remaining gases were sent to a flare. Fig. 1 il-
studies are focusing on oil composition. Buekens et al. [14] lustrates the measurement system.
reviewed the catalytic plastic pyrolysis for converting plastic The reactor was flushed with argon before each measurement to
wastes into gasoline-range hydrocarbons and it was concluded that eliminate the air from the system. The heating procedure started
fractional distillation of pyrolytic oil is preferred for gasoline range after the argon flush. The electric shell heater temperature was set
hydrocarbon separation. A similar conclusion was drawn by to 700  C. The pyrolysis runs were typically stopped when the
Miandad et al. [13] as post treatment of pyrolysis oil through temperature inside the reactor reached z540  C as the cracking
distillation and refining or blending with conventional fuel is reactions ended by this temperature, based on TG and DTG analysis
required to utilize the liquid oil as a transport fuel. Although oil of the plastic waste samples in an inert atmosphere. Fig. 2 shows
purification is matured from laboratory to industrial scale and is the results of TG and DTG analysis, where the temperature is nar-
widely used worldwide, this step makes the whole pyrolysis system rowed to a range of 300e700  C for better illustration. Based on the
more complex; therefore one direction of research is to investigate TG curves the solid residues are low in case of HDPE (1.6%), LDPE
pure pyrolysis oil blended with commercial fuel. Kumar et al. [28] (4.0%), PP (3.9%) and PS (4%), while PET (20.0%) and PUR (19.0%)
studied the effects of waste plastic oil blends on a multi cylinder shows significantly higher solid residue content. The main ther-
spark ignition engine. The engine operated without any modifica- mogravimetric data are summarized in Table 1.
tions with 0e20% pyrolysis oil in gasoline and it was noticed that Ultimate, proximate and HHV analyses of solid wastes were also
NOx emission increased and hydrocarbon emission decreased as performed based on EN ISO 16948:2015 (determination of total
the concentration of pyrolytic oil in gasoline increased. Measure- content of carbon, hydrogen and nitrogen), MSZ 24051:2001
ments were performed using the same engine in another study [29] (determination of sulfur content using high-temperature tube
with 15% pyrolysis oil and 5% ethanol blended with gasoline. By furnace combustion method) and MSZ 19954:1971 (determination
adding ethanol to the mixture the engine performance was of combustion heat and heating value) standards. The moisture and
improved and NOx emissions might be better controlled compared ash content was measured by gravimetric method following EN ISO
to the mixture without ethanol. Demirbas [30] investigated thermal 18134e3:2016 (determination of moisture content) and EN ISO
pyrolysis of PS, PP and PE and it was concluded, that as the liquid 18122:2016 (determination of ash content) standards. The volatile
products from plastic wastes contain gasoline (C8eC10) and light and fixed carbon contents were determined by TG analysis
gas oil (C10eC20) fractions, these products would be ideally suited
to further processing in a petrochemical refinery, with the gases
used directly as fuels. Kassargy et al. [31] performed catalytic py-
rolysis of PE, PP and PE/PP blend wastes and the collected oils were
separated into different fractions by distillation. The gasoline frac-
tion (distillation cut at 160  C) was tested in a DRESSER CFR engine
suitable for RON measurement. It was concluded that although the
gasoline obtained by 50% PP and 50% PE pyrolysis has a very low
content in aromatics, it presents a relatively high octane number
(RON ¼ 89.8). Faussone [4] presented two cases of industrial scale
transportation fuel production plants utilizing plastic waste as a
fuel source; one located in South East Asia, the other located in
eastern Europe. An important conclusion of the study is that
upgrading the pyrolysis oil into transportation fuels is necessary to
meet fuel standards and get 100% fuel from the waste source
suitable for transportation. It was noted, that if the obtained fuel is
not fully satisfying the fuel requirement then increasing the num-
ber of distilled cuts and subsequently blending them at different
proportions might be an easy route to meet the transportation fuel
parameters and better control the fuel properties.
The main goal of this paper is to investigate the thermal py-
rolysis of HDPE, LDPE, PP, PS, PET and PUR plastic wastes and
determine the yields of transportation fuels by performing atmo-
spheric distillation of the pyrolysis oil. Additionally, the extracted
gasoline fractions were investigated with GC-MS and tested in a
traditional spark-ignition engine without any modifications or
blending as there is lack of information such tests. Fuel consump-
tion and exhaust gas emission (NOx, CO) were measured and Fig. 1. Schematic illustration of the measurement system.

et al. / Energy 189 (2019) 116353
Z. Dobo 3

Fig. 2. TG and DTG analysis of plastic wastes used in this study. MOM Derivatograph C/PC was used for measurement with a heating rate of 10  C/min in a nitrogen atmosphere.

