Ind. Eng. Chem. Res.

2004, 43, 5133-5145

5133

Pilot-Scale Pyrolysis of Scrap Tires in a Continuous Rotary Kiln

Reactor
S.-Q. Li,*, Q. Yao, Y. Chi, J.-H. Yan, and K.-F. Cen
Department of Thermal Engineering, Tsinghua University, Beijing 100084, China, and
Department of Energy Engineering, Zhejiang University, Hangzhou 310027, China

The pilot-scale pyrolysis of scrap tires in a continuous rotary kiln reactor was investigated at
temperatures between 450 and 650 C. As the reactor temperature increased, the char yield
remained constant with a mean of 39.8 wt %. The oil yield reached a maximum value of 45.1 wt
% at 500 C. The pyrolytic derived oils can be used as liquid fuels because of their high heating
value (40-42 MJ/kg), excellent viscosity (1.6-3.7 cS), and reasonable sulfur content (0.97-1.54
wt %). The true-boiling-point distillation test showed that there was a 39.2-42.3 wt % light
naphtha fraction in the pyrolytic oil. The volatile aromatics were quantified in the naphtha
fraction using gas chromatography-mass spectrometry. The maximum concentrations of benzene,
toluene, xylene, styrene, and limonene in the oil were 2.09 wt %, 7.24 wt %, 2.13 wt %, and 5.44
wt %, respectively. The abundant presence of aromatic groups was also confirmed by functional
group Fourier transform infrared analysis. The concentration of polycyclic aromatic hydrocarbons
such as fluorine, phenanthrene, and anthracene increased with increasing temperature. The
pyrolytic char was composed of mesopores with a Brunauer-Emmett-Teller (BET) surface area
of about 89.1 m2/g. The char after carbon dioxide activation had a high BET surface area of 306
m2/g at 51.3% burnoff. The relationship between the surface area and the carbon burnoff was
almost linear. Both the original pyrolytic char and the activated char have good potential for
use as adsorbents of relatively large molecular species.
1. Introduction
The disposal of scrap tires is currently a major
environmental and economical issue. Recent estimates
of the annual arisings of scrap tires in North America
are about 2.5 million tonnes, in European Union about
2.0-2.5 million tonnes, and in Japan about 0.5-1.0
million tonnes.1,2 In China, more than 1.0 million
tonnes/year of tires are generated, which results in
about 0.22 million tonnes of used tires/year.3 Unfortunately, most of these scrap tires are simply dumped in
the open and in landfills in our country. Open dumping
may result in accidental fires with highly toxic emissions or may act as breeding grounds for insects.
Landfills full of tires are not acceptable to the environment because tires do not easily degrade naturally. In
recent years, many attempts have been made to find
new ways to recycle tires, i.e., tire grinding and crumbling to recycle rubber powders and tire incineration to
supply thermal energy. However, grinding is quite
expensive because it is performed at cryogenic temperatures and requires energy-intensive mechanical equipment, while incineration may produce hazardous polycyclic aromatic hydrocarbons (PAHs) and soot during
the combustion process.4
Pyrolysis as an attractive method to recycle scrap
tires has recently been the subject of renewed interest.
Pyrolysis of tires can produce oils, chars, and gases, in
addition to the steel cords, all of which have the
potential to be recycled. Within the past 2 decades, most
experiments have been conducted using laboratory-scale
batch units to characterize oil, char, and gas products.
* To whom correspondence should be addressed. E-mail:
[email protected].

Tsinghua University.

Zhejiang University.

Some conclusions from these laboratory-scale studies

are as follows:
(1) Pyrolytic char has potential as a low-grade carbon
black for a reinforcing filler or a printing ink pigment,4-6
as a carbon adsorbent after proper activation,7-9 and
as a solid or slurry fuel.10
(2) Pyrolytic oil, a mixture of parafins, olefins, and
aromatic compounds, possesses a high calorific value
(43 MJ/kg) and can be used directly as fuel or can be
added to petroleum refinery feedstocks.11-15 Oils can
also be properly cut based on their evaporating temperatures to solely produce valuable chemical feedstocks
(i.e., benzene, xylene, toluene, and D-limonene), or some
of the chemicals can be extracted with residue used as
fuel.16-22
(3) Pyrolytic gas contains high concentrations of
methane, butadiene, and other hydrocarbons, which
results in a high calorific value (35-40 MJ/kg) sufficient
to heat the pyrolysis reactor.23-25 The gas is generally
not sold as a commercial product but used as a process
heat resource because of its low yield (10 wt %).
However, these laboratory-scale studies differ greatly
from the field-scale practical applications. Thus, the
pilot-scale studies, which bridge between laboratoryscale data and large-scale applications, are of importance. However, the available literature on the continuous pilot-scale tire pyrolysis test is limited. According
to reports, the reactors that can fulfill continuous tire
pyrolysis include fluidized beds, vacuum moving beds,
two-stage moving beds, ablative beds, and rotary kilns.
Representative results for each reactor type are listed
below.
(1) Vacuum Moving-Bed Process (Laval University). Roy and co-workers have focused on the vacuum
pyrolysis process for more than 20 years. The reactor
development has gone through a 1 kg batch vacuum

