6

Development of a Process for the

Continuous Conversion of Waste
Plastics Mixtures to Fuel
TAKAO MASUDA AND TERUOKI TAGO
Division of Chemical Process Engineering, Graduate School of Engineering, Hokkaido
University, N13 W8 Kita-ku, Sapporo 060–8628, Japan

1 INTRODUCTION

The amount of waste plastics discarded each year is constantly increasing and is causing
serious pollution-related problems worldwide. Chemically recycling such waste might
provide cheap and abundant sources of useful chemicals and energy. Among a number
of potential approaches to recycling, chemical approaches, which convert waste plastics
to useful hydrocarbons, have been recognized as being of particular promise.
Presently, about half of the waste generated in Japan is derived from factories as
specific plastics, such as polyolefins, polystyrene (PS) [1–4], poly(ethylene terephthalate)
(PET) [5–6], and polyvinylchloride (PVC) [2, 3, 7–11]. To date, several studies have
focused on the individual degradation of these waste plastics. The remaining half of the
waste plastics in Japan is generated from households in the form of plastic mixtures and
waste-containing kitchen refuse. PVCs can be removed from such waste mixtures by
utilizing PVCs with a higher density than that of other plastics. However, there remain
certain unresolved issues, such as the lack of an efficient continuous chemical recycling
process for application to a mixture of remaining waste plastics. Specifically, it remains
problematic that each type of plastics has different pyrolytic properties, as is the fact
that large amount of sublimate materials, such as terephthalic acid (sublimation point
∼300◦ C), are produced from the degradation of some plastics, such as PET. Terephthalic
acid is precipitated as a hard, solid body around valves and pipelines, i.e. at locations
where the temperature is below 300◦ C.
Waste plastics are primarily generated by small-to-medium-size enterprises and house-
holds. Therefore, it is preferable to build chemical plants for the recycling of waste

Feedstock Recycling and Pyrolysis of Waste Plastics: Converting Waste Plastics into Diesel and Other Fuels Edited by
J. Scheirs and W. Kaminsky  2006 John Wiley & Sons, Ltd ISBN: 0-470-02152-7
162 T. MASUDA AND T. TAGO
plastics locally, i.e. in each self-governing region or prefecture. However, in this respect,
it should be noted that it is costly to use nitrogen or hydrogen as the carrier gas.
Because steam is cheaper and easy to handle, it is a potential candidate carrier gas in
such plants.
The chemical recycling of waste plastics consists of two processes: the first is the
degradation of waste plastics for the production of heavy oils, and the second is a catalytic
cracking process that converts the heavy oils into useful hydrocarbons. To achieve these
recycling goals, it remains necessary to develop efficient chemical recycling processes
that can operate in a steam atmosphere.
In this chapter, a chemical recycling method for the mixtures of waste plastics is
described. First, a number of possibilities are examined that would realize the formation
of pyrolytic properties with a uniform reaction rate among oxygen-containing plastics;
furthermore, it would be advantageous to achieve the degradation of such plastics with
the production of a minimal carbonaceous residue [5–6]. To this end, a new pyrolytic
reactor [12] has been proposed that has been shown to achieve high hold-up, a high level
of heat transfer, and good contact between melted plastics and steam, thus accelerating
hydrolysis. Using this reactor, a mixture of waste plastics can be degraded and further
decomposed over an FeOOH catalyst in a steam atmosphere. Second, chemical processes
are described here that can achieve a catalytic cracking of the heavy oil, achieved by
the pyrolysis and hydrolysis of the waste plastics, and producing useful fuels [13–15].
The catalyst used in the recycling process must exhibit high and stable activity during
the catalytic cracking of the heavy oil in a steam atmosphere [16]. Finally, the rationale
for and usefulness of the proposed chemical recycling method are both validated by the
experimental results obtained in a pilot-scale plant.

2 RECOVERY OF HEAVY OIL FROM WASTE PLASTIC

The waste plastics generated from households are in the form of plastic mixtures; among
such mixtures, the amount of PET generated as a household waste has rapidly increased
as the production of PET bottles has increased. PET is formed by the ester bonding
of terephthalic acid and ethylene glycol. When PET is heated above 380◦ C, pyrolysis
suddenly starts, yielding oxygen-containing hydrocarbons and a significant amount of
carbonaceous residue [17–19]. Accordingly, in order to develop a recycling process for
the conversion of waste plastics into useful liquid hydrocarbons, a method for the degrada-
tion of a mixture of waste plastics without the production of such a carbonaceous residue
is required.

2.1 DEGRADATION OF VARIOUS PLASTICS

The relationship between the rate of PET degradation and the molar fraction of steam
in the carrier gas was investigated using a thermogravimetric apparatus equipped with
a thermobalance. Figure 6.1 shows the remarkable change in the fraction of unreacted
PET as the molar fraction of steam in the carrier gas was varied. When a pure nitro-
gen stream (steam molar fraction = 0%) was used, the degradation of PET was initiated
CONTINUOUS PROCESS FOR FUEL PRODUCTION 163
1.0
Heating rate:
5°C min−1
0.8 Molar fraction

Unreacted fraction, 1−x

of steam = 0.7
0.6 0.5
0
0.3
0.4

0.15
0.2

0
200 300 400 500 600
Reaction temperature T / °C

Figure 6.1 Change in the fraction of unreacted PET in a steam atmosphere. Heating
rate: 5◦ C min−1 . (Reproduced with permission from Elsevier)

at about 380◦ C and was terminated at about 530◦ C, leaving about 16% carbonaceous
residue. When the carrier gas contained steam, the fraction of carbonaceous residue was
drastically reduced from 16% (100% nitrogen) to below 2% (70 mol% steam). Further-
more, the initiation temperature for pyrolysis decreased by about 30◦ C. Figure 6.2 shows
the relationship between the molar fraction of steam in the carrier gas and the amount
of carbonaceous residue remaining at 520◦ C. The circles, triangles, and squares represent
the results obtained at heating rates of 2.5, 5.0, and 10◦ C min−1 , respectively. All data lie
on a single curve, suggesting that the amount of carbonaceous residue depends only on
the partial pressure of steam, and not on the rate of increasing temperature. The amount
of residue thus decreases with increases in the partial pressure of steam. Less than 1%
residue was found to remain under 100% steam carrier gas conditions.

20
Plastic: PET Heating rate
2.5°C min−1
15 5.0°C min−1
Yield of residue / wt%

10°C min−1

10

0
0 20 40 60 80 100
Fraction of steam in the carrier gas
/ mol%

Figure 6.2 Relationship between the molar fraction of steam in the carrier gas and the
yield of residue at 520◦ C. (Reproduced with permission from Elsevier)
164 T. MASUDA AND T. TAGO

C=O C=C C−O −(CH2)n−

(ester) (aromatics) (ester) (s −bond)

Unreacted PET
A.U.

PET at 60%
conversion in nitrogen

PET at 70%
conversion in steam
2000 1600 1200 800 400
Wave number /cm−1

Figure 6.3 FTIR spectra of unreacted PET and of residues obtained from PET at 60%
conversion in both nitrogen and steam (70 mol%) atmospheres. (Reproduced with
permission from Elsevier)

Figure 6.3 shows the FT-IR spectra of residues obtained in nitrogen at a PET conversion
of 60%, and in a steam atmosphere at a conversion of 70%. This figure also shows the
spectrum for unreacted PET. PET is formed by the ester bonding of terephthalic acid
and ethylene glycol. In the case of thermal pyrolysis in nitrogen, the dehydration of the
chemical bonds and the random scission of the main chain of PET occurred, resulting
in a reduction in the peak strengths corresponding to bonds involving the oxygen atom
(C–O, C=O) and σ-bonds, leaving a hydrogen-poor residue. On the other hand, when
the degradation of PET was conducted in the presence of steam, the C–O bond and σ-
bond were preferentially weakened, and a peak corresponding to free aromatics (C=C)
appeared. These aromatics were considered to be terephthalic acids at the ends of the main
chain of PET. Hence, steam was found to accelerate the hydrolysis of PET, producing
monomers of PET such as terephthalic acid.
Figure 6.4(a, b) shows the thermogravimetric curves (TG-curves) for plastics at a heat-
ing rate of 5◦ C min−1 in nitrogen and in steam, respectively. Seven types of plastics were
used in this series: polyethylene (PE), polypropylene (PP), polycarbonate (PC), poly-
butyleneterephthalate (PBT), polystyrene (PS), nylon-6 (N6) and nylon-6,6 (N6,6). The
curves for the degradation of PET are also shown in the figure for comparison. The TG
curves were remarkably different among the plastics in a nitrogen stream, due to differ-
ences in the main chains of the plastics. Furthermore, polyester resins, such as PC, PBT,
and PET, yielded large amounts of carbonaceous residue at 527◦ C, due to dehydration
during the degradation in nitrogen. In contrast, in a steam atmosphere, the TG curves of
PC, PBT, and PET were remarkably altered by shifts to lower temperature regions and
by reducing the amount of carbonaceous residue, as compared with those in a nitrogen
atmosphere. Accordingly, it was concluded that the dominant mechanism of degradation
of polyester resins changed from thermal pyrolysis to hydrolysis by introducing steam as
the carrier gas.
Under the nitrogen steam conditions, PC showed the lowest degradation rate. On the
other hand, the plastic resin with the lowest degradation rate was PE in a steam atmo-
sphere. Therefore, the size of a degradation reactor could be designed by considering the
degradation rate of PE alone.
CONTINUOUS PROCESS FOR FUEL PRODUCTION 165
1.0