Table 1
Solid residue, maximum rate of volatile matter evolution (DTGmax) and temperature analysis was to be found as 3.5% (relative) for each component. The
of maximum volatile matter released (Tmax) of plastic waste materials used in this pyrolysis vapors exiting the reflux were cooled down to ambient
study, based on TG and DTG analysis in nitrogen atmosphere. temperature by utilizing a water cooled heat exchanger and the
Material Residue, % DTGmax,%/min Tmax, C condensed liquids were collected. The condensates were then
stored at room temperature in an air tight container until further
HDPE 1.6 28.6 496
LDPE 4.0 29.9 497 utilization or analysis.
PP 3.9 29.5 475 The collected pyrolysis oil was processed by atmospheric
PET 20.0 18.8 450 distillation to determine and separate the transportation fuel con-
PS 4.0 24.2 433
tent. The oil was heated in a spherical glass retort and the vapors
PUR 19.0 7.8 368
were condensed in a water-cooled condenser. The retort temper-
ature was measured by K-type thermocouple. The vials used for
collecting the extracts were changed at 200  C and 305  C tem-
performed with MOM Derivatograph C/PC (heating rate of 10  C/ peratures and the 20e200  C product was considered as the gas-
min, air atmosphere). The results are summarized in Table 2. oline yield and the 200e305  C content was considered as the
Although the sulfur content is very low in plastic samples, it can diesel yield. The collected gasoline from plastic waste pyrolysis oil
accumulate either in gas, liquid or solid fraction. Typically, the was tested in a traditional spark ignition engine (type: Honda GC-
volatile content of the plastics used in this study is high, while the 135; four stroke; single cylinder; air cooled; displacement:
moisture and ash contents are low. These results are in good 135 cm3) [32] equipped with a carburetor. The carburetor mixes the
agreement with the literature [2]. liquid fuel with air which is then delivered into the cylinder and
The maximum temperature of pyrolysis oil vapors exiting the combusted in the ignition period to provide rotational mechanical
reflux was set to 200  C by adjusting the cooling air flow rate and energy. The engine was connected to an electric generator acting as
temperature. The materials with a boiling point higher than the a load for the engine at 3000 rpm. Fuel consumption and NOx
reflux exiting temperature should flow back into the rector for emission were monitored and compared to reference measure-
further molecule scissoring. As a result, the amount of gasoline ments performed using commercial gasoline with a research oc-
range hydrocarbons in the pyrolysis oil should be increased, how- tane number of 95. Horiba PG-250 type flue gas analyzer was used
ever, the more evaporation/condensation cycles result in more for NOx measurement, while the CO content of flue gas was
C1eC4 products as well, appearing in the gas phase. Gas samples analyzed by gas chromatography. The 20e200  C distillates were
were taken from the gas stream after the second heat exchanger qualitatively analyzed by GC-MS (Agilent 7890 A GC; detector:
and the composition was analyzed using gas chromatography Agilent 5975C MSD; column: Agilent 19091J-433 HP-5 5% Phenyl
(model: Dani Master; TCD detector with 3 columns: Restek RT-Q- Methyl Siloxan 30 m  0.25 mm x 0.25 mm).
Bond 30 m  0.32 mm x 10 mm, Restek RT-Q-Bond
15 m  0.53 mm x 20 mm and Restek RT-Msieve 5A
30 m  0.53 mm x 50 mm; FID detector with 1 column: Rt-Alumina
BOND/Na2SO4 30 m  0.53 mm x 10 mm). The accuracy of the gas

Table 2
Ultimate, proximate and HHV analysis of the plastic waste materials used in this study including the standard deviation of the measurements.

Material Ultimate, % Proximate, % HHV, MJ/kg

a
C H N S O Moisture Volatile Fixed carbon Ash

HDPE 84.89 ± 0.65 13.95 ± 0.84 0.08 ± 0.02 0.02 ± 0.01 0.30 0.32 ± 0.04 98.35 ± 0.10 0.90 ± 0.14 0.43 ± 0.03 46.3 ± 1.08
LDPE 80.55 ± 0.18 13.21 ± 0.39 0.07 ± 0.02 0.02 ± 0.01 0.20 0.27 ± 0.03 91.16 ± 0.19 2.89 ± 0.22 5.68 ± 0.33 43.2 ± 0.53
PP 85.56 ± 0.60 13.66 ± 1.16 0.06 ± 0.02 0.03 ± 0.01 0.20 0.30 ± 0.04 93.42 ± 0.07 6.09 ± 0.02 0.19 ± 0.05 44.8 ± 0.17
PET 63.48 ± 0.49 4.65 ± 0.06 0.05 ± 0.01 <0.01 31.45 0.35 ± 0.09 83.81 ± 0.95 15.83 ± 1.20 0.01 ± 0.01 22.3 ± 0.65
PS 89.72 ± 0.10 8.26 ± 0.31 0.07 ± 0.01 <0.01 0.00 0.30 ± 0.06 93.36 ± 0.52 4.71 ± 0.60 1.63 ± 0.10 41.0 ± 0.12
PUR 66.33 ± 0.26 7.67 ± 0.01 5.69 ± 0.09 <0.01 18.79 1.34 ± 0.20 42.37 ± 0.54 56.11 ± 0.68 0.18 ± 0.11 28.4 ± 0.11
a
By difference.
4
et al. / Energy 189 (2019) 116353
Z. Dobo