10.1021/ie030115m CCC: $27.50 2004 American Chemical Society

Published on Web 06/29/2004

5134 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004

vessel, a 15 kg/h semicontinuous vacuum hearth, and

finally a 200 kg/h pilot-scale continuous moving bed.5,25,26
The advantage of the vacuum process is that the
evaporating volatiles can be immediately removed from
the reactor, which minimizes secondary reactions such
as thermal cracking, repolymerization, and recondensation. As a result, the oil yield is dramatically increased
at the expense of char and gas. Roy et al. also analyzed
end uses of pyrolytic char and oil. For the solid products,
they concluded that the Brunauer-Emmett-Teller
(BET) surface area and structure of the carbon black
after pyrolysis only changed a little. That is, a small
amount of carbonaceous deposits formed on the pyrolytic
char, which limits its use only as reinforcing filler of
low-grade rubbers.5,27 The use of pyrolytic char as an
additive for road bitumen was also discussed.28 For the
liquids, Roy et al. concluded that the distillation of oils
improved its economic value. For instance, the lighter
fractions can be used as a source of high-value chemicals
such as BTX (benzene, toluene, and xylene) and limonene and as an extender oil in rubber formulation,14-17
while the heavy fractions can be used as additives in
road bitumen or as a feedstock for coke production.5
(2) Hamburg Fluidized-Bed Process (University
of Hamburg). Kaminskys group launched the study
of waste polymer pyrolysis in the 1970s. The well-known
Hamburg process, using an indirectly heated fluidized
bed, was developed with the original aim to yield basic
chemicals such as BTX and carbon black for reinforcing
at temperatures in excess of 700 C.19 Laboratory-scale
plants have been built with capacities of 0.60-3.0 kg/h
of plastic, followed by pilot-scale plants with capacities
of 10-40 kg/h of plastic and 120 kg/h of used tires at
the University of Hamburg.29,30 In recent years, research
efforts have been conducted at lower temperatures in
order to improve the oil yield and to reduce the energy
input.4
(3) BBC Continuous Ablative Process. Black and
Brown reported limited details of the continuous ablation reactor from patent applications.31,32 Ablation is
achieved by sliding contact of the rubber particles on a
hot metal surface, which resulted in a high yield of
liquids because of the fast heating of the tire. BBC
Company (Canada) is one representative.33 Their tests
in a 50 kg/h pilot-scale reactor demonstrated efficient
heat and mass transfer, and a smaller test unit is
presently operating at a throughput of 10-25 kg/h. In
addition, a 1500-2000 kg/h commercial plant was sold
to the Castle Capital Company for use in Halifax, Nova
Scotia, Canada. Black and Brown31 found that liquid
yields of 54 wt % can be obtained at 470-540 C with a
0.88 s residence time and a 1.3 mm feeding size. The
BTX composition in oils was less than 3.5 wt %. A study
by Helleur et al.34 reported that pyrolytic char can be
converted to activated char with steam or carbon
dioxide. The activated char exhibited excellent quality
for the removal of organics and heavy metals from
aqueous solutions.
(4) Two-Stage Moving-Bed Process. The process
developed by the Universite Libre de Bruxelles (ULB)
is based on a two-stage pyrolysis mechanism.18 During
the first stage, the tire is depolymerized at a relatively
low temperature (500 C), while during the second
stage, the pyrolytic volatiles are postcracked at temperatures of 750-800 C. Their objective was just to
recycle the BTX and activated carbon. Cypres and
Bettens18 had obtained oil yields of 37.0-42.2 wt %,

Table 1. Proximate and Ultimate Analyses of Scrap Tire

(Air-Dried Basis)
case
proximate analysis (wt %)
moisture
ash
volatile matter
fixed carbon
ultimate analysis (wt %)
carbon
hydrogen
nitrogen
sulfur
oxygen
low heat value (kJ/kg)

this work

Roy et al.26

1.14
4.35
62.24
32.28

0.50
6.10
65.20
28.70

84.08
6.71
0.49
1.51
1.73
34923

81.50
7.10
0.50
1.40
3.40
36800

solid yields of 41.7-45.3 wt %, and gas yields of 6.019.5 wt %. The contents of BTX in oils are quite high,
with values of 36.4, 16.8, and 6.95 wt %, respectively.
(5) Continuous Rotary Kiln Process. A rotary kiln
pyrolyzer offers many unique advantages over other
types of reactors. For instance, the slow rotation of the
inclined kiln enables well mixing of wastes, and thereby
uniform pyrolytic products. Also the residence time of
solids can be easily adjusted to provide the optimum
conditions of pyrolysis reaction. Solid wastes of various
shapes, sizes, and calorific values can be fed into a
rotary kiln either in batches or continuously. The typical
rotary kiln processes include the Kobe Steel commercial
1 tonnes/h plant, the Italian ENEA Research Center
Trisaia pilot-scale plant, and the Kassel University
laboratory-scale setup.35-37 These rotary kiln reactors
widely exist in the process for pyrolysis or gasification,
but the technical- and pilot-scale testing data are quite
scarce. Only available data by Kawakami et al. were,
in fact, less systemic.38
Within the past years, we have been devoted to the
research and development of a rotary kiln pyrolyzer.
Pyrolysis of various solid wastes in a laboratory-scale
batch kiln39,40 and pilot-scale cold-model test in a
continuous rotary kiln41,42 has been successively performed. On the basis of these works, a continuous pilotscale rotary kiln reactor has been self-designed and
successfully operated.43 The objective in this paper is
to study the pilot-scale pyrolysis characteristics of scrap
tires in a continuous rotary kiln. Influences of the
reactor temperature on the product distribution as well
as the product properties and chemical compositions are
intensively discussed.
2. Experimental Section
2.1. Tire Samples. Shredded scrap tires were used
with particle sizes of 13-15 mm including the fabric
cords but not the steel. The proximate and ultimate
analyses on the air-dried basis and the calorific value
of scrap tires are listed in Table 1. The data of Roy and
Unsworth26 are given for comparison. The properties for
two tires were quite similar, except for a small difference
in the content of the carbon element.
2.2. Pilot-Scale Pyrolysis Process. The pilot-scale
process development unit is shown in Figure 1. It
consisted of a pyrolytic rotary kiln main reactor and
peripheral systems including a supply system (a storage
bin with a screw feeder), a tar condenser and reservoir,
a solid residue collection tank, a flue gas cleaner, a
demister filter, a gas burner, and an effluent gas
sampling system.43
The kiln was designed for continuous operation with
tire powder conveyed from a sealed container to the

Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5135

Figure 1. Schematic of a pilot-scale continuous rotary kiln pyrolysis reactor: 1, purge-gas tank; 2, PID controller; 3, supply and screw
feed; 4, mechanically frictional seal; 5, gear wheel; 6, external electric heater; 7, thermocouples; 8, computer; 9, primary packed condenser;
10, filter; 11, postcombustor; 12, gas sampling; 13, gas flowmeter; 14, induced fan; 15, secondary condenser with ice-water mixture; 16,
collector of solid residue.
Table 2. Operational Parameters of Each Run in the Pilot-Scale Tire Pyrolysis Test
reactor
temp/C