Fraction of unreacted resin /-

0.8 PS
N6,6 N6
0.6 PBT
PE
0.4
PC
Carrier gas: N2 PP
0.2 PET
Heating rate: 5 °C•min−1
0
200 300 400 500 600
Temperature / °C
(a)

1.0
Fraction of unreacted resin /-

PS PE
0.8
N6
0.6 PET
PP
0.4 PBT PC
Carrier gas: steam N6,6
0.2
Heating rate: 5 °C•min−1
0
200 300 400 500 600
Temperature / °C
(b)

Figure 6.4 Change in the fraction of unreacted plastics used in this study at a heating
rate of 5◦ C min−1 : (a) in nitrogen; (b) in steam. (Reproduced with permission from
Elsevier)

2.2 CATALYTIC CRACKING OF WASTE PLASTICS WITHOUT RESIDUE

The degradation of PET in a steam atmosphere is effective at reducing carbonaceous

residue, and this type of degradation produces terephthalic acid and oxygen-containing
compounds. However, terephthalic acid is precipitated as a hard solid body around valves
and pipelines, because this compound is a sublimate material (sublimation point ∼300◦ C).
Therefore, in this case, sublimate materials such as terephthalic acid should be converted
into liquid hydrocarbons. In order to convert this produced terephthalic acid, a catalyst
for cracking carbonyl groups of terephthalic acid is required. For this reason, transition
metal oxides, which easily form carbonyl complexes, were considered to be suitable
catalyst materials. Moreover, the transition metal oxides such as iron oxide and nickel
oxide are inexpensive. In this section, a method for the decomposition of terephthalic
acid into useful liquid hydrocarbons is described. The screening of potential catalysts was
conducted using transition metal oxide catalysts.
The catalytic cracking of PET was carried out using a fixed-bed type of reactor in
a steam atmosphere. The reactor was heated to the desired temperature in a mixture of
166 T. MASUDA AND T. TAGO
steam and nitrogen as the carrier gas. After feeding PET particles into the reactor, the
particles were heated to the reaction temperature, and then underwent the hydrolysis in a
steam atmosphere to be decomposed to terephthalic acid and lighter hydrocarbons. These
molecules were cracked through the catalyst bed. The liquid products were condensed in
two condensers cooled with ice and water. Lighter hydrocarbons were collected in a gas
pack. The amount of carbonaceous residue was calculated from the difference between
the mass of the catalyst before and after the experiment. The reaction was conducted at
the temperatures ranging from 450 to 530 K.
Figure 6.5 shows the yields ([wt%]) of the reaction of PET using several transition
metal oxide catalysts under the following conditions: a temperature of 500◦ C, a time
factor (the ratio of the mass of the catalyst W , to the PET feed rate F ) of 0.317 h, and a
particle size of 0.21–0.25 mm. Fe2 O3 did not show activity, hence these results have been
omitted. With respect to the reduction of terephthalic acid, FeOOH, nickel hydroxide and
nickel oxide showed the decomposition activity of terephthalic acid. However, a large
amount of benzoic acid, which is also a sublimate material (sublimation point ∼100◦ C),
was produced over nickel hydroxide and nickel oxide. Because these nickel compounds
are more expensive than FeOOH, FeOOH was considered to be a suitable catalyst for the
decomposition of terephthalic acid.
Figure 6.6 shows the change in product yield with increase in the time factor W /F . The
amount of sublimate materials (terephthalic acid and benzoic acid) decreased remarkably
with increase in the time factor, and no sublimate materials were observed after approxi-
mately 0.5 h. Moreover, carbon dioxide was produced, and the yield of the carbon dioxide

Gaseous
Benzoic acid Other oil compounds
5.5% Acetoaldehyde Rresidue
Phenol
Acetophenone Benzene Carbon dioxide
4.8 FeOOH
24.1 12.9 7.7 5.9 37.2

Terephthalic acid Benzoic acid

Ni(OH)2
31.9 36.1 3.1 8.2 13.5 3.1

29.6 36.6 3.3 8.8 10.3 4.1 5.5 NiO

Benzophenone Acetic acid

Liquid products Gas products

0 20 40 60 80 100
Product yield / wt%

Figure 6.5 Production yields of the reaction of PET over transition metal
catalysts. T = 500◦ C, W /F = 0.32 h−1 , particle size = 0.21–0.25 mm, molar fraction of
steam = 0.94%. (Reproduced with permission from Elsevier)
CONTINUOUS PROCESS FOR FUEL PRODUCTION 167
Other gaseous
Residue compounds
100

80 Carbon dioxide
Product yield / wt%

60 Other liquid
compounds

40 Phenol
Benzene

20 Benzoic acid
Acetophene
Terephthalic acid
0
0 0.081 0.159 0.317 0.634
W /F/ (kg-cat)· (kg-PET)−1·h

Figure 6.6 (a) Change in product yield with an increase in the time factor, W /F .
T = 500◦ C, particle size = 0.21–0.25 mm, molar fraction of steam = 0.94%. (Reproduced
with permission from Elsevier); (b) Change in product yield with an increase in the time
factor. The solid curves represent the calculated results

Terephthalic acid

Carbon dioxide Benzoic acid

Lighter hydrocarbons

Acetophenone

Other liquid compounds

Figure 6.7 Reaction pathway proposed for the reaction of terephthalic acid produced
by the hydrolysis of PET. (Reproduced with permission from Elsevier)

increased with decrease in the yield of the sublimate materials. Figure 6.7 shows the puta-
tive reaction mechanisms [5], i.e. the carbonyl groups of terephthalic acid are decomposed
to yield carbon dioxide and benzoic acid, and benzoic acid undergoes further reactions to
produce acetophenone and carbon dioxide. Some of the acetophenone was converted to
benzene and phenol, which are components classified under other liquid compounds. Two
reaction pathways, from acetophenone and other liquid compounds to carbon dioxide, are
negligibly small. In the reaction path from terephthalic acid to benzoic acid, benzene and
phenol were also produced. As a result, since sublimate materials such as terephthalic
acid and benzoic acid were successfully decomposed using an FeOOH catalyst, serious
pipe blocking at source plants could be avoided.
168 T. MASUDA AND T. TAGO
The structure and morphology of FeOOH treated at 500◦ C in a steam atmosphere
was analyzed by X-ray diffraction and transmission electron microscopy, respectively.
The X-ray diffraction analysis showed that FeOOH was transformed to Fe2 O3 during the
treatment in the steam atmosphere. In contrast, the TEM observation revealed micropores
of 1 nm diameter in the untreated FeOOH, which was not observed with commercial
Fe2 O3 . Moreover, the pores increased in diameter to approximately 5–100 nm after the
steam treatment. This morphology was thought to be the result of the dehydration of
FeOOH. It is possible that many active sites were generated on the surface of the pores,
resulting in the observation that treated FeOOH showed high activity, even though its
crystal structure is the same as that of Fe2 O3 [6, 20].
Terephthalic acid is a useful source material of PET, as well as benzoic acid and
benzoates. However, in order to recycle the terephthalic acid, produced further purification
is required, because other organic compounds are also produced as impurities in the
degradation process of waste plastic mixtures, e.g. PE and PET mixtures described in
Section 2.3.