3. Discussion

3.1. Plastic waste pyrolysis

The mass distribution of the six plastic waste materials pyro-

lyzed in a batch reactor is summarized in Table 3. The table also
contains the HHV analysis of the pyrolysis oils and the ultimate
analysis of the remained chars. The PP produced the highest oil
yield (83.4%) with the highest HHV value and the lowest gas yield
compared to the other materials used. Pyrolysis oil was not
observed in case of PET. Instead, around 0.2% solid powder was
found in the product container with a yellowish color. A similar
result was found in the study published by Sogancioglu et al. [5],
where zero percent of liquid product was presented in case of PET.
Sharuddin et al. [33] also reported similar experience during PET
pyrolysis and the study reveals that the yellowish substance might
be benzoic acid. Muddy and dark-green liquid was observed during
PUR pyrolysis. The high nitrogen concentration of PUR oil can be
explained with the solid PUR waste composition, where the ni-
trogen content is significant. The sulfur content of each oil sample
was below the detection limit in each case indicating that further
purification to remove sulfur is likely unnecessary. The HHV of each
pyrolysis oils are similar to the solid waste materials including the
HHV of the solid powder obtained during PET pyrolysis. Photos of
the pyrolysis oils can be seen in Fig. 3.
Based on the ultimate analysis of remained char it was
concluded, that the sulfur content of solid waste material is
enriched in the char portion. Sulfur was under the detection limit in
case of PET, PS and PUR both in the initial solid waste and the solid
product as well. The sulfur concentrations in solid pyrolysis prod-
ucts are about 60, 10 and 22 times greater than the concentrations
in initial plastic waste materials in case of HDPE, LDPE and PP,
respectively. This kind of sulfur enrichment in solids is favorable
indicating that the pyrolysis gas and oil presumably do not contain
a significant amount of sulfur. On the other side, the relatively high
carbon content of the solid residues gives the possibility of utilizing
them either by combustion or gasification as the heating values are
high, but such energetic utilization of HDPE, LDPE and PP chars
would likely need SO2 or H2S removal from generated gases to meet
emission requirements. As the sulfur content was below the
detection limit in case of PET, PS and PUR, the solid residues of these
waste plastics have a great potential for further energetic utilization
from the sulfur point of view.
The importance of energetic utilization of solid residues is also
supported by HHV analysis. The HHV was calculated based on the
following formula, where the combustible components are taken
Fig. 3. WPO samples obtained from the batch reactor equipped with temperature
controlled reflux. From left to right: HDPE, LDPE, PP, PS, PUR.
into account [34]:
HHV½MJ=kg ¼ 32:8 C þ 141:7 H þ 9:2 S (1)
where C, H and S are the elemental compositions of solid residues
in weight fraction. It can be concluded that the HHV is mostly set by
carbon content. As the solid residues of PP and PUR contained high
amount of carbon, the HHVs were the highest for these materials.
Consequently, the low carbon content in the case of LDPE resulted
in a low HHV as well.
The typical compositions of collected pyrolysis gases are sum-
marized in Table 4. High amount of C1eC4 hydrocarbons were
observed during HDPE (85.0%), LDPE (65.5%), PP (74.7%) and PS
(75.3%) pyrolysis, while high amount of CO and CO2 was observed
during PET and PUR pyrolysis. In case of PP, high monomeric yield
can be seen (38% C3H6), while PS provided high methane yield
(39.6%) and the concentration of hydrocarbons at PS significantly
decreases as the carbon number increases. C1eC3 is dominant at
LDPE and HDPE and smaller amount of C4 is present despite the fact
that random chain scissoring is the main decomposition mecha-
nism. Distribution of C4 between pyrolysis gas and liquid was
observed elucidating the smaller concentration of C4 in the pyrol-
ysis gases. The high amount of CO and CO2 during PET can be
explained with the oxygen atoms in the molecular structure as it
partially or fully oxidizes the combustible materials. A similar
oxidizing method can be observed during PUR pyrolysis. As the
initial material was porous, the argon flush of the reactor was not
efficient enough to eliminate all the air from the system; thus the
air stuck in the pores added more oxygen to the process resulting in
higher CO2 yield in the dry pyrolysis gases. H2S was not detected in
either gas which is a good indicator that sulfur was not presented in
the gas phase. This has the benefit that the gas can be burnt without

Table 3
Mass distribution, HHV and ultimate analyses of different pyrolysis products.