kiln rotating
rate/rpm

screw rotating
rate/rpm

kiln
slope/deg

pressure at the
kiln outlet/Pa

450
500
550
600
650

0.45
0.60
0.75
0.90
0.90

5
5
5
5
5

2
2
2
2
2

-10 to -20
-10 to -20
-10 to -20
-10 to -20
-10 to -20

electrically heated rotary kiln by a screw feeder. The

feeding rate was regulated from 10 to 30 kg/h. The kiln
diameter was 0.3 m, and the overall length was 3.0 m.
Three individual proportional-integral-derivative (PID)controlled heaters with a total power of 30 kW were
used to heat the kiln. The effective heated length of the
kiln was about 1.8 m.
The solids were transported in the kiln as a result of
inclination and rotation. At the kiln exit, the residue
char fell into a sealed 6 L container. The solid residence
time was adjusted by changing the kiln rotation to meet
the desired condition for the complete pyrolysis. The
evolved oil vapors and gases were quickly removed from
the reactor by a special induced fan to reduce the
residence time. The condenser, with four coils cooled by
water and a trap refrigerated by ice, was used to recover
pyrolytic oils into the reservoirs. The noncondensable
gases passed to the scrubbing unit to remove the acids
and finally passed to the burner. Two mechanical
friction-type seals, with two wearable sliding rings and
graphite between the rings, were used to seal the
reaction chamber at both the kiln inlet and outlet. The
pressure force on the seal was adjusted by varying the
tension in the springs.
2.3. Operating Conditions. Before the experiment,
nitrogen was used to purge the reactor for 3-5 min to
remove the air. The system was operated at a slightly
negative pressure of -20 to -10 Pa to prevent leakage
of pyrolytic gases. The operating pyrolysis temperature
was varied from 450 to 650 C. The residence time of
the solids must be chosen long enough to complete the
pyrolysis. Because the measurement of the solid residence time during pyrolysis in a hot kiln is very difficult,
the selection of the solid residence time can be optimized
in the following program: conducting the pyrolysis test
at a certain rotating rate with the char yield weighted,
altering the rotating rate in a step of 0.5 rpm and
repeating test, judging whether the difference between
char yields in those two tests is less than 1%, and finally
determining the rotating rate, i.e., the final residence
time. The detailed experimental conditions for each
pyrolysis run are presented in Table 2. Each run in
Table 2 was reproduced. The feeding rate was 12-15
kg/h, with a total of 50 kg of tire in each run.

2.4. Pyrolytic Oil Characterization. 2.4.1. Determination of the Fuel Properties. The fuel properties
of pyrolytic oils were analyzed according to the National
Standards of the Peoples Republic of China (PRC). The
analyses included the relative density or API gravity
by the pycnometer method (GB/T 2540-1981), the viscosity by the Ostwald method (GB/T 265-1988), the flash
point by the closed-cup method (GB/T 261-1983), the
moisture by the Karl Fischer method (GB/T 111331989), the carbon residue by the Conradson method
(GB/T 17144-2001), and the calorific value and the
sulfur content by the oxygen bomb method (GB/T 3881964).
2.4.2. Functional Group Compositional Analysis.
The functional group compositional analysis of the oils
was carried out by Nicolet-5DX type Fourier transform
infrared (FT-IR) spectroscopy. A thin uniform layer of
oils was placed on the sample cell, and peak heights
were normalized to the major C-H peak.
2.4.3. True-Boiling-Point (TBP) Distillation of
Oils. The FY-II crude oil TBP autoclave, developed by
Fu-shun Institute of Petroleum (PRC), was used for the
automatic distillation test. The distillation included
atmospheric pressure distillation (theoretical plate number, 14-18; reflux ratio, 5:1; ibp, 200 C) and reduced
pressure distillation (1.33 kPa; reflux ratio of 2:1; 200350 C). In each run, about 5-6 kg of pyrolytic oils was
fed into the autoclave. As the temperature was increased from room temperature to 350 C, the distillable
fraction was cut in steps of 25 C. Then each fraction
was weighed to calculate its yield.
2.4.4. Oil Analysis of Aromatic Class. After TBP
distillation, the oil was classified as the naphtha fraction
(ibp 200 C) and the heavy fraction (bp > 200 C).
The naphtha oil can be directly analyzed by Finnigan
Voyager gas chromatography-mass spectrometry (GCMS). The analysis was performed on a DB-5 capillary
column (30 m 0.2 mm i.d. 25 m), using a 200:1
split ratio and a carrier helium gas at 1.0 mL/min. The
distilled naphtha fraction, with an interval of 50 C, was
performed by GC-MS with a different temperature
program. As for the oil fraction of bp < 100 C, the oven
temperature was initially set at 35 C for 10 min, then
programmed to 150 C at 8 K/min, then moved to 250

5136 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004
Table 3. Pyrolytic Product Yields of Scrap Tires at Temperatures of 450-650 C and Their Comparison with Various
Typical Processes
pilot-scale
reactor
rotary kiln

vacuum processa,5,16
two-stage processb,18
fluidized-bed process4
ablative process31

hydrocarbon vapor yield/(wt %)

run no.

temp/C

char yield/
(wt %)

oil yield

gas yield

total volatile

AT01
AT02
AT03
AT04
AT05
H036
H045
H018
Rad-X
Rallye
V-10

450
500
550
600
650
480 (431)a
534 (510)
520 (500)
450 + 800
450 + 800
450 + 800
550
600
450

43.9
41.3
39.9
39.3
38.8
39.3
38.4
33.4
44.8
44.2
41.7
34.0
40.0
39.1

43.0
45.1
44.6
42.7
42.9
53.7
49.9
56.5
38.5
39.7
41.3
56.8
50.9
52.9

13.1
13.6
15.5
18.0
18.3
7.0
11.7
10.1
16.8
16.1
17.0
9.2
9.1
8.0

56.1
58.7
60.1
60.7
61.2
60.7
61.6
66.6
55.2
55.8
58.3
66.0
60.0
60.9

In a vacuum process by Roy et al., 480 (431) represents the heating medium and bed temperatures. b In the two-stage process by
Cypres et al., 450 + 800 represents the primary reactor and secondary reactor temperatures; Rad-X, Rallye and V-10 are types of waste
tire.