2.3 CONTINUOUS DEGRADATION OF WASTE PLASTICS MIXTURES FOR

THE RECOVERY OF HEAVY OIL

The dominant mechanisms of polyester resin degradation (such as polycarbonate, poly-

butyleneterephthalate and polyethyleneterephthalate) changes from thermal pyrolysis to
hydrolysis by the introduction of steam as the carrier gas. PET was successfully degraded
by hydrolysis in a steam atmosphere, yielding an amount of pure terephthalic acid that
could be predicted from the chemical formula of PET, and leaving carbonaceous residue
of less than 1%. From economic and energetic viewpoints, it is both inexpensive and
easy to employ steam as the carrier gas in chemical recycling plants. Furthermore, the
temperature required to initiate was found to decrease approximately 70◦ C when polyester
resins were degraded by hydrolysis in a steam atmosphere [12].
When the mixture of waste plastics was degraded by accelerating the hydrolysis of the
plastics with steam and by decomposing the generated sublimate materials over FeOOH
catalyst, the following reaction conditions are needed:
1. A good contact of melted plastics with steam to accelerate the hydrolysis of plastics.
2. Large rate of the heat transfer to heat plastics up to a desired temperature.
3. High hold-up of plastics in a reactor to achieve an enough reaction time for degrading
plastics.
4. Contact of vapor of sublimate materials with FeOOH catalyst to decompose the mate-
rials.
Bockhorn et al. [2, 3] proposed circulated-spheres reactor for the pyrolysis of waste
plastics. This reactor enabled one to remove gaseous products from the reaction zone and
to achieve high heat transfer rates, namely condition (2) described above. The concept
of this circulated-spheres reactor has the possibility to attain conditions (1) and (3) by
improving the reactor.
Based the discussions above, a new pyrolytic reactor can be proposed that would be
capable of achieving a high hold-up, high heat transfer, and good contact between melted
plastics and steam for the acceleration of hydrolysis. Using this novel reactor system,
CONTINUOUS PROCESS FOR FUEL PRODUCTION 169
a mixture of waste plastics was degraded, and further decomposition over an FeOOH
catalyst was also achieved.
This novel reactor is a pyrolytic reactor system using stirred heat-medium-particles
(Figure 6.8). This pyrolytic reactor system is composed of a series of three types of
reactor. The first reactor uses stirred heat-medium-particles (reactor 1), the second is
a tank reactor (reactor 2), and the remaining reactor is a fixed-bed reactor (reactor 3).
Reactor 2 is located under reactor 1, and is separated from reactor 1 by a stainless steel
net. The seizes of these reactors are described in the figure. The expected reaction behavior
is described below.
In reactor 1, plastic particles are fed into the top of a bed of glass beads as the heat-
medium-particles, and the particles are then melted and adhered to the beads. The glass
beads are stirred slowly by two equipped impellers, one of which is the propeller-type of
impeller, and the other is an anchor-type of impeller located at the bottom of the bed of
glass beads. The propeller-type of impeller is turned to lift the particles. In this manner,
the glass beads at the top layer of the bed of glass beads are replaced continuously by

For pyrolysis For upgrading

Particles of
plastics Oil

N2

Motor N2

Pre-heating zone of steam

Glass beads N2
Impeller Water

Reactor 1
f: 55 mm
H: 100 mm

Catalyst Sampling zone

Reactor 2
f: 55 mm
Gas pack
H: 70 mm
Condenser
Stainless steel net
Reactor 3
T=400~500ºC
f: 35 mm
Muffle furnace H: 80 mm

(a)

Figure 6.8 (a) Schematic view of the proposed reactor system for the chemical recycling
of plastics; (b) photograph of the proposed reactor system. (Reproduced with permission
from Elsevier)
170 T. MASUDA AND T. TAGO

Reactor 1
Inlet of plastic
particles

Muffle furnace Reactor 2

Reactor 3

(b)

Figure 6.8 (continued )

other beads lying beneath the top layer. These stirred beads increase the rate of heat
transfer. Following the string of the glass beads, the melted plastics are transported over
the bed of glass beads, resulting in a high hold-up of plastics in the reactor and good
contact with the steam, which functions as the carrier gas. The melted plastics on the
glass beads are decomposed by hydrolysis with steam and the random scission of C–C
bonds. Some of the melted plastics on the glass beads are carried to the bottom of the bed
of glass beads, and are dripped onto reactor 2 (tank reactor). In reactor 2, the unreacted
plastics undergo further decomposition, yielding various gaseous compounds. Reactor 3
is filled with an FeOOH catalyst. Gaseous compounds, including the vapors of sublimate
materials, are passed through the FeOOH catalyst bed, where they undergo the catalytic
degradation.
A mixture of PE and PET was used as a model waste plastic mixture. The ratio of
PE/PET was 15:2, which is the ratio of the amounts of the two plastics discarded in
Kyoto, Japan. After continuous degradation with and without stirring of the glass beads,
the glass beads were collected from reactor 1 (Figure 6.8). When the glass beads were
not stirred, which corresponded to a trickle-bed type of reactor, a massive carbonaceous
residue remained with the glass beads on the top of the bed of glass beads. For this reason,
the heat transfer rate was low and the melted plastics did not make good contact with
the steam. On the other hand, there was not a massive amount of residue and glass beads
slightly colored yellow, when the glass beads were stirred by rotating the impellers at
rates exceeding 8 rpm. Furthermore, no residue nor oil was in reactor 2, the tank reactor,
CONTINUOUS PROCESS FOR FUEL PRODUCTION 171
(Figure 6.8). These results indicate that the proposed reactor system is useful for the
continuous degradation of plastics.
Figure 6.9 shows the dependency of the product yields, based on data collected at the
outlet of reactor 3 (see Figure 6.8), on the ratio of the catalyst mass W to the feed rate
of the plastics F and the time factor W /F increased from 0 to 1.5 h. Reactions were
conducted under temperature conditions ranging from 500◦ C, at a catalyst weight W of
about 2.0 × 10−2 kg. When the FeOOH catalyst was not loaded into reactor 3, a large
amount of yellow wax was obtained. When the FeOOH catalyst was loaded into reactor 3,
oil and carbon dioxide were produced. The amount of carbon dioxide and that of gaseous
hydrocarbons increased, and the yield of oil decreased, as the amount of FeOOH catalyst
loaded increased. This suggests that the FeOOH catalyst is capable of catalysis, leading
to the decomposition of a wax via oxidation by oxygen atoms from the lattice of FeOOH
and/or from H2 O.
When a wax is decomposed via oxidation by the oxygen atoms from the lattice of
FeOOH catalyst, the catalytic activity of an FeOOH catalyst will decrease. Therefore, the
gaseous product yields of the reaction were measured by sampling at different intervals
at a temperature of 500◦ C and the time factor of 1 h. The main gaseous products were
carbon dioxide (∼3 wt% for 140 min), n-C4 H10 (2 wt%), n-C3 H8 (2 wt%), C2 H6 (0.5
wt%), C2 H4 (1.5 wt%), and CH4 (0.5 wt%). Except for at the beginning of the reaction,
there were only negligible changes in the product yields. The amount of oxygen required
for producing carbon dioxide during a reaction time of 140 min was evaluated, and was
found to be larger than that generated through the phase change of iron from Fe2 O3 to
Fe3 O4 . These results suggest that the wax was decomposed by reaction with H2 O over
the FeOOH catalyst, and the catalytic activity of the FeOOH catalyst remained stable in
the steam atmosphere.

100
CO2 Coke
80
Product yields / wt%

Wax
60
Oil

40

20

Gaseous hydrocarbons
0
0 0.32 1.0 1.5
Time factor, W/F/ h

Figure 6.9 Effect of an FeOOH catalyst loaded in reactor 3 (see Figure 6.8) on prod-
uct yield; temperature = 500◦ C, carrier gas: steam. (Reproduced with permission from
Elsevier)
172 T. MASUDA AND T. TAGO

50
C20+
Feedstock: PE/PET = 15/2
40 Catalyst: FeOOH
W/F = 1 h
T= 500
Yield / wt%

30

20

10

0
51 0 15 20
Carbon number / -

Figure 6.10 Carbon number distribution of heavy oil produced by degradation of a

mixture of polyethylene (PE) and poly(ethylene terephthalate) (PET), at a weight ratio of
PE/PET = 15/2, temperature = 500◦ C, time factor W /F = 1 h, carrier gas: steam. (Repro-
duced with permission from Elsevier)

Figure 6.10 shows the carbon number distribution of the products obtained when the
temperature was 500◦ C and the time factor was 1 h. The oil produced was considered to
correspond to heavy oil, as based on the carbon number distribution. Therefore, it would
be necessary to upgrade this oil for practical applications.

3 UPGRADING OF WASTE-PLASTICS-DERIVED HEAVY OIL OVER

CATALYSTS
In the series described thus far, it was found that the degradation of waste plastics proceeds
efficiently by both thermal pyrolysis and hydrolysis in a steam atmosphere. A wax and
carbonaceous residue produced by the hydrolysis of PET are decomposed by reaction
with steam over an FeOOH catalyst, the activity of which remains stable in a steam
atmosphere. However, the liquid product from generated from the process mentioned
above contains a large amount of heavy oil, as shown in Figure 6.10. Both catalysts and
chemical processes are required for efficiently upgrading the quality of the heavy oil.