Unit HDPE LDPE PP PET PS PUR

Loaded plastic waste g 300 300 400 95 127 65

Char remained % 0.5 2.1 3.3 23.6 4.6 28.7
Pyrolysis oil % 73.9 58.8 83.4 0.2 79.2 36.9
Gas (by difference) % 25.6 39.1 13.3 76.2 16.2 34.4

Pyrolysis oil ultimate C % 85.37 ± 1.13 85.47 ± 0.37 84.97 ± 1.22 e 89.63 ± 2.47 74.50 ± 0.20
H % 14.14 ± 0.67 14.25 ± 0.32 14.46 ± 0.24 e 9.60 ± 0.43 8.04 ± 0.14
N % <0.01 <0.01 <0.01 e <0.01 9.11 ± 0.21
S % <0.01 <0.01 <0.01 e <0.01 <0.01
Pyrolysis oil HHV MJ/kg 45.8 ± 0.50 45.3 ± 0.64 47.1 ± 1.90 20.8 ± 0.34 41.5 ± 0.25 27.9 ± 0.87
Char ultimate C % 49.7 ± 0.912 32.1 ± 0.311 89.3 ± 0.264 62.9 ± 0.101 59.7 ± 1.181 82.5 ± 0.606
H % 2.77 ± 0.047 1.71 ± 0.048 1.98 ± 0.089 4.88 ± 0.081 2.27 ± 0.020 4.36 ± 0.071
N % 0.47 ± 0.025 0.60 ± 0.063 0.35 ± 0.017 0.11 ± 0.002 0.14 ± 0.015 7.08 ± 0.083
S % 1.82 ± 0.021 0.25 ± 0.041 0.66 ± 0.010 <0.01 <0.01 <0.01
Char HHVa MJ/kg 20.3 12.9 31.9 27.4 22.6 33.0
a
Calculated value based on ultimate analysis.

et al. / Energy 189 (2019) 116353
Z. Dobo 5

Table 4
Typical composition of collected dry pyrolysis gases and the results of calculated HHV based on the HHV of individual components.

Component Limit of detection (LOD), ppm Concentration, vol%

HDPE LDPE PP PET PS PUR

CO2 440 1.6 1.37 1.37 35.0 6.52 76.1

H2S 1070 <0.107 <0.107 <0.107 <0.107 <0.107 <0.107
H2 5000 1.24 12.3 <0.5 <0.5 11.5 3.85
CO 920 0.82 0.58 0.73 40.0 6.48 1.80
CH4 240 16.8 15.3 11.9 1.74 39.6 4.74
C2H6 2 19.4 13.7 15.0 0.05 11.8 2.43
C2H4 2 16.5 15.4 2.82 1.62 12.0 0.48
C3H8 2 12.1 8.00 6.0 0.03 4.33 1.13
C3H6 1 17.9 9.77 38.0 0.07 6.56 5.29
C2H2 1 <0.0001 <0.0001 <0.0001 0.04 <0.0001 <0.0001
C4H10 (iso) 4 0.15 1.94 0.36 <0.0004 0.12 0.03
C4H10 (norm) 1 0.11 1.43 0.25 <0.0001 0.09 0.02
C4H8 1 2.08 0.00 0.41 0.01 0.78 0.11
Other 11.3 20.2 23.2 21.4 0.2 4.0
HHV, kJ/m3 59.4 46.0 56.3 6.6 43.2 10.2

any sulfur removal or post-combustion treatment. It is interesting

to note that PP and PET pyrolysis gases did not contain H2. Based on
the composition of different pyrolysis gases the HHV values were
calculated. In case of high C1eC4 content (HDPE, LDPE, PP and PS)
the HHV is very high, while the low C1eC4 hydrocarbons and high
CO and CO2 content give lower HHV for the PUR and PET gases.

3.2. Pyrolysis oil distillation

The collected pyrolysis oils were separated into three temper-

ature ranges by atmospheric distillation. The yields of trans-
portation fuels from pyrolysis oils are summarized in Table 5. Fig. 4. Photos of distillation products in 20e200  C range. From left to right: HDPE,
Distillation was not performed in case of PET as no oil was observed LDPE, PP, PS, PUR.

during PET pyrolysis. WPO obtained from HDPE, LDPE and PS

contained high amount of transportation fuels and the gasoline
range hydrocarbons are dominating. PS pyrolysis oil had 89% of Table 6
liquid fraction in the 20e200  C range providing the highest gas- Transportation fuel yields from solid plastic waste.
oline yield. PP pyrolysis oil contained less amount of total trans-
Temperature range,  C Yield, g/kgwaste
portation fuels (80.2%) compared to HDPE, LDPE and PS. Although
the 20e200  C fraction was 63.9% during PUR pyrolysis oil distil- HDPE LDPE PP PET PS PUR