C at 25 K/min, and finally held for 5 min. The program

of the 100-150 C oil fraction was started at 50 C for
10 min, moved to 250 C at 8 K/min, and held for 5 min.
The program of the 150-200 C fraction was started at
70 C for 10 min, moved to 250 C at 8 K/min, and held
for 10 min. The MS operating conditions were as
follows: ion source, 270 C; electron energy, 70 eV, with
a range of m/z 50-500 m and a scan time every 1.0
ms.
The heavy oil (>200 C) was processed according to
the China SY5119-86 standard, the method for the
classes of soluble organics in rock and for the major
chemical classes in crude oil. The heavy fraction was
first solved using n-hexane, and the insoluble material
was a bituminous fraction. The glass columns packed
with silica gel and Al2O3 were used to separate oil into
chemical class fractions followed by GC-MS. The
column was then sequentially eluted with sequential
elutions of hexane, CH2Cl2/hexane, and ethanol/CHCl3
to produce aliphatic, aromatic, and polar fractions. The
aromatic fraction was analyzed by GC-MS to determine
identification of PAH. The Wax-10 capillary column (30
m 0.25 mm i.d.), with a 30:1 split ratio and a 1.0 mL/
min carrier helium, was used. The temperature profile
was 75 C for 3 min, followed by a 3 K/min heating to
180 C, then a 10 K/min heating to 270 C, and 20 min
of holding.
2.5. Pyrolytic Char Characterization. 2.5.1. Proximate and Ultimate Analyses. The pyrolytic char was
characterized using the following PRC standards: the
proximate analysis by GB/T 212-2001, the ultimate
analysis by GB/T 476-2001, and the size distribution by
GB/T 12496.2-1999.
2.5.2. Activation of Tire-Derived Chars. A nearly
isothermal, externally heated horizontal quartz-tube
reactor (28 mm i.d. 500 mm) was used to activate the
tire-derived char. Prior to activation, the char was sieved
to generate a 2.5-7.0 mm size fraction and dried to 105
C. About 15 g of char was placed in the reactor, with
carbon dioxide used as the activated agent. The experiments were conducted between 850 and 950 C. If not
specifically indicated, the activating-agent flow rate was
0.3 L/min and the activation time was 4 h. The tests
studied the influences of the activation temperature on
the degree of burnoff. The burnoff was defined as

burnoff (wt % daf) ) (w1 - w2)/w1 100

(1)

where w1 and w2 are the char mass (dry ash free basis)
before and after activation, respectively.
Only char obtained at a pyrolytic temperature of 550
C was selected for activation because temperatures in
excess of 500 C had little effect on the char characterization, as shown in the results.
2.5.3. Surface Area and Pore Structure of Pyrolytic or Activated Char. The surface areas of both the
pyrolytic char and the activated carbon were determined
by nitrogen adsorption at 77 K using a BET isotherm.
The instrument for the nitrogen adsorption experiments
was made by Quantachrome Corp. In addition, a
mercury intrusion porosimeter by Quantachrome Corp.
was used as a complementary tool to study the pore
structures of pyrolytic chars.
3. Discussion and Results
3.1. Tire Pyrolysis Reaction Scheme. From the
point of view of kinetics, tire pyrolysis is quite complex
and consists of more than hundreds of chemical reactions. Although each individual reaction is difficult to
determine, these reactions can be generally classified
into three groups: primary pyrolysis reaction (250-520
C), secondary postcracking reaction of pyrolytic volatiles (600-800 C) that strongly affect BTX yields, and
char gasifying reaction with CO2/H2O/O2 in the gases
(750-1000 C). The importance of each reaction group
is dependent on two parameters, the temperature and
time of residence, as indicated by the rates of reaction
in Arrhenius form. These multigroup reaction schemes
can be used to understand the pyrolysis results in the
following parts.
3.2. Yields of Pyrolytic Products. The yields of
pyrolytic char, oil, and gas (in difference) are presented
in Table 3 for temperatures of 450-650 C. Data for
other typical pilot-scale continuous pyrolyzers (vacuum,
fluidized-bed, ablative, and two-stage bed) are also given
in Table 3 as a contrast.
First, the solid char yield remains essentially constant
with a mean of 39.8 wt %, except for the relatively high
value of 43.9 wt % at a temperature of 450 C. The data
suggested that the kinetics of the primary pyrolysis
reaction did not complete at low temperature (450
C).44,45 Compared with other typical processes, Roy et
al.5,16 reported that the vacuum moving-bed process
with a feedstock of 21-42 kg/h had a char yield of 38.439.3 wt %. The two-stage process by Cypres and

Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5137
Table 4. Physical-Chemical Properties of Pyrolytic Oils in a Rotary Kiln and Comparison with the Vacuum Process
vacuum process5

rotary kiln
pyrolysis run
characteristics
density (kg/Nm3)
gravity (API)
viscosity at 50 C (cS)
flash point (C)
LHV (MJ/kg)
moisture (wt %)
carbon residue (wt %)
ash (wt %)
ultimate analysis
C (wt %)
H (wt %)
N (wt %)
S (wt %)
O + others (wt %)
H/C atomic ratio

450 C

500 C

550 C

600 C

650 C

car tire (no. H20)

truck tire (no. H22)

0.941
18.2
2.87
27.5
41.9
0.52
1.36
traces

0.962
15.0
2.44
17.0
41.7
0.88
1.78
traces

0.987
11.3
3.66
30.0
41.0
1.32
3.09
traces

0.955
16.1
1.63
17.5
41.6
0.80
2.01
traces

0.982
11.9
2.00
13.5
41.0
0.85
3.26
traces

0.950
17.4
9.7
28
43.7
0.3
1.30
traces

0.939
19.3
17.8
22
44.8
1.5
1.20
0.005

84.32
10.92
0.48
0.97
3.31
1.55

84.26
10.39
0.42
1.54
3.39
1.48

84.55
9.59
0.64
1.26
3.96
1.36

86.14
9.54
0.70
1.27
2.35
1.33

86.1
9.06
0.84
1.11
2.89
1.26

85.8
10.7
0.5
0.8
2.2
1.49

87.0
11.1
0.7
1.0
0.2
1.54

Figure 2. TBP distillation test of tire-derived oil at temperatures

of 500 and 600 C.

Bettens18 can produce 41.7-44.8 wt % char with three

different brands of tire at a temperature of 450 C.
Kaminsky and Mennerich,4 using a fluidized-bed reactor, showed that the char yield was about 40.0 wt % at
600 C. In addition, a char yield of about 39.1 wt % was
obtained in a continuous ablative process developed by
BBC company, Canada.31
Comparison with other processes indicates that the
char yield is not sensitive to the reactor types of various
heating rates, which differs greatly from the pyrolysis
of coal or biomass. This is because that scrap tire is a
synthetic chemical feedstock according to a certain
formula ratio. Tires are made of a complex blend of
elastomers, processing oils, carbon black, and a few
additives including mineral fillers, vulcanization agents,
palticizer, etc. Because the pyrolytic vapors (elastomers
and processing oils) produced during tire pyrolysis by
thermogravimetry (TG) are about 62-65 wt %,44,45 the
yield of pyrolytic solid residue should theoretically be
35-38 wt %. In the practical process, the char yield is
a little higher than this value due to the slight hydrocarbon deposits in the char, for example, about 39.8 wt
% in this experiments.
As shown in Table 3, the volatiles, consisting of oils
and gases, are constant at a value of 60 wt % with
various temperatures. However, the oil yields reached
a maximum yield of 45.1 wt % at 500 C and then