3.1 CATALYTIC CRACKING OF HEAVY OIL OVER SOLID-ACID CATALYSTS

The most widely used conventional chemical methods are pyrolysis [21–25] and cat-
alytic cracking [13, 26–30]. The latter yields products with a smaller range of carbon
numbers and of a higher quality than products generated by the former method. Several
types of solid acid catalysts, which are known to be effective for catalytic cracking (e.g.
HZSM-5, HY and rare earth metal-exchanged Y-type (REY) zeolite and silica–alumina
(SA)) were evaluated by catalyst screening tests and are listed in Table 6.1. The acidic
CONTINUOUS PROCESS FOR FUEL PRODUCTION 173
Table 6.1 Properties of the catalysts. (Reproduced with permission from Elsevier)

Properties Catalyst
REY HY Silica–alumina HZSM-5 HZSM-5
(SA) (65) (1000)
Si/Al 4.8 4.8 13% 65 1000
alumina
Pore size (A) 7.4 7.4 60–100 5.3×5.6 5.3×5.6
Amount of total
acid sitesb 1.048 1.207 0.560 0.235 0.067
[mol · kg-cat−1 ]
Amount of strong
acid sitesc 0.375 0.441 0.187 0.122 0.031
[mol · kg-cat−1 ]
Supercage Yes (11.8A) Yes (11.8A) No No No
a
Measured by conventional TPD experiment
b
Based on the total amount of ammonia desorbed in the TPD experiment
c
Based on the amount of ammonia desorbed above 300◦ C in the TPD experiment

properties of the catalysts were measured from the temperature-programmed desorption

spectra of ammonia (NH3 -TPD) method. The feed oil was obtained by the pyrolysis of
solid polyethylene plastics at 450◦ C. To ensure homogeneity and to remove the lighter
hydrocarbons, the oil was distilled at 473 K. Table 6.2 shows the results of the elementary
analysis of the oil, which was used as a feed oil in the catalytic reforming reaction. A
continuous-flow, fixed-bed reactor was utilized for the catalytic reforming of the oil. The
reaction was carried out at 400◦ C under a nitrogen stream. The oil was fed at a constant
weight hourly space velocity (WHSV) of 1.0 kg-oil·kg-cat−1 ·h−1 for all catalysts.
Figure 6.11 shows the product yields for each catalyst. The products are classified into
four lumps, i.e. gas (carbon number 1–4), gasoline (5–11), heavy oil (above 12), and
a carbonaceous residue referred to as coke. In the figure, PE oil represents the feed oil
and contains a 34% gasoline fraction. The feed oil was effectively cracked by solid acid
catalysts. The gasoline yield was highest with REY zeolite. HZSM-5(65) yielded the

Table 6.2 Analysis of the feed oil. (Repro-

duced by permission of the American Chem-
ical Society)

wt% mol%
Heavy oil (≥C12 ) 95
Gasoline (C5 –C11 ) 5
Elemental analysis
H 13.7 66.0
C 82.0 32.9
N 0.0 0.0
O 4.3 1.1
H/C 2.00

RON of liquid [−] 4.5.

174 T. MASUDA AND T. TAGO

Gasoline Gas Coke C12+oil

HY

REY

SILICA ALUMINA

HZSM-5(1000)

HZSM-5(65)

PR OIL

0 20 40 60 80 100
Yield [wt%]

Figure 6.11 Distribution of product yields under nitrogen. WHSV = 1, T = 400◦ C and
t = 3 h. (Reproduced with permission from Elsevier)

largest quantity of gaseous compounds and the lowest amount of gasoline, which was
even lower than the gasoline fraction in the feed oil. While coke loading was highest
with HY zeolite, both HZSM-5 zeolites generated a negligibly small amount of coke
deposition. As the amount of coke deposition increased, the catalytic activity decreased.
Table 6.3 summarizes the yield of gaseous products according to the carbon number. In
this table, C2 = to C5 = indicate olefins corresponding to their respective carbon numbers.

Table 6.3 Yield of each gaseous product according to carbon number. The values in brackets
refer to the corresponding olefins. (Reproduced with permission from Elsevier)

Catalyst
Carbon REY HY Silica–alumina HZSM-5 HZSM-5
Number (SA) (65) (1000)
C1 0.09 0.20 0.07 0.12 0.01
C2 0.76 1.39 0.23 6.32 0.43
(C2=) (0.69) (1.39) (0.17) (6.01) (0.40)
C3 7.71 9.85 7.99 30.92 11.00
(C3=) (5.85) (3.25) (6.91) (29.90) (11.00)
C4 16.28 20.50 17.35 23.80 14.21
(C4=) (12.22) (15.51) (11.24) (13.78) (8.27)
C5 6.60 9.23 3.51 3.84 5.62
(C5=) (0.0) (0.0) (2.30) (1.80) (1.52)
C6 2.81 1.55 0.06 1.48 1.09
C7 1.23 1.02 0.0 0.62 0.41
C8 0.40 0.46 0.0 0.14 0.75
Total yield, (wt%) 35.88 44.20 29.20 69.04 33.52
CONTINUOUS PROCESS FOR FUEL PRODUCTION 175
A significant amount of C3 –C5 gaseous compounds were produced. With the exception
of ZSM-5(65) zeolite which favored the production of the C3 fraction, the other catalysts
yielded C4 as the main component.
HZSM-5 zeolite has channels within its crystals. The size of the channel is nearly equal
to that of benzene and is too small for easy penetration by oil molecules. Hence, the ends
of only certain molecules can penetrate the channels and undergo cracking. This leads to
a higher yield of gaseous products and a lower gasoline yield, indicating that HZSM-5
zeolite is not suitable for the cracking reaction of heavy oil. In contrast, the HY and REY
zeolites have larger pores. Therefore, the oil molecules can penetrate into the pores and
undergo cracking. Moreover, the existence of rare earth metals in REY zeolite results in
a decrease in the amount of stronger acid sites (see Table 6.1). This in turn leads to a
reduction in the deactivation rate and the amount of coke loading in comparison with that
obtained with HY zeolite. Accordingly, REY zeolite has the proper acidic properties and
pore size, and is suitable for the reaction with heavy oil.

3.2 PRODUCTION OF HIGH-QUALITY GASOLINE OVER REY ZEOLITES

The effects of the reaction conditions and the catalytic properties of REY zeolites on
reaction product yields and on the quality of the gasoline can now be examined.
Four types of REY zeolite (Si/Al = 4.8) with different crystal sizes and acidic prop-
erties were used. The physical and chemical properties of the fresh zeolites are given in
Table 6.4. Polyethylene plastics-derived heavy oil, shown in Table 6.2, was used as the
feed oil. The cracking reaction was conducted in a tubular reactor filled with catalyst par-
ticles under the following conditions: time factor W /F = 0.2–3.0 kg-cat kg oil−1 h−1 and
reaction temperature = 300–450◦ C. The lumping of reaction products were gas (carbon
number 1–4), gasoline (5–11), heavy oil (above 12), and a carbonaceous residue referred
to as coke. The index of the gasoline quality used was the research octane number (RON),
which was calculated from Equation 6.1 [31].

RON = −1.0729YNP2 + 0.7875YIP1 + 0.0978YIP2 + 0.3395YCP + 0.4049YAR + 69.0306

(6.1)

Table 6.4 Physical and chemical properties of the fresh REY zeolites. (Reproduced by permission
of the American Chemical Society)

Catalyst
Property REY-1 REY-2 REY-3 REY-4
Si/Al 4.8 4.8 4.8 4.8
Crystal size (µm) 0.1 1.0 0.1 0.1
Amount of total acid sitesa 2.91 2.44 2.99 2.78
(mol kg cat−1 )
Amount of strong acid sitesb 0.79 0.57 0.66 0.70
(mol kg cat−1 )
a
Based on the total amount of ammonia desorbed in the TPD experiment
b
Based on the amount of ammonia desorbed above 300◦ C in the TPD experiment
176 T. MASUDA AND T. TAGO

100
450°C

Conversion of heavy oil [%]

80
350°C
60
300°C

40 300°C
400°C 350°C
20 400°C
450°C
0
0 1 2 3 4
Time factor, W/F [kg-cat kg-oil−1 h]

Figure 6.12 Dependence associated with the reaction temperature and the time factor,
W /F , on the conversion of heavy oil: zeolite crystal size = 0.1 µm. (Reproduced by
permission of the American Chemical Society)

where Yi is the weight fraction of the ith component in the gasoline fraction. The subscript
NP2 denotes the n-paraffins without C5 , IP1 the total isoparaffins from C5 to C7 , IP2 the
total isoparaffins without C5 –C7 , CP the total cycloparaffins, and AR the total aromatics.
The relationship between the conversion of heavy oil and the time factor, W /F , at
different reaction temperatures is shown in Figure 6.12. Conversion was defined as the
mass fraction of heavy oil (components above C12 ) converted to gasoline, gas, and coke.
This value was calculated from the following equation:

mass of heavy oil at the outlet

conversion = 1 − (6.2)
mass of heavy oil at the inlet

As the reaction temperature increased, the reaction proceeded. Although conversion

was greater as the temperature increased, there was no significant difference between the
conversions at 400 and 450◦ C. The effects of temperature on the respective yields of the
reaction products are shown in Table 6.5. As the temperature increased, the amount of
unreacted heavy oil decreased, and the yields of both gas and coke increased. The gasoline
yield reached a maximum value at 400◦ C, and then decreased with further increases in
temperature.