lation, it contained a significant amount of water as seen in Fig. 4 as 20e200 373 408 421 e 705 235
well. Thus, the water content of the liquid was removed by simple 200e305 305 152 248 e 11 32
Total 678 560 669 e 716 267
gravimetric separation and only the upper part was considered for
possible transportation fuel.
The yields of transportation fuels from solid plastic wastes are
summarized in Table 6. The yields were calculated by utilizing both The 20e200  C distillation fractions were qualitatively
the results seen in Tables 3 and 5. PS provided the highest trans- described by GC-MS. The chromatograms are shown in Fig. 5, while
portation fuel yield and gasoline yield as well, similarly to the re- the different components and their properties are listed in Table 8.
sults found in Table 5. It can be clearly seen that the gasoline range HDPE and LDPE gave very similar results, only 2,4-dimethyl-1-
hydrocarbons are dominating, and the gasoline to diesel ratio in heptene (peak #12) was the difference as it was not detected in
each case changes. While HDPE provided similar amounts of gas- the case of LDPE. Both HDPE and LDPE 20e200  C fractions con-
oline and diesel, PS pyrolysis gave almost gasoline range hydro- tained mostly paraffins and olefins. Typically, paraffins have lower
carbons. The 23.5% 20e200  C fraction obtained at PUR oil octane numbers compared with the olefins [35]. For example, RONs
distillation includes the water portion, which is roughly 50% by are 62, 25, 0 and 0 for n-pentane, n-hexane, n-heptane and n-oc-
volume of the distillate. tane, respectively [36]. On the other side, RONs are 88, 76, 55 and 29
for 1-pentene, 1-hexene, 1-heptene and 1-octane, respectively [36].
The gasoline obtained from PP WPO contains mostly branched
hydrocarbon chains, while high aromatic content was observed in
Table 5
Transportation fuel yields from pyrolysis oil. PS gasoline presumably giving high octane ratings for the liquid. For
example, RONs are 101, 121, 103 and 108 for benzene, toluene,
Temperature range,  C Yield, g/kgoil
styrene and ethylbenzene, respectively. Although only hydrocar-
HDPE LDPE PP PET PS PUR bons were identified in HDPE, LDPE, PP and PS gasoline fractions,
20e200 504 695 505 e 890 639 PUR 20e200  C fraction is a mixture of other hydrocarbons as well
200e305 413 259 297 e 14 90 containing O, N and Cl elements. Acetone (contains O) blended with
Total 917 954 802 e 904 729 gasoline was successfully combusted in a spark-ignition engine
6
et al. / Energy 189 (2019) 116353
Z. Dobo
Fig. 5. GC-MS analysis of the 20e200  C extracts: chromatograms.
leading to a boost in both fuel conversion and engine performance
as reported by Elfasakhany [37]. On the other side, 1,2-
dichloropropane was detected in the 20e200  C fraction of PUR
WPO. The presence of Cl degrades the quality of the pyrolysis
product and such impurities may restrict the use of the pyrolysis oil
[38]. The impurities of the PUR gasoline makes the utilization more
complex. However, after purification and/or blending with other
fuels may give sufficient solution. Based on the PUR gasoline den-
sity seen in Table 7, the liquid contains mostly nitrogenated hy-
drocarbons as these components have densities similar to the
measured value.
Table 7
Densities of neat 20e200  C distillates measured at 25  C.
Material HDPE LDPE
Density (20e200  C fraction), g/cm3 750 762
3.3. Engine performance test
Engine tests were performed to investigate the fuel consump-
tions and emissions and compare to commercial gasoline with
RON ¼ 95. The water content of 20e200  C fraction obtained from
PUR pyrolysis was removed and the remained part was fed into the
engine. As the engine was not able to run on this liquid, the PUR
was excluded from the investigation list of materials in terms of
engine tests. Additional purification steps are necessary to remove
the hydrocarbons containing N and Cl, which can be done by
optimizing the distillation cuts. More research is needed to evaluate
the possibilities of using gasoline made from PUR WPO.
The volumetric fuel consumption change in the case of HDPE,
LDPE, PP and PS gasoline is shown in Fig. 6. It can be seen, that the
PP PET PS PUR
Lower (H2O solution) Upper
744 e 886 1006 990

et al. / Energy 189 (2019) 116353
Z. Dobo 7

Table 8
GC-MS analysis of the 20e200  C extracts: major compounds and properties.