decreased, as the temperature increased from 450 to 650

C. The gas yield correspondingly increased from 13.1
to 18.3 wt % as a result of the serious secondary
postcracking of vapors at higher temperature. The
maximum yield of oils at 500 C can be attributed to
the balance for the competition between the primary
pyrolysis reaction and secondary postcracking reaction.
The oil yield in the current kiln process is compared
with that of other reactor processes in Table 3. Although
the reactor configuration has little influence on the char
yield, the oil yield varied greatly for the different reactor
types. The vacuum process by Roy and co-workers5,16
had a maximum oil yield of 53.7 wt % at 431 C. The
Hamburg fluidized-bed process by Kaminsky and Mennerich4 produced 56.8 wt % oil at 550 C. The continuous
ablative process had a maximum oil yield of 52.9 wt
%.30,31 The maximum oil yield of 41.3 wt % was obtained
by Cypres and Bettens18 using the two-stage moving
bed, which is lowest among all of the processes. The
different oil yields of various reactors due to the different degrees of vapor postcracking can mainly be attributed to the difference of the vapor residence time
in the high-temperature zone in each process.
It is noted that Kawakami et al.,38 using a similar
rotary kiln reactor, obtained a maximum oil yield of 53.0
wt % at 500 C, which was higher than the data in this
experiment. The effective heated length is 1.8 m, and
the length of the kiln outlet is 0.6 m. It can be explained
that the ratio of length to diameter (L/D ) 8) in this
paper might be larger than that from the study of
Kawakami et al. The larger the ratio of L/D, the higher
the residence time of the vapor. A high vapor residence
time in the high-temperature zone caused the reduction
of the oil yield in this paper. According to the authors
experiences, the concentrations of valuable chemicals
such as BTX and limonene in the oils exerted more
important impacts on the economic viability of the
process than the total oil yield. The quantification of
the valuable chemicals was essential. It would be
discussed in the following part.
3.3. Characterization of Pyrolytic Oils. The pyrolytic oils are derived from (1) processing oils in the
original tire, (2) evaporating and decomposing of elastomers such as natural rubber, butadiene rubber, or
styrene-butadiene rubber, and (3) other organic additives in tires. The tire-derived oils are mostly unrefined
oils with a wide range of boiling fractions.
3.3.1. Fuel Properties. Table 4 presents the fuel
properties of the pyrolytic oils carried out in the pilot-

5138 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004

scale rotary kiln. The properties of oils from the vacuum

process by Roy et al.5 are also given as a contrast.
The calorific value of the tire oils is similar for all of
the tests, with values of 41.0-41.9 MJ/kg, which is
higher than that of scrap tires (34.9 MJ/kg). Moreover,
this high calorific value of the derived oil is comparable
with that of light fuel oil, indicating the potential for
use as liquid fuels for industrial furnaces and power
plants.
The viscosity of pyrolytic oils was 2.0-3.7 cS (at 50
C). It is similar to that of diesel fuel (1.8-8.0 cS at 20
C) but at least 1 order of magnitude lower than that of
marine or furnace fuel oils. The fuel viscosity is important because it affects the fuel flow through pipes, the
fuel atomization, and the performance and wear of
diesel pumps. The viscosity of oils derived from the
vacuum process by Roy et al.5 was about 4 times that
of the viscosity of oils derived from the current rotary
kiln process. The oils derived from a fixed bed by
Cunliffe and Williams1 had a viscosity of 2.38 cS at 60
C, which agrees well with the pyrolytic oils in this
paper. Generally, the shorter residence time of vapor
caused by the vacuum pump reduced the secondary
postcracking reaction and improved the oil yield but also
increased the amounts of long-chain molecules in the
oils, which resulted in the high oil viscosity and limited
its usage. The oil density was 0.94-0.99 kg/m3 over a
temperature range of 450-650 C, while the corresponding API gravity was 11.3-18.2. The density of
pyrolytic oils was higher than that of diesel oils (0.78
kg/m3) and approached that of marine fuel oils (0.98 kg/
m3). The effects of temperature on both the oil viscosity
and density were indistinct within the error limits of
measurement.
The flash point of liquid fuel indicates the temperature at which the oil begins to evolve vapors in sufficient
quantity to form a flammable mixture with air. As seen
in Table 4, the flash point of the tire-derived oils was
13.5-30 C with no particular trend with increasing
temperature. Roy et al.5 reported a value of 22-28 C,
and Cunliffe and Williams1 reported 14-18 C. The
flash point of the tire-derived oil was lower than that
of the petroleum-refined fuels. For example, light diesel
oil has a required minimum flash point of 45 C and a
heavy gas oil of 65 C. The low flash point of the tire oil
can be attributed to the wide range of unrefined oils in
the mixture including some light low-boiling-point
hydrocarbons. The low flash point will complicate storage of the oils. The moisture of 0.5-1.3 wt % will also
be noticeable during the pyrolytic oil usage.
The residue carbon in pyrolytic oil was 1.3-3.3 wt %.
The corresponding value reported by Roy et al. was 1.21.3 wt %, and that by Williams was 0.5-2.2 wt %.5-13
A typical diesel fuel would have a carbon residue of
approximately 0.2 wt %; however, fuel oils used in a very
large diesel engine may have carbon residues of up to
12 wt %. Thus, the tire-derived oil should be used in a
large diesel engine or industrial boiler rather a microscale typical diesel.
The temperature had only a slight effect on the
elemental content of pyrolytic oils. As the temperature
increased, carbon increased slowly from 84% to 86% and
hydrogen decreased rather appreciably from 10.9 to 9.0.
The hydrogen/carbon (H/C) ratio was about 1.26-1.55,
which is somewhat lower than that of diesel oil (1.735)
and light fuel oil (1.740). With increasing temperature,
the H/C ratio decreased; especially, the drop gradient

Table 5. Carbon, Hydrogen, and Nitrogen Profiles in

Each Distillate Fraction for Pyrolyic Oils Produced at
600 C
distillant

carbon

hydrogen

nitrogen

H/C ratio

total oils
50-100 C fraction
100-150 C fraction
150-200 C fraction
200-250 C fraction
250-300 C fraction
300-350 C fraction