Table 6.5 Effect of temperature on product yield at W/F = 0.75 kg-cat kg oil−1 h. (Reproduced
by permission of the American Chemical Society)

Temperature Heavy oil Gasoline Gas Coke

(◦ C) (wt%) (wt%) (wt%) (wt%)
300 55.45 37.22 6.96 0.37
350 30.77 48.04 20.76 0.43
400 15.84 51.93 31.71 0.52
450 11.43 39.61 48.42 0.54
CONTINUOUS PROCESS FOR FUEL PRODUCTION 177
Figure 6.13(a) shows the effect of the reaction temperature on the relationship between
gasoline yield and the conversion of heavy oil. The gasoline yield increased with increas-
ing conversion to the maximum value, and then decreased significantly. This suggests that
the gasoline formed by the cracking of heavy oil subsequently undergoes further cracking,
which in turn yields gaseous products and coke. Thus, gasoline is an intermediate prod-
uct. The maximum gasoline yield, which is related to the rates of gasoline formation and
cracking, was observed at about 400◦ C. The same optimum temperature has been found
for the catalytic cracking of gas oil [32]. Figure 6.13(b) shows the relationship between
the gas yield and the conversion of heavy oil at various reaction temperatures. Because
almost all of the data lie on a single curve, the reaction temperature had no significant
effect on the gas yield at a constant conversion level. The yield of gas products increased

60
300°C 350°C
400°C
50
350°C
Gasoline yield [wt%]

40 400°C
450°C
30 300°C

20 450°C

10

0
0 20 40 60 80 100
Conversion of heavy oil [%]
(a)

100
300°C
80 350°C
Gas yield [wt%]

400°C
60
450°C

40

20

0
0 20 40 60 80 100
Conversion of heavy oil [%]
(b)

Figure 6.13 Effect of the reaction temperature on the relationship between product yield
and the conversion of heavy oil (zeolite crystal size = 0.1 µm): (a) gasoline yield; (b) gas
yield; (c) coke yield. (Reproduced by permission of the American Chemical Society)
178 T. MASUDA AND T. TAGO

1.2
300°C
300°C
0.9 350°C

Coke yield [wt%]

400°C
450°C 350°C
0.6

0.3
450°C

400°C
0
0 20 40 60 80 100
Conversion of heavy oil [%]
(c)

Figure 6.13 (continued )

as the reaction progressed. Figure 6.13(c) shows the effect of different reaction tempera-
tures and conversion levels of heavy oil on the coke yield. At the same conversion level,
high temperatures reduced coke formation. The difference between the coke yields at 400
and 450◦ C was small. Similar findings have been reported for the catalytic cracking reac-
tion of gas oil, in which coke formation proceeded well at reaction temperatures below
400◦ C [32].
The effect of reaction temperature on gasoline quality and its main components are
shown in Figure 6.14. Below 400◦ C, the RON value increased with temperature due to
an acceleration of the formation rate of the IP1 fraction and the cracking rate of the NP2
fraction. Above 400◦ C, however, the cracking of IP1 proceeded (i.e. a reduction in the
yield of IP1), leading to a decrease in the RON value. On the basis of the gasoline, coke,

120
Composition [wt%], RON [-]

90

RON
60

AR
30
NP2
IP1
0
250 300 350 400 450 500
Temperature [°C]

Figure 6.14 Effect of the reaction temperature on the RON value of the gasoline
and main components (zeolite crystal size = 0.1 µm, W /F = 0.75 kg-cat kg-oil−1 h):
NP2 = n-paraffins without C5 , IP1 = C5 –C7 isoparaffins, AR = aromatics. (Reproduced
by permission of the American Chemical Society)
CONTINUOUS PROCESS FOR FUEL PRODUCTION 179
120
RON

Composition [wt%], RON [-]

90

AR
60 IP1

30

NP2
0
0 1 2 3
−1
Time factor, W/F [kg-cat•kg-oil •h]

Figure 6.15 Effect of the time factor, W /F , on the RON value of the gasoline and
the main components (zeolite crystal size = 0.1 µm, reaction temperature = 400◦ C):
NP2 = n-paraffins without C5 , IP1 = C5 –C7 isoparaffins, AR = aromatics. (Reproduced
by permission of the American Chemical Society)

Table 6.6 Comparison of commercial gasoline and the gasoline obtained from heavy oil derived
from waste plastics. (Reproduced by permission of the American Chemical Society)

Gasoline obtained Regular High-octane

Component (optimized) gasoline gasoline
IP1 (wt%) 40.44 29.47 36.78
AR (wt%) 29.03 33.23 50.53
NP2 (wt%) 12.36 15.41 9.96
RON (−) 100.92 90.44 108.01

and gas yields, as well as the RON value, the most favorable reaction temperature was
determined to be approximately 400◦ C.
The effect of the time factor, W /F , on gasoline quality and its main components
obtained at 673 K is shown in Figure 6.15. Below a W /F value of 0.75 kg-cat kg-oil−1
h, the increase in the RON value was due to the significant increase in the IP1 fraction
and the large reduction in the NP2 fraction. Above a W /F value of 1 kg-cat kg-oil−1 h,
only the reaction of IP1 to AR took place, producing a slight decrease in the RON value.
These results suggest that the optimum W /F value for the production of gasoline of the
highest quality is in the range 0.75–1 kg-cat kg-oil−1 h.
Table 6.6 compares the contents of the main components of regular and high-octane
gasoline with those of gasoline obtained under the optimal conditions, namely, tempera-
ture = 673 K, time factor = 0.75 − 1 kg cat kg oil−1 h, crystal size of the REY zeolite
catalyst = 0.1 µm, and the number of strong acid sites on the used catalyst = 0.28 mol
kg−1 . The gasoline obtained under the optimum contained a larger amount of IP1 and
a smaller amount of AR than the corresponding amounts in commercial gasoline. The
amount of NP2 in the gasoline obtained in this study was between that of regular and
high-octane gasoline.
180 T. MASUDA AND T. TAGO

3.3 KINETICS OF THE CATALYTIC CRACKING OF HEAVY OIL OVER REY

ZEOLITES

A rare earth metal-exchanged Y-type (REY) zeolite catalyst was found to be an effective
catalyst for the catalytic cracking of heavy oil. The influence of the reaction conditions
and the catalytic properties of REY zeolite on the product yield and on gasoline quality
have been described above. In this section, a reaction pathway is proposed for the catalytic
cracking reaction of heavy oil, and a kinetic model for the cracking reaction is developed
[14,33].
Figure 6.16 shows the typical relationship between product distribution and the time
factor, W /F , at different temperatures [13]. The experimental conditions are described in
Section 3.2. As the W /F value increased, heavy oil was cracked to produce gasoline and
gaseous products. Moreover, the gasoline product subsequently underwent further cracking
to yield gaseous products. Hence, the gasoline yield was shown to have a maximum value

100
300°C
80 Heavy oil
Composition [wt%]

60
Gasoline
40

20 Gas
Coke
0
0 1 2 3 4
Time factor, W/F [kg-cat•kg-oil−1•h]
(a)

100
350°C
80
Composition [wt%]

Heavy oil

60
Gasoline

40 Gas

20
Coke
0
0 1 2 3 4
Time factor, W/F [kg-cat•kg-oil−1•h]
(b)

Figure 6.16 Kinetic runs performed using a catalyst with a crystal size of 0.1 µm:
(a) 300◦ C; (b) 350◦ C; (c) 400◦ C; (d) 450◦ C. (Reproduced by permission of the American
Chemical Society)
CONTINUOUS PROCESS FOR FUEL PRODUCTION 181
100
400°C
Heavy oil
80

Composition [wt%]
Gas
60 Gasoline

40

20
Coke
0
0 1 2 3 4
Time factor, W/F [kg-cat•kg-oil−1•h]
(c)

100
450°C Gas
80
Composition [wt%]

Heavy oil
60

40
Gasoline
20
Coke
0
0 1 2 3 4
Time factor, W/F [kg-cat•kg-oil−1•h]
(d)

Figure 6.16 (continued )

that appeared at a low W /F value and a high reaction temperature. The yield of coke
gradually increased and was considered to be the product of gasoline and heavy oil.
Figure 6.17 illustrates a possible reaction pathway that could account for the product
distributions shown in Figure 6.16. The proposed reaction pathway separately takes into
account the heavy oil, gasoline, gas, and coke lumps and is considered to represent the
product distribution.
The experimental conditions were set up to ensure that both the heat and mass transport
limitations across the film would be negligible. Moreover, limitations due to intraparti-
cle diffusion were assumed to be insignificant. The mass balance equation of the ith
component can be written as follows:
dFi
= ri (i = A, B, C, D) (6.3)
dW
where Fi is the mass flow rate of the ith lump (kg h−1 ); W is the mass of catalyst (kg);
ri is the production rate of the ith lump per unit mass of catalyst (kg (kg cat)−1 h−1 ); and
suffixes A, B, C and D refer to heavy oil, gasoline, gas, and coke lumps, respectively.
182 T. MASUDA AND T. TAGO