Peak Name Formula Retention time, min Molecular weight, g/mole Boiling point,  C

HDPE, LDPE 1 2-butene C4H8 1.580 56.11 1

2 1-pentene C5H10 1.692 70.13 30
3 pentane C5H12 1.708 72.15 36.1
4 1-hexene C6H12 1.976 84.16 63
5 hexane C6H14 2.010 86.18 68
6 2,4-hexadiene C6H10 2.339 82.14 80.0
7 1-heptene C7H14 2.623 98.19 93.6
8 heptane C7H16 2.698 100.21 98.4
9 1-methyl-cyclohexene C7H12 3.543 96.17 110.3
10 1-octene C8H16 3.800 112.24 121
11 octane C8H18 3.917 114.23 125.6
12 2,4-dimethyl-1-heptene C9H18 4.564 126.24 134
13 1-nonene C9H18 5.377 126.24 147
14 nonane C9H20 5.511 128.26 150.8
15 1-decene C10H20 7.046 140.27 170.6
16 decane C10H22 7.185 142.29 174.1
17 1-undecene C11H22 8.656 154.29 192.7
18 undecane C11H24 8.785 156.31 195.9
19 1-dodecene C12H24 10.164 168.32 213
20 dodecane C12H26 10.282 170.33 216.2
21 1-tridecene C13H26 11.577 182.35 232.8
22 tridecane C13H28 11.684 184.37 234.5
PP 1 2-butene C4H8 1.580 56.11 1
2 pentane C5H12 1.708 72.15 36.1
3 2,3-dimethylbutane C6H14 1.890 86.18 57.9
4 2-methyl-1-pentene C6H12 1.976 84.16 62.1
5 2,4-hexadiene C6H10 2.232 82.14 80.0
6 2,4-dimethyl-pentene C7H14 2.291 98.19 81
7 2,4-dimethyl-1,4-pentadiene C7H12 2.350 96.17 89.6
8 4-methylheptane C8H18 3.468 114.23 117
9 2-methyl-1-heptene C8H16 3.741 112.21 118.9
10 2,4-dimethyl-1-heptene C9H18 4.596 126.24 134
11 1,3,5-trimethylcyclohexane C9H18 4.821 126.24 140.6
PS 1 benzene C6H6 2.425 78.11 80.1
2 toluene C7H8 3.516 92.14 110.6
3 ethylbenzene C8H10 4.933 106.17 136
4 styrene C8H8 5.468 104.15 145
5 cumene C9H12 5.971 120.19 152.4
6 propylbenzene C9H12 6.458 120.19 159
7 a-methylstyrene C9H10 6.950 118.18 166
PUR 1 acetone C3H6O 1.697 58.08 56.1
2 allyl alcohol C3H6O 1.821 58.08 96
3 benzene C6H6 2.425 78.11 80.1
4 1,2-dichloropropane C3H6Cl2 2.724 112.99 96
5 toluene C7H8 3.521 92.14 110.6
6 2,5-dimethyl-1,4-dioxane C6H12O2 3.987 116.16 134.4
7 2,6-dimethyl-1,4-dioxane C6H12O2 4.393 116.16 135.4
8 ethylbenzene C8H10 4.928 106.17 136
9 xylene C8H10 5.062 106.17 144
10 styrene C8H8 5.404 104.15 145
11 cumene C9H12 5.966 120.19 152.4
12 propylbenzene C9H12 6.458 120.19 159
13 1-ethyl-2-methylbenzene C9H12 6.618 120.19 164.1
14 1,3,5-trimethylbenzene C9H12 6.704 120.19 165
15 aniline C6H7N 6.902 93.13 184.1
16 benzonitrile C7H5N 6.998 103.04 191
17 toluidin C7H9N 8.415 107.15 200

fuel consumption was lower in each case compared to the reference
measurement. PS showed the highest fuel consumption decrease
with 9.1% and 9.4% at idling and under load, respectively. HDPE and
LDPE also showed extreme fuel consumption reduction compared
to the reference despite the fact, that components with lower oc-
tane numbers were identified in the liquid. Additionally, the overall
operation of the engine with HDPE gasoline was smoother. A less
significant change was observed during PP pyrolysis and the engine
behaved similarly compared to the reference measurement.
The NOx emission of the engine (Fig. 7) was higher in each case
except PP, where a slight decrease was noticeable. While PS showed
extreme low fuel consumption, its NOx emission was the highest
(more than doubled under load conditions). As there was minor
fuel consumption change at PP tests, a minor change is observed in
terms of NOx emission as well. Consequently, both HDPE and LDPE
generated a significantly higher amount of NOx as their fuel con-
sumption was lower.
In terms of CO emissions (Fig. 8), PP gasoline generated the
highest CO. Extreme low CO concentration was measured when PS
gasoline was applied, the CO decreased 96% when idling and 91%