86.14
87.03
89.83
88.50
86.56
87.22
88.28

9.54
9.93
9.50
9.51
9.06
9.27
9.24

0.70
0.46
0.56
1.13
1.71
1.49
1.12

1.33
1.37
1.27
1.29
1.26
1.28
1.26

between 500 and 550 C is sharper than that in other

zones. This implies that the postcracking of evolved
vapor, including chain cracking, dehydrocyclization, and
aromatization, becomes serious from this zone onward.
Thus, the aromatics in the tire-derived oil increased, as
indicated by the reducing H/C ratio. In addition, the
nitrogen content increased with increasing temperature
because the nitrogen in the char or gas participates in
the aromatization process to form polycyclic aromatic
nitrogen hydrocarbons (PNAHs). The sulfur content was
0.97-1.54 wt % and similar to that of light fuel oil,
which is typically about 1.4-1.5 wt %.
Comparison of the pyrolytic oils from the rotary kiln
and the vacuum process5 indicated that both the nitrogen and sulfur contents are similar, the hydrogen
contents are similar below 500 C but differ above 500
C, the carbon content in the rotary kiln is lower than
that in the vacuum process by 1-2%, and the oxygen
content is higher than that in the vacuum process.
Table 5 shows the carbon, hydrogen, and nitrogen
contents in different distillate fractions of pyrolyic oil
at 600 C. The H/C ratio was quite constant with a mean
of 1.27, except for a value of 1.37 in the 50-100 C
distillate fraction. Therefore, the aromatization degrees
of the fractions in excess of 100 C were similar. The
nitrogen content in the low-boiling fractions (<150 C)
was lower than that in the high-boiling fractions (>150
C), which suggests that the nitrogen exists mainly as
a high-boiling-point PNAH such as caprolectum, quinoline, etc.
3.3.2. TBP Distillation Test of Pyrolytic Oil. The
TBP distillation test is one of the best approaches to
reflect the true distillation of oil. It plays an important
role in developing oil refinery or extraction processes.
Although many researchers have emphasized the importance of the TBP distillation test, little literature is
found about the TBP test because it needs a great
amount of pyrolytic oils (about 5 kg). Roy et al.5 once
reported the simple test of the pyrolytic oil distillation.
Figure 2 shows the results of the TBP distillation test
of the pyrolytic oils at temperatures of 500 and 600 C.
The pyrolytic oils had fractions with a wide range of
boiling points, mainly 39.2-42.3 wt % light naphtha (ibp
200 C), 32.4-33.2 wt % medium fractions that can
be used as diesel oil (200-350 C), and 25.5-28.5 wt %
heavy fractions (>350 C). Comparison of these results
with those of vacuum pyrolysis in Table 6 indicated that
(1) the light fraction (39-42.3 wt %) is higher than that
in the vacuum pyrolysis oil (26.8 wt %) and (2) the
contents of medium fractions of both studies are similar
at about one-third of the total oil. Kawakami et al.38
reported a relatively low light fraction of 24.5 wt %. It
is concluded that the significant postcracking of the oil
resulted in not only a low oil yield but a highly valuable
light naphtha fraction as well.
Figure 3 shows the distillate fractions at intervals of
25 C of pyrolytic oil produced at 600 C. The light

Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5139
Table 6. Distillate Fractions for Pyrolytic Oils and Crude Oils
oil fraction
tire-derived pyrolytic oils (%)
crude oils produced in China (%)

oil samples

light fraction
(ibp 200 C)

medium fraction
(200-350 C)

heavy fraction
(>350 C)

600 C
500 C
vacuum process5
Daqing
Shengli
Xinjiang

42.3
39.2
-26.8
10.7
7.6
15.4

33.2
32.4
30.7
17.3
26.3
26.0

24.5
28.4
42.5
69.0
66.0
58.6

Figure 3. Distillate fractions at intervals of 25 C for tire-derived

oil produced at 600 C.

naphtha fraction (ibp 200 C) and the medium

fraction (200-350 C) have normal distribution, with
mean values of 150 and 275 C, respectively. This
distribution favors the cut of pyrolytic oil as the naphtha
and diesel oil in potential applications.
In addition, Table 6 also compares the distillation
fractions of the pyrolytic oil and three crude oils in
China (Daqing, Shengli, and Xinjiang). The pyrolytic oil
is much lighter than the crude oil. The crude oil has a
58.6-69.0 wt % heavy fraction (>350 C), while the
pyrolytic oil only contains a 24.5-28.4 wt % heavy
fraction. Therefore, the properties of the pyrolytic oil
are superior to those of the crude oil.
3.3.3. Functional Group Compositional Analysis
of Pyrolytic Oil. The group composition of the pyrolytic
oil was determined by FT-IR spectroscopy. Figure 4
illustrates the spectra of the tire-derived pyrolytic oils
at temperatures of 450-650 C. Using the FT-IR spectra
of the oil at 600 C as an example, the following
conclusions can be drawn:
(1) The dCH stretching vibrations at 3100-3000 cm-1
indicated the presence of aromatic or alkene groups. The
aromatic groups can be distinguished from the alkenes
by the breathing vibration, which was defined as the
conjugated CdC vibrations occurring in the vicinities
of both 1600 and 1500 cm-1. In pyrolytic oil spectra, two
conjugated peaks at 1604 and 1495 cm-1 confirmed the
presence of the aromatic groups. Furthermore, single,
polycyclic, and substituted aromatic groups can be
identified by the absorbance peaks of C-H cyclic
deforming vibrations at 900-675 cm-1 as well as the
resonating peaks of C-H and C-C deforming vibrations
at 2000-1600 cm-1.
(2) The absorbance peaks at both 1675-1575 and
950-875 cm-1 represented the CdC stretching vibrations, which confirm the presence of alkenes in the
pyrolytic oil.