HEAVY OIL

k2 k3
GAS k1 COKE
k4 k5

GASOLINE

Figure 6.17 Reaction pathway proposed in this study. (Reproduced by permission of

the American Chemical Society)

Figure 6.18 shows the Arrhenius plots using the evaluated kinetic parameters. The data
were found to lie on a straight line for each parameter. The slopes of these straight lines
gave the activation energies, which are listed in Table 6.7. The activation energy for the
reaction of gas formation from heavy oil k2 is 75.5 kJ mol−1 and is comparable with
other data for gasification reactions: 58.6 kJ mol−1 in the case of a CaX catalyst (Ca
ion-exchanged X-type zeolite catalyst) [34], and 61.5 kJ mol−1 in the case of the silica
alumina [35] for the gasification of a polymer waste, and 75 kJ mol−1 for the reaction
of gas oil [36]. The difference in the activation energies between gaseous formation k2

104

103 k1

102 k2
k1, k2, k3[m6(kg kg-cat h)−1]
k4, k5, k3[m3(kg-cat h)−1]

k4
101

100

k3
10−1

k5
10−2

10−3
1.2 1.4 1.6 1.8 2.0
T −1× 103 [K−1]

Figure 6.18 Arrhenius plots of the kinetic parameters. (Reproduced by permission of

the American Chemical Society)
CONTINUOUS PROCESS FOR FUEL PRODUCTION 183
Table 6.7 Activation energies using a catalyst with a crystal
size of 0.1 µm. (Reproduced by permission of the American
Chemical Society)

Rate Rate
constants Ea (kJ mol) constants Ea (kJ mol)

k1 50.7 k4 35.1
k2 75.5 k5 42.1
k3 18.5

and gasoline formation k1 accounts for the fact that the selectivity of gaseous products
increases, while that of gasoline decreases with increases in temperature, especially at
temperatures above 400◦ C.

3.4 USAGE OF STEAM AS A CARRIER GAS

As it is both inexpensive and easy to handle, steam is a potential candidate carrier gas for
waste plastic recycling in chemical plants. Furthermore, as mentioned in Section 2.1, the
degradation temperatures for polyester resins are remarkably shifted to low-temperature
regions, and the amount of carbonaceous residue produced in the degradation process is
reduced in a steam atmosphere, as compared with that in a nitrogen atmosphere. Accord-
ingly, the preparation of a catalyst that could demonstrate both stable activity for the
catalytic cracking of PE-derived heavy oil, but that would also remain stable in a steam
atmosphere, was examined [16].
Nickel is a well-known catalyst component and is thought to play an important role
in the transformation of the hydrogen of steam to hydrocarbons. Hence, a part of the
rare earth metal in REY is exchanged with Ni to become prepared Ni and the rare earth
metal-exchanged Y-type zeolite catalyst (Ni–REY) [14, 15]. The physical and chemical
properties of the catalysts are listed in Table 6.8. The polyethylene plastics-derived heavy
oil shown in Table 6.2 was used as the feed oil.
A continuous-flow, fixed-bed reactor was utilized for the catalytic cracking of the heavy
oil. Reactions were conducted under temperature conditions ranging from 300 to 600◦ C,
at a catalyst weight W of about 1.0 × 10−3 kg and a feed oil mass flow rate F of about
1.0 × 10−3 kg h−1 . In order to examine the catalysis of Ni in Ni–REY for hydrogenation,
experiments using hydrogen as the carrier gas were also conducted.

Table 6.8 Physical and chemical properties of the catalyst samples. (Reproduced with permission
from Elsevier)

Ni–REY
REY Ni content 1.0 wt% 3.0 wt% HY MFI
0.5 wt%
SiO2 /Al2 O3 4.8 4.8 4.8 4.8 4.8 65
Crystal size (mm) 0.1 0.1 0.1 0.1 0.1 1.1
Number of strong acid sites (mol kg−1 ) 0.375 0.441 0.122
184 T. MASUDA AND T. TAGO
The selectivity of the products of the reaction in nitrogen and hydrogen stream at 400◦ C
was compared among the catalysts used at a feed oil conversion of about 80%. Hydrocar-
bons with carbon numbers ranging from 1 to 4 were regarded as gaseous compounds, and
those with carbon numbers from 5 to 11 were regarded as gasoline fractions. MFI-type
zeolite (ZSM-5) yielded the largest quantity of gaseous compounds (selectivity: 64%) and
the lowest amount of gasoline (35%) because its pores are too small for the penetration
of heavy oil, and reactions such as dewaxing are likely to proceed inside. As HY-type
zeolite had a large umber of strong acid sites, and excessive cracking occurred, a low
yield of gasoline was obtained (46%). The usage of REY, on which the number of strong
acid sites is lower than that on the HY-type zeolite, gave a higher yield of gasoline (57%)
as compared with that of HY. On the other hand, when the reaction was conducted using
the Ni–REY catalyst in hydrogen stream, the selectivity towards gasoline was found to
be the highest at 64%, at the same temperature and conversion of the feed oil.
The reactions over HY, MFI, and REY zeolite catalysts in nitrogen proceed over their
acid sites. During the reactions, paraffins are first decomposed to yield lighter olefins
and paraffins. The produced olefins are preferentially adsorbed on acid sites and undergo
further reaction, yielding gaseous compounds and aromatics. When a Ni–REY catalyst is
employed in a hydrogen atmosphere, the decomposition of hydrogen to hydrogen atoms
also proceeds over Ni in the catalyst. These hydrogen atoms diffuse on the pore surface
(e.g. the spillover phenomenon) and when they make contact with the olefins adsorbed on
the acid sites, the hydrogenation of the olefins occurs. The paraffins formed in this manner
easily desorb from the acid sites without undergoing any further reactions. Therefore, the
reactions using Ni–REY in this series showed the highest selectivity towards gasoline.
In order to optimize the Ni content of Ni–REY, the compositions of the obtained
gasoline fraction were measured and were divided into lumps as shown in Figure 6.19.
Figure 6.20 shows the RON value calculated from the data in Fig. 6.19 using
Equation (6.1). The results obtained with commercial regular-grade and high-RON-grade
gasoline are also shown in these figures for comparison. Commercial gasoline with
a high RON value contains large amounts of isoparaffins (IP1), aromatics (AR), and
cycloparaffins (CP), as well as a small amount of n-paraffins (NP2). For the complete
combustion of gasoline, the AR content must be low. Therefore, high-quality gasoline
must contain large amounts of IP1 and CP and small amounts of NP2 and AR. Using
Ni–REY with a 0.5 wt% Ni content, gasoline with larger amounts of CP and IP1 and
smaller amounts of AR and NP2 was obtained, as compared with a commercial high-RON-
grade gasoline. Furthermore, the RON value of the gasoline produced over Ni–REY (0.5
wt%) was larger than that of commercial gasoline. As the content of Ni increased, the
number of strong acid sites decreased, resulting in a reduction in cracking. This explains
why the amount of IP1 decreased and that of NP2 increased with the increase in the Ni
content (from 0.5 to 3.0 wt%), leading to the reduction of the RON value of the gasoline
produced in this manner. Based on the above information, it was concluded that Ni–REY
with a Ni content of 0.5 wt% could be used for the following experiments.
The effects of reaction temperature on the conversion of the heavy oil and the gaso-
line yield are shown in Figure 6.21 for reactions using Ni–REY in steam and hydro-
gen streams. At temperatures below 400◦ C, both the conversion and the gasoline yield
increased as temperature increased. However, at temperatures above 400◦ C, excessive
CONTINUOUS PROCESS FOR FUEL PRODUCTION 185
100

Cycloparaffin Aromatics (AR)

(CP)
80

n-paraffin (NP)
Composition/ wt%

60

40

iso-paraffin
20 (IP2)
iso-paraffin (IP1)

0
HY MFI REY 0.5 1.0 3.0 Regular High RON
(Ni content / wt%) grade grade

in N2 Ni – REY in H2 Commercial
gasoline

Figure 6.19 Comparison of gasoline fractions produced by catalytic cracking over

the various catalysts used (T = 400◦ C, W /F = 1 h). (Reproduced by permission from
Elsevier)

120
High RON grade
110 gasoline
RON value / -

100

90
Regular
80
grade
gasoline
70
HY MFI REY 0.5 1.0 3.0
(=Ni content / %)
in N2
Ni – REY in H2

Figure 6.20 Research octane umber (RON value) of the gasoline fraction produced with
the catalysts used. (Reproduced by permission from Elsevier)

cracking proceeded, followed by coke formation, leading to a decrease in both conver-

sion and gasoline yield. This tendency was clearly observed in the case of hydrogen
stream. Strong acid sites, on which coke formation proceeds and is accelerated at higher
temperatures, are partially covered by water molecules in a steam atmosphere. Therefore,
the catalyst is able to maintain activity above 400◦ C in a steam atmosphere.
186 T. MASUDA AND T. TAGO

100 100

80 80

Yield of gasoline fraction/%

Conversion of feed oil / %
In steam

60 60
In H2

40 40
In steam
In H2
20 20
Catalyst : Ni–REY

0 0
200 300 400 500 600 700
Reaction temperature/ºC

Figure 6.21 Dependence of both conversion and gasoline yield on reaction temperature
(W /F = 1 h)

Figure 6.22 shows the typical carbon number distribution of products obtained using
Ni–REY in steam at 400◦ C and with a time factor W /F of 1 h. The results obtained with
MFI-type (ZSM-5) and REY zeolites in N2 are also shown for comparison. Although
steam was used as a carrier gas, Ni–REY gave the largest amount of fuel, e.g. gasoline,
kerosene, and gas oil, thus suggesting the potential use of steam as a carrier gas.