under load compared to the reference measurement. This indicates
better combustion performance compared to the other fuels used
in this study. HDPE and LDPE gasoline likely increased the CO
emission except LDPE gasoline test under engine idling, where a
8
et al. / Energy 189 (2019) 116353
Z. Dobo
Fig. 6. Fuel consumption change of the engine using various plastic gasoline compared
to reference measurements. Typical fuel consumption of reference measurement when
idling (0 W) and under load was 0.51 l/h and 0.61 l/h, respectively.
Fig. 7. NOx emission change of the engine using various plastic gasoline compared to
reference measurements. Typical NOx emission of reference measurement when idling
(0 W) was 46.3 ppm and the NOx emission at 500 W electrical load was 77.9 ppm.
slight decrease was observed. Overall, the CO emissions increased
during HDPE, LDPE and PP tests, while PS significantly decreased
the CO.
4. Conclusion
Pyrolysis of HDPE, LDPE, PP, PS, PET and PUR plastic wastes was
investigated. The pyrolysis runs were performed in a batch reactor
equipped with temperature-controlled reflux to enhance the gas-
oline content of pyrolysis oils. The temperature of the cracked
plastic oil vapor exiting the reflux was not exceeding 200  C. The
pyrolysis oils were upgraded by atmospheric distillation to separate
Fig. 8. CO emission change of the engine using various plastic gasoline compared to
reference measurements. Typical CO emission of reference measurement when idling
(0 W) was 3.61 vol% and the CO emission at 500 W electrical load was 2.78 vol%.
the transportation fuels. The gasoline yield was 37.3%, 40.8%, 42.1%
and 70.5%, while the diesel yield was 30.5%, 15.2%, 24.8% and 1.1%
for HDPE, LDPE, PP and PS, respectively. No liquid product was
observed during PET pyrolysis.
The neat gasoline fractions collected during atmospheric
distillation were investigated by GC-MS and tested in a SI engine.
Stable operation was observed when HDPE, LDPE, PP and PS gas-
oline were applied, while the engine was not able to run with the
neat PUR gasoline fraction (after dehydration). Further purification
of PUR gasoline may give fuel sufficient enough to power the en-
gine. The volumetric fuel consumption was lower by 9.1e9.4%
when the PS gasoline was used compared to the reference mea-
surement (commercial fuel, RON ¼ 95). HDPE gasoline reduced the
fuel consumption by 6.1e7.8%, LDPE by 5.6e6.3% and PP by
1.2e2.0%, compared to the reference measurement. The highest
NOx emission was measured during PS gasoline test when 82e147%
more NOx was generated compared to the reference measurement.
On the other side, CO emission decreased by 91e96% when PS
gasoline was fed. PP showed 3.7e5.4% NOx reduction and this was
the only case when the NOx concentration was lower compared to
the reference measurement. Although significantly lower CO
emission was measured during PS gasoline test, HDPE, LDPE and PP
gasoline likely generated more CO. The results show that the py-
rolysis of different plastic wastes is a viable method to reduce the
accumulation of plastics in landfills and provide value-added liquid
transportation fuels.
Acknowledgment
This research was supported by the European Union and the
Hungarian State, co-financed by the Regional Development Fund in
the framework of the GINOP-2.3.4-15-2016-00004 project, aimed
to promote the cooperation between the higher education and
industry.”
References
[1] Geyer R, Jambeck JR, Law KL. Production, use, and fate of all plastics ever
made. Sci. Adv. 2017;3:1e5.
[2] Sharuddin SDA, Abnisa F, Daud WMAW, Aroua MK. A review on pyrolysis of