30
Gas Gasoline Reaction conditions:
Kerosene T = 400 °C
W/F = 1 h
Gas oil

20
MFI in N2
Yield/%

REY in N2
Feed oil
Ni-REY in steam
10

0
5 10 15 20 25
Carbon number/-

Figure 6.22 Carbon number distribution of products of the catalytic cracking of oil
obtained by the pyrolysis of PE. (Reproduced with permission from Elsevier)
CONTINUOUS PROCESS FOR FUEL PRODUCTION 187
From the results discussed above, the combination of Ni–REY as the catalyst and
steam as the carrier gas was considered preferable for achieving a realistic recycling
process. In order to confirm the long-term stability of Ni–REY, repeated sequences of
reaction and regeneration were conducted. Reaction and regeneration were conducted at
400◦ C for 3 h in a steam atmosphere, and at 500◦ C for 3 h in an air stream, respectively.
This experiment was also conducted using an MFI-type zeolite catalyst for comparison
by exchanging the carrier gas in the reaction from steam to nitrogen. Figure 6.23 shows
changes in the conversion of the feed oil and the selectivity towards gasoline for REY
and Ni–REY catalysts during the sequences of reaction and regeneration. The carrier
gas in the reaction was steam for the REY and Ni–REY catalysts. Both the activity
and the selectivity remained almost constant for REY and Ni–REY, indicating that both
catalysts are stable in a steam atmosphere. The stable activity observed with REY and
Ni–REY could be ascribed to the stability of the acid sites on the REY and Ni–REY
catalysts [37]. Furthermore, Ni–REY showed higher selectivity towards gasoline (78%).
A small amount of hydrogen was detected in the exit gas of the reactor. This finding
suggested that Ni plays an important role in the hydrogen transportation from steam to
hydrocarbons, resulting in a high selectivity towards gasoline. The above findings thus
confirmed that PE-derived heavy oil can be efficiently upgraded to useful fuels when it
is reacted over Ni–REY in a steam atmosphere.
In order to verify the high efficiency of Ni–REY in an actual system, the catalytic
cracking of the oil derived from a mixture of PE and PET was conducted in a bench-
scale reactor (see Figure 6.8). The Ni–REY catalyst was placed in reactor 3, shown in
Figure 6.7. Figure 6.24 represents the carbon number distribution of the feed oil and the
product of the reaction over Ni–REY in a steam atmosphere. The fraction of aliphatic
heavy oil was completely converted to gasoline and kerosene, indicating that the Ni–REY
catalyst can be used for the upgrading of oil derived from a mixture of PE and PET. The

100
MFI
Selectivity towards gasoline/%
Conversion of heavy oil/%

80
100
Ni-REY REY
60
Ni-REY 80
40
REY 60
20 MFI
40
0
0 1 2 3 4 5
Number of sequences of reaction
and regeneration/-

Figure 6.23 Changes in the catalytic activity of the MFI-type zeolite, Ni–REY, and REY
catalysts during the repetition of a sequence of reaction and regeneration (T = 400◦ C,
W /F = 1 h), carrier gas in the reaction: nitrogen for MFI, steam for REY and Ni–REY.
(Reproduced with permission from Elsevier)
188 T. MASUDA AND T. TAGO

50
Catalyst : Ni-REY
100% Steam atmosphere
40 T = 400 ºC, W/F = 1.0 h

Gasoline
Yield/wt% 30
Kerosene Feed oil

20 Aromatics
Paraffins + Olefins

10

0
5 10 15 20+
Carbon number/-

Figure 6.24 Carbon number distribution of products obtained by the catalytic cracking
of oil derived from a mixture of PE and PET (reaction conditions: T = 400◦ C, W /F = 1 h).
(Reproduced with permission from Elsevier)

products with carbon numbers from C5 to C11 (i.e. the gasoline fraction) contained about
30% aromatics, almost the same result observed in the case of the products obtained from
the reaction over Ni–REY in H2 (see Figure 6.19). This finding suggests that cracking,
followed by hydrogenation, proceeds in a steam atmosphere due to the existence of Ni in
the catalyst.

4 CONTINUOUS PRODUCTION OF FUELS FROM WASTE PLASTICS

4.1 CONTINUOUS PRODUCTION OF FUELS

First, by decomposing waste plastics in a steam stream, it was shown that waste plastics
could be degraded without producing a carbonaceous residue. Furthermore, it was found
that the decomposition of oxygen-containing plastics can proceed at a uniform reaction
rate. Second, a new reactor was proposed to achieve high hold-up, high heat transfer, and
good contact between melted plastics with steam in order to accelerate hydrolysis. Using
this reactor, a mixture of waste plastics can be degraded, and further decomposition over
an FeOOH catalyst can also be achieved. Finally, the oil obtained was upgraded to a
variety of fuels, indicating gasoline and kerosene, over a Ni–REY catalyst in a steam
atmosphere. On the basis of these laboratory-scale experimental results, a pilot-scale plant
was built and the validity of the proposed chemical recycling process was examined.
Figure 6.25(a) shows a novel process for the continuous production of fuels from waste
plastics. The proposed process consists primarily of three reactors. A mixture of waste
plastics is fed into a pyrolytic reactor with heat-medium-particles stirred by a helical
impeller (Figure 6.25(b)), where melted plastics are hydrothermally decomposed with
steam and the random scission of C–C bonds. The produced mixture of a heavy oil
containing wax and sublimate material is carried by steam stream to the next reactor,
which is filled with an FeOOH catalyst (i.e. a catalytic hydrolysis reactor). The gaseous
CONTINUOUS PROCESS FOR FUEL PRODUCTION 189
compounds, including the vapors of the wax and sublimate materials, are passed through
an FeOOH catalyst bed, and the oxidative decomposition over the FeOOH catalyst with
steam proceeds. The quality of the oil thus obtained is further upgraded over Ni–REY
zeolite catalysts in the catalytic cracking reactor, and fuels, such as gasoline, kerosene,
and gas oil are produced. The weights of the FeOOH catalyst and the Ni–REY zeolite
catalyst were 45 and 20 kg, respectively. The feed rate of the waste plastics, F , was
45 kg/h (the time factor in the pyrolytic reactor, W/F = 1.0 h).

Cooling water
400°C
Catalytic hydrolysis Catalytic cracking
reactor reactor
(FeOOH catalyst) (zeolite catalyst)

500 ºC
Waste plastics Cooling
Products water
+ steam 200 °C Cooling water
Superheater

Pyrolytic
reactor
Wax
Superheated
steam Gas to Incinerator
Oil
Boiler

Residue
Water
Water

Plastics, steam

Ceramic balls

Impeller of
spiral in shape

Gaseous products

Figure 6.25 (a) Flow scheme of the continuous production of fuels from waste plastics;
(b) schematic diagram of the pyrolytic reactor used in the pilot plant. (Reproduced with
permission from Elsevier)
190 T. MASUDA AND T. TAGO
Table 6.9 Product yield at the outlet of the catalytic hydrolysis reactor and the catalytic cracking
reactor in the pilot-scale plant

Gas (C1 –C4 ) Fuel (C5 –C19 ) Heavy oil (C20+ ) Others
(wt%) (wt%) (wt%) (wt%)
Outlet of reactor with FeOOH 2.1 46.7 50.3 0.9
Outlet of reactor with zeolite 13.5 52.4 33.9 0.2

The product yields at the outlet of the catalytic hydrolysis reactor and at the catalytic
cracking reactor are listed in Table 6.9. The product yields of liquid fuel (gasoline-
kerosene) and heavy oil reached to 46.7 and 50.3 wt%, respectively, even at the outlet of
the catalytic hydrolysis reactor. The catalytic cracking reactor yielded 13.5 wt% gaseous
fuel, 52.4 wt% liquid fuels, and 33.9 wt% heavy oil, indicating that the heavy oil obtained
at the outlet of the catalytic hydrolysis reactor efficiently upgraded the quality of the fuel
oil. Using this method, the storage of carbonaceous residue and sublimate materials in
reactors, valves, and pipelines can be avoided. Moreover, recycling heavy oil at the outlet
of the catalytic cracking reactor into the pyrolytic reactor can enable an improvement in
the yield of liquid fuel to 72.4 wt%. Accordingly, the validity of the proposed process for
the continuous conversion of waste plastic mixtures to various fuels was demonstrated on
the basis of this pilot-scale plant.