et al. / Energy 189 (2019) 116353
Z. Dobo
plastic wastes. Energy Convers Manag 2016;115:308e26.
[3] Plastics e the facts 2018. An analysis of European plastics production, demand
and waste data. https://www.plasticseurope.org/application/files/6315/4510/
9658/Plastics_the_facts_2018_AF_web.pdf.
[4] Faussone Gian Claudio. Transportation fuel from plastic: two cases of study.
Waste Manag 2017;73:416e23.
[5] Sogancioglu M, Ahmetli G, Yel E. A comparative study on waste plastics py-
rolysis liquid products quantity and energy recovery potential. Energy Pro-
cedia 2017;118:221e6.

pez A, de Marco I, Caballero BM, Laresgoiti MF, Adrados A. Influence of time
[6] Lo
and temperature on pyrolysis of plastic wastes in a semi-batch reactor. Chem
Eng J 2011;173:62e71.
[7] Kalargaris I, Tian G, Gu S. The utilization of oils produced from plastic waste at
different pyrolysis temperatures in a DI diesel engine. Energy 2017;131:
179e85.
[8] Murata K, Sato K, Sakata Y. Effect of pressure on thermal degradation of
polyethylene. J Anal Appl Pyrolysis 2004;71:569e89.
[9] Miranda R, Yang J, Roy C, Vasile C. Vacuum pyrolysis of commingled plastics
containing PVC I. kinetic study. Polym Degrad Stab 2001;72:469e91.
[10] Diaz Silvarrey LS, Phan AN. Kinetic study of municipal plastic waste. Int J
Hydrogen Energy 2016;41:16352e64.
[11] Butler E, Devlin G, McDonnell K. Waste polyolefins to liquid fuels via pyrol-
ysis: review of commercial state-of-the-art and recent laboratory research.
Waste and Biomass Valorization 2011;2:227e55.
[12] Miandad R, Barakat MA, Aburiazaiza AS, Rehan M, Nizami AS. Catalytic py-
rolysis of plastic waste: a review. Process Saf Environ Prot 2016;102:822e38.
[13] Miandad R, Barakat MA, Rehan M, Aburiazaiza AS, Ismail IMI, Nizami AS.
Plastic waste to liquid oil through catalytic pyrolysis using natural and syn-
thetic zeolite catalysts. Waste Manag 2017;69:66e78.
[14] Buekens AG, Huang H. Catalytic plastics cracking for recovery of gasoline-
range hydrocarbons from municipal plastic wastes. Resour Conserv Recycl
1998;23:163e81.
[15] Kunwar B, Cheng HN, Chandrashekaran SR, Sharma BK. Plastics to fuel: a
review. Renew Sustain Energy Rev 2016;54:421e8.
[16] Kumar S, Singh RK. Recovery of hydrocarbon liquid from waste high density
polyethylene by thermal pyrolysis. Braz J Chem Eng 2011;28:659e67.
[17] Miranda R, Yang J, Roy C, Vasile C. Vacuum pyrolysis of commingled plastics
containing PVC I. kinetic study. Polym Degrad Stab 2001;72:469e91.
[18] Kunwar B, Cheng HN, Chandrashekaran SR, Sharma BK. Plastics to fuel: a
review. Renew Sustain Energy Rev 2016;54:421e8.
[19] Seth D, Sarkar A. Thermal pyrolysis of polypropylene: effect of reflux-
condenser on the molecular weight distribution of products. Chem Eng Sci
2004;59:2433e45.
[20] Sogancioglu M, Yel E, Ahmetli G. Investigation of the effect of polystyrene (PS)
waste washing process and pyrolysis temperature on (PS) pyrolysis product
quality. Energy Procedia 2017;118:189e94.
[21] Sinn Hansjo €rg, Walter Kaminsky, Janning Jo € rg. Processing of plastic waste and
scrap tires into chemical raw materials, especially by pyrolysis. Angew Chem
Int Ed 1976;15:660e72.
[22] Font R, Fullana A, Caballero JA, Candela J, García A. Pyrolysis study of
9
polyurethane. J Anal Appl Pyrolysis 2001;58e59:63e77.
[23] Damodharan D, Sathiyagnanam AP, Rana D, Rajesh Kumar B, Saravanan S.
Combined influence of injection timing and EGR on combustion, performance
and emissions of DI diesel engine fueled with neat waste plastic oil. Energy
Convers Manag 2018;161:294e305.
[24] Mani M, Nagarajan G, Sampath S. An experimental investigation on a DI diesel
engine using waste plastic oil with exhaust gas recirculation. Fuel 2010;89:
1826e32.
[25] Mani M, Nagarajan G, Sampath S. Characterization and effect of using waste
plastic oil and diesel fuel blends in compression ignition engine. Energy
2011;36:212e9.
[26] Kaimal Viswanath K, Vijayabalan P. A detailed study of combustion charac-
teristics of a DI diesel engine using waste plastic oil and its blends. Energy
Convers Manag 2015;105:951e6.
[27] Panda Achyut K, Murugan S, Singh RK. Performance and emission charac-
teristics of diesel fuel produced from waste plastic oil obtained by catalytic
pyrolysis of waste polypropylene. Energy Sources, Part A Recovery, Util En-
viron Eff 2016;38:577e85.
[28] Vijaya Kumar Kareddula, Kumar Puli Ravi. Effects of waste plastic oil blends
on a multi cylinder spark ignition engine. In: MATEC Web of Conferences,
108; 2017.
[29] Kumar Kareddula Vijaya, Kumar Puli Ravi. Influence of plastic oil with ethanol
gasoline blending on multi cylinder spark ignition engine. Alexandria Eng J
2018;57:2585e9.
[30] Demirbas Ayhan. Pyrolysis of municipal plastic wastes for recovery of
gasoline-range hydrocarbons. J Anal Appl Pyrolysis 2004;72:97e102.
[31] Kassargy Chantal, Awad Sary, Burnens Gae €tan, Kahine Khalil,
Tazerout Mohand. Gasoline and diesel-like fuel production by continuous
catalytic pyrolysis of waste polyethylene and polypropylene mixtures over
USY zeolite. Fuel 2018;224:764e73.
[32] GC135 manual: https://engines.honda.com/support-and-service/owners-
manuals/gc135.
[33] Sharuddin Shafferina Dayana Anuar, Abnisa Faisal, Daud Wan Mohd Ashri
Wan, Aroua Mohamed Kheireddine. Energy recovery from pyrolysis of plastic
waste: study on non-recycled plastics (NRP) data as the real measure of
plastic waste. Energy Convers Manag 2017;148:925e34.
[34] Lackner Maximilian, Arp
a
d B. Palota

s, franz winter. Combustion. Wiley; 2013.
978-3-527-33376-9.
[35] Cheng Yu, Hu Erjiang, Deng Fuquan, Yang Feiyu, Zhang Yingjia,
Tang Chenglong, Huang Zuohua. Experimental and kinetic comparative study
on ignition characteristics of 1-pentene and n-pentane. Fuel 2016;172:
263e72.
[36] William L. Kubic. A group contribution method for estimating cetane and
octane numbers. Report. 2016.
[37] Elfasakhany Ashraf. Performance and emissions analysis on using acetone-
gasoline fuel blends in spark-ignition engine. Eng Sci Technol, an Int. J.
2016;19:1224e32.
[38] Lu Peng, Huang Qunxing, Bourtsalas ACThanos, Themelis Nickolas J, Chi Yong,
Yan Jianhua. Review on fate of chlorine during thermal processing of solid
wastes. J Environ Sci 2019;78:13e28.