REFERENCES

1. D. A. Anderson and E. S. Freeman (1961) The kinetics of the thermal degradation

of polystyrene and polyethylene, J. Poly. Sci., 54, 253–260 (1961).
2. H. Bockhorn, J. Hentschel, A. Hornung and U. Hornung Environmental engineer-
ing: stepwise pyrolysis of plastic waste, Chem. Eng. Sci., 54, 3043–3051 (1999).
3. H. Bockhorn, A. Hornung and U. Hornung Stepwise pyrolysis for raw material
recovery from plastic waste, J. Anal. Appl. Pyrol., 46, 1–13 (1998).
4. T. Kuroki, T. Sawaguchi, N. Ikebayashi, T. Ikemura and N. Sakikawa Pyrolysis
of polystyrene-prediction of product yield, Bull. Chem. Soc. Jpn., 11, 1766–1772
(1976).
5. T. Masuda, Y. Niwa, A. Tamagawa, S. R. Mukai, K. Hashimoto and Y. Ikeda Deg-
radation of waste poly(ethylene terephthalate) in a steam atmosphere to recover
terephthalic acid and to minimize carbonaceous Residue, Polym. Degrad. Stab., 58,
315–320 (1997).
6. T. Masuda, Y. Niwa, K. Hashimoto and Y. Ikeda Recovery of oil from waste poly
(ethylene terephthalate) without producing any sublimate materials, Polym. Degrad.
Stab., 61, 217–224 (1998).
7. C. H. Wu, C. Y. Chang, J. L. Hor, S. M. Shih and L. W. Chen Pyrolysis kinetics
of waste PVC at high conversion, Chem. Engng. J., 55, 87–92 (1994).
8. M. Bisi, C. Nicolella, E. Palazzi, M. Rovatti and G. Ferraiolo HCl formation via
pyrolytic degradation of polyvinyl chloride (PVC): an empirical approach to kinetic
modeling, Chem. Engng. Technol., 17, 67–72 (1994).
CONTINUOUS PROCESS FOR FUEL PRODUCTION 191
9. N. Lingaiah, M. A. Uddin, A. Muto and Y. Sakata Hydrochlorination of chlorinated
hydrocarbons over metal-carbon composite catalysts prepared by modified carboth-
ermal reduction method, J. Chem., Soc. Chem. Communication, August, 1657–1658
(1999).
10. Y. Sakata, Y. Shiraga, M. A. Uddin, A. Muto and K. Murata Dechlorination of
waste plastic derived oil using a carbon and iron oxide composite Gorden Research
Conf. January 1999.
11. T. Yoshioka, K. Furukawa and A. Okuwaki Chemical recycling of rigid-PVC by
oxygen oxidation in NaOH solution at elevated temperature, Polym. Degrad. Stab.,
67, 285–290 (1999).
12. T. Masuda, T. Kushino, T. Matsuda, S. R. Mukai, K. Hashimoto and Y. Yoshida
Chemical recycling of mixture of waste plastic using a new reactor system
with stirred heat-medium-particles in steam atmosphere, Chem. Eng. Journal, 82,
173–181 (2001).
13. A. R. Songip, T. Masuda, H. Kuwahara and K. Hashimoto Test to screen catalysts
for reforming heavy oil from waste plastics, Applied Catalysis B Environmental, 2,
153–164 (1993).
14. A. R. Songip, T. Masuda, H. Kuwahara and K. Hashimoto Production of high
-quality gasoline by catalytic cracking over rare-earth metal exchanged Y-type zeo-
lites of heavy oil from waste plastics, Energy and Fuels, 8, 136–140 (1994).
15. A. R. Songip, T. Masuda, H. Kuwahara and K. Hashimoto Kinetic studies for cat-
alytic cracking of heavy oil from waste plastics over REY zeolite, Energy and Fuels,
8, 131–135 (1994).
16. T. Masuda, H. Kuwahara, S. R. Mukai and K. Hashimoto Production of high quality
gasoline from waste polyethylene derived heavy oil over Ni–REY catalyst in steam
atmosphere, Chem. Eng. Sci., 54, 2773–2779 (1999).
17. I. C. McNeill and M. Bounekhel Thermal degradation studies of terephthalate poly-
esters: 1. Poly(alkylene terephthalates), Polym. Degrad. Stab., 34, 187–204 (1991).
18. G. Montaudoa, C. Puglisib and F. Samperi Primary thermal degradation mecha-
nisms of PET and PBT, Polym. Degrad. Stab., 42, 13–28 (1993).
19. Y. Sakata, M. A. Uddin, K. Koizumia and K. Muratab Thermal degradation of poly-
ethylene mixed with poly(vinyl chloride) and poly(ethyleneterephthalate), Polym.
Degrad. Stab., 53, 111–117 (1996).
20. T. Masuda, Y. Kondo, M. Miwa, T. Shimotori, S. R. Mukai, K. Hashimoto,
M. Takano, S. Kawasaki and S. Yoshida Recovery of useful hydrocarbons from
oil palm waste using ZrO2 supporting FeOOH catalyst, Chem. Eng. Sci., 56(3),
897–904 (2001).
21. J. K. Koo, S. W. Kim and Y. H. Seo Characterization of aromatic hydrocarbon for-
mation from pyrolysis of polyehtlene polystyrene mixtures, Resources Conservation
and Recycling, 5(4), 365–382 (1991).
22. L. Wheatley, Y. A. Levendis, P. Vouros Exploratory-study on the combustion and
PAH emissions of selected municipal waste plastics, Environmental Science and
Technology, 27(13), 2885–2895 (1993).
23. S. F. Sodero, F. Berruti and L. A. Behie Ultrapyrolytic cracking of polyethylene – a
high yield recycling method, Chem. Eng. Sci., 51(11), 2805–2810 (1996).
192 T. MASUDA AND T. TAGO
24. S. Lovett, F. Berruti and L. A. Behie Ultrapyrolytic upgrading of plastic wastes and
plastics heavy oil mixtures to valuable light gas products, Industrial and Engineering
Chemistry Research, 36(11), 4436–4444 (1997).
25. S. Ide, T. Ogawa, T. Kuroki and T. Ikemura Controlled degradation of Polystyrene,
Journal of Applied Polymer Science, 29(8), 2561–2571 (1984).
26. C. Vasile, H. Pakdel, B. Mihai, P. Onu, H. Darie and S. Ciocalteu Thermal and cat-
alytic decomposition of mixed plastics, Journal of Analytical and Applied Pyrolysis,
57(2), 287–303 (2001).
27. C. Vasile, P. Onu, V. Barboiu, M. Sabliovschi and G. Moroi Catalytic decompo-
sition of polyolefins.2. Consideration about the composition and the structure of
reaction-products and the reaction mechanism on silica-alumina cracking catalyst,
Acta Polymerica, 36(10), 543–550 (1985).
28. Y. Ishihara, H. Nanbu, T. Ikemura and T. Takesue Catalytic decomposition of poly-
ethylene using a tubular flow reactor system, Fuel 69(8), 978–984 (1990).
29. D. P. Serrano, J. Aguado and J. M. Escola Catalytic cracking of a polyolefin mix-
ture over different acid solid catalysts, Industrial and Engineering Chemistry Re-
search, 39(5), 1177–1184 (2000).
30. G. Manos, I. Y. Yusof, N. Papayannakos and N. H. Gangas Catalytic cracking of
polyethylene over clay catalysts. Comparison with an ultrastable Y zeolite, Industrial
and Engineering Chemistry Research, 40(10), 2220–2225 (2001).
31. G. Proticlovasic, N. Jambrec, D. Deursiftar and M. V. Prostenik Determination of
catalytic reformed gasoline octane number by high-resolution gas-chromatography,
Fuel 69(4), 525–528 (1990).
32. A. Corma and B. W. Wojciechowski The chemistry of catalytic cracking, Catalysis
Reviews-Science and Engineering, 27(1), 29–150 (1985).
33. L. L. Oliveira and E. Biscaia Catalytic cracking kinetic models. Parameter esti-
mation and model evaluation, Industrial and Engineering Chemistry Research, 28,
264–271 (1989).
34. A. Ayame, Y. Uemichi, T. Yoshida and H. Kanoh J. Jpn. Petrol. Inst., 22(5),
280–287 (1979).
35. Y. Yemichi, A. Ayame, T. Yoshida and H. Kanoh J. Jpn. Petrol. Inst., 23(1),
280–287 (1980).
36. V. W. Weekman Jr. and D. M. Nace AIChE J., 16(3), 397–404 (1970).
37. K. Hashimoto, T. Masuda and M. Arai Influence of solid acidity of catalyst on
selectivity of isomerization of n-hexane over Y type zeolite supporting platinum
catalyst, Sekiyu Gakkaishi, 33, 166–172 (1990).