(3) The two distinctly intensive peaks at 2906 and

2869 cm-1 represented the C-H stretching vibrations
and were indicative of alkanes. Furthur detail concerning the group compositions can be deduced from the
C-H deformations at 1500-1300 cm-1 and the C-H
cyclic deformations at 1000-650 cm-1. For instance, the
C-H deformations for the tire pyrolytic oil occurred at
1455 cm-1, while the C-H cyclic deformations at 775735 cm-1. Thus, it is deduced that the C-H group in
the pyrolytic oil was probably -C-(CH3)n-C- (n < 4)
or mC-CH3.
(4) The remarkably wide absorbance peaks between
3700 and 3200 cm-1 represented the polar compounds
in the pyrolytic oil. These peaks can be possibly attributed to the resonance of the O-H and N-H groups,
which indicated an acylamino group. Moreover, the
presence of CdO<, indicated by the absorbance peak
at 1705 cm-1, enabled the O-H group spectra to become
wide. However, Williams et al.13,24 only found the CdO<
group but not the O-H group in the spectra of the tirederived oil.
Figure 4 also shows the FT-IR spectra for the oils in
relation to the pyrolysis temperature (450-650 C). The
FT-IR spectra peaks for all of the oils produced at
different temperatures were normalized to the main
CH2 peak in each spectrogram, and consequently relative changes of the functional group composition were
determined. As the pyrolysis temperature increased, the
intensities of some peaks changed dramatically, indicating that the corresponding group composition changed.
For instance, the aromatic groups, indicated by absorbance peaks at 3100-3000, 1604 (1495), and 900-675
cm-1, increased in intensity with increasing temperatures. In addition, the 3000-2800 cm-1 absorbance
peak, indicative of alkanes, and the 1675-1575 and
950-875 cm-1 peaks, indicative of alkenes, showed a
decrease in intensity as the temperature increased from
450 to 650 C. The FT-IR results agree well with the
aforementioned reaction mechanisms that the increasing temperature improves the secondary postcracking
of volatile products and further aromatization of the
pyrolytic oil.
3.3.4. Volatile Aromatic Hydrocarbon Analysis.
Table 7 presents the light volatile aromatics in pyrolytic
oils. The high concentration of single-ring aromatics
such as benzene, toluene, xylene, dimethylbenzene,
styrene, and indene was suggested as a potential high
value feedstock in the plastics/polymer industry, which
could offset the disposal cost. The high pyrolytic temperature promoted the production of volatile aromatics.
As the temperature increased from 450 to 600 C, the
benzene concentration increased from 0.40 to 2.09 wt
%, the toluene dramatically from 2.27 to 7.24 wt %, the
xylene from 1.54 to 2.13 wt %, the 1,2-dimethylbenzene
from 2.34 to 6.74 wt %, and the 1,3-dimethylbenzene
from 0.54 to 1.46 wt %. However, as the temperature
further increased from 600 to 650 C, there was even a
small drop in the BTX concentration. This implies that

5140 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004

Figure 4. FT-IR spectra of tire-derived pyrolytic oils at various temperatures.

Table 7. Concentrations of Volatile Aromatic
Hydrocarbons in Pyrolytic Oils (wt %)
tire pyrolysis temperature (C)
volatile hydrocarbons

450

500

550

600

650

benzene
toluene
xylene
1,2-dimethylbenzene
styrene
1,3-dimethylbenzene
trimethylbenzene
C10-based aromatics
indene
limonene

0.401
2.269
1.538
2.398
1.207
0.539
4.626
3.815
0.831
5.440

1.341
2.789
1.864
4.319
1.302
1.085
7.325
4.937
0.381
1.883

1.488
5.162
2.051
5.641
1.423
1.406
6.265
5.442
0.900
0.419

2.107
7.244
2.127
6.774
1.436
1.553
6.331
5.014
0.447
0.122

2.093
7.056
2.015
6.095
2.640
1.464
6.099
4.073
0.382
0.070

Diels-Alder aromatization would occur and make singlering aromatics convert to more PAHs. Within temperature ranges of 450-650 C, the styrene concentration
always increased. Indene, used to produce indenecoumarone resins in industrial applications, had a
maximum concentration of 0.90 wt % at 550 C.
Limonene, with a high value because of its applications in the formulation of industrial solvents, resins,
and adhesives, has attracted the most interest. Table 7
indicates that limonene had a high concentration of 5.44
wt % at lower temperature and dramatically decreased
to 0.07 wt % as the temperature increased from 450 to
650 C. From the point of view of reaction kinetics, some
researchers found that the degradation of limonene to
form a range of BTX and styrene occurred above 500
C.1,17 This greatly agreed well with the limonene data
in Table 7. So, it is concluded that the heat transfer in
the rotary kiln was well enough to have a rapid response
to the reaction kinetics.
The rotary kiln process in this paper can produce
more volatile aromatics than other one-stage processes
aforementioned. Within similar temperature ranges, the
concentrations of benzene, toluene, xylene, styrene, and
limonene in a laboratory-scale fixed bed were up to
0.29, 1.77, 1.68, 0.36, and 3.13 wt %, respectively.1 In a
vacuum process, the concentrations of benzene, toluene,
xylene, and limonene were 0.68, 1.86, 1.60, and 4.00 wt
%. However, the maximum concentrations of benzene,
toluene, xylene, styrene, and limonene in this experiment were 2.09, 7.24, 2.13, and 5.44 wt %, respectively.
Because the length of the kiln can be varied, the rotary
kiln pyrolyzer is flexible with regard to the gas residence

time as well as time/temperature profile. In this paper,

the richer single-ring aromatics are caused by the larger
kiln length.
3.3.5. PAH Analysis. Table 8 presents the concentrations of several main PAHs in the tire-derived oils
in relation to the temperature. The increasing temperature enhanced the formation of PAHs. For instance,
as the temperature increased from 450 to 650 C, the
concentrations of the fluorene and alkylated fluorene
increased from 1.17 to 2.57 wt %, those of the phenanthrene homologues from 2.15 to 6.51 wt %, and those of
the anthracene homologues from 1.13 to 2.15 wt %. As
far as the two-ring naphthalene and its alkylated
compounds were concerned, their concentrations first
increased from 8.61 to 13.14 wt % as the temperature
increased from 450 to 500 C. However, there was a drop
from 13.14 to 10.06 wt % as the temperature further
increased to 650 C. It is deduced that a two-ring
alkylated naphthalene might take an active part in
Diels-Alder aromatization to form a three-ring PAH
such phenanthrene and anthracene above 500 C. The
naphthalene homologues included naphthalene, methylnaphthalene, ethylnaphthalene, dimethylnaphthalene, and trimethylnaphthalene. The recycling of naphthalene from oils is less possible than BTX. Because the
concentration of alkylated naphthalene was high, two
modifications of the rotary kiln pyrolyzer should be
considered to maximize the production of single-ring
aromatics in the future: (1) multiple heating zones with
different temperatures along the kiln length in order
to have the desired time/temperature profiles; (2) a
solid-bed catalyst process adjacent to the kiln outlet.
The concentrations of four-ring PAHs were relatively
low, e.g., total alkylated pyrene (0.49-0.83 wt %),
fluoranthene (0.07-0.15 wt %), and benzo[b]fluorine
(0.06-0.21 wt %). The increment of four-ring PAHs in
relation to the temperature was less clear than that of
three-ring PAHs.
3.4. Characterization of Pyrolytic Chars. The
pyrolytic char, 38.8-43.9 wt %, is mainly derived from
(1) the commercial carbon black filler used in tire
production (30 wt %); (2) the carbonaceous deposits
formed by condensation and dealkylation reactions of
hydrocarbons, which is absorbed on the char surface;
(3) the residue hydrocarbons from original elastomers
or processing oils; (4) other inorganic additives such as

Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5141
Table 8. Concentrations of PAHs in Pyrolytic Oils (wt %)
tire pyrolysis temperature (C)
volatile hydrocarbons

450

500

550

600

650

naphthalene homologues
fluorene homologues
phenanthrene homologues
anthracene homologues
pyrene homologues
fluoranthene
benzo[b]fluorene

8.618
1.169
2.151
1.130
0.830
0.070.
0.135

13.144
1.012
2.356
0.736
0.756
0.086
0.058

11.298
1.513
3.398
1.091
0.489
0.066
0.105

10.143
2.270
3.644
1.362
0.544
0.079
0.120

10.058
2.568
6.510
2.152
0.685
0.148
0.214

Table 9. Size Distributions of Tire-Derived Char for Various Pyrolysis Temperatures

pyrolysis reactor temperature (C)
characteristics
size distribution
<0.074 mm
0.074-0.125 mm
0.125-0.25 mm
0.25-0.335 mm
0.335-1.00 mm
1.00-1.25 mm
1.25-2.0 mm
2.0-2.5 mm
2.5-5.1 mm
5.1-7.mm
7.0-10 mm
>10 mm
average diameter
weight-averaged
surface-averaged

450

500

550

600

650

0.04
1.08
3.03
4.14
21.48
3.54
16.72
10.75
22.46
1.23
11.87
3.67

0.30
1.55
3.15
6.31
18.38
5.17
14.84
9.65
24.65
1.11
10.63
4.27

0.55
2.18
4.74
5.40
21.37
3.53
14.35
8.87
22.25
1.70
10.69
4.37

0.95
3.20
3.14
3.01
14.18
4.59
13.61
9.58
27.69
2.03
14.03
3.99

0.45
2.18
3.88
4.09
18.21
4.50
14.15
10.17
25.79
1.19
11.82
3.57

3.09
1.00

3.08
0.86

3.02
0.74

3.40
0.75

3.14
0.82

Table 10. Proximate Analysis of Pyrolytic Char at Temperatures of 450-650 C (Air-Dry Basis)
pyrolysis reactor temperature (C)
proximate analysis (wt %)
moisture
ash
volatile matter
fixed carbon
calorific value/(kJ/kg)
V/FC ratio

450

500

550

600

650

3.40
12.51
16.61
67.47
31.2
0.246

2.35
12.32
16.14
69.19
31.5
0.232

1.28
14.58
6.92
77.22
30.0
0.090

1.98
14.30
5.86
77.93
30.4
0.075

1.48
13.94
6.27
78.32
30.1
0.080

zinc and calcium compounds. The properties of pyrolytic

char differ from those of commercial carbon black
because of the above complex sources.
3.4.1. Size Distribution of Char. The original 15
mm tire powder particle underwent a thorough thermal
cracking in the kiln. Table 9 presents the size distribution of the tire-derived char in relation to the pyrolysis
temperature. The weight- and surface-averaged diameters are also given. The data show that 83-84% of the
char particles had fractions with sizes of less than 5.1
mm. The fraction above 10 mm was less than 4.5%. The
weight- and surface-averaged diameters were 3.02-3.40
and 0.74-1.00 mm, respectively. The pyrolysis temperature had little effect on the char size distribution
within 450-650 C. It is concluded that the rotary kiln
reactor can produce more uniform char than other
processes, where the kiln rotation plays a crucial role
in the thorough mixing of char.
3.4.2. Proximate and Ultimate Analyses of Char.
(1) Proximate Analysis. Table 10 shows the proximate
analysis (air-dry basis) of pyrolytic char for temperatures from 450 to 650 C. The volatile matter content
decreased while the fixed carbon content increased with
increasing temperature. The V/FC ratio, defined as the
ratio of volatile matter to fixed carbon, underwent a slow
drop under 500 C, then a sharp drop between 500 and
550 C, and finally again a slow drop over 550 C. It is
concluded that the temperature in excess of 550 C
favors the evaporation of volatile matters because the

primary pyrolysis reaction has been completed. In

addition, Roy et al.27 reported that the volatile matter
content of pyrolytic char in a vacuum process was 2.17.9 wt %, which is similar to that for pyrolytic char over
550 C in this paper.
The calorific value of pyrolytic char was 30.0-31.5
MJ/kg but slowly decreased with increasing pyrolysis
temperature, among which the drop between 500 and
550 C is relatively large. That is, the calorific value is
in relation to the volatile matter content. Collins et al.10
reported a calorific value of 30 MJ/kg and an ignition
temperature of approximately 510 C for the tire-derived
char.
The ash content of pyrolytic char, 12.32-14.58 wt %,
is high compared to that of most activated carbons used
for flue gas or wastewater cleanup. The high ash content
for the tire-derived char has also been reported previously. For example, Roy et al.27 reported the ash content
of pyrolytic char at 10.6-12.2 wt %, Darmstadt et al.46
at 14.7 wt %, and Merchant and Petrich9 at 11.1-11.9
wt %. The major components of the ash of pyrolytic char
were zinc and calcium compounds.47
(2) Ultimate Analysis. Table 11 shows the elemental
analysis of pyrolytic char for various temperatures with
additional contrast data. The chars contained about
81.00-82.17 wt % carbon on a dry basis, which contributed more than 90% to the organic matters of the char.
The hydrogen content decreased slightly with increasing
temperature, indicating lower formation of solid hydro-

5142 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004
Table 11. Ultimate Analysis of Pyrolytic Chars at Various Temperatures (Dry Basis) and Comparison with Various
Typical Processes
elemental composition (wt %)
reactor
rotary kiln

vacuum process (Roy et al.27)

ablative process (Helleur et al.34)
batch fixed-bed process (Cunliffe and Williams47)
a

pyrolysis temp/C

carbon

hydrogen

nitrogen

sulfur

oxygen

ash

H/C

450
500
550
600
650
500 (10.3 kPa)a
500 (0.3 kPa)b
550
500
600

82.13
82.17
80.82
81.00
81.03
82.56
82.6
80.3
83.8
83.9

2.10
2.28
1.46
1.38
1.98
0.97
0.7
1.3
0.8
0.7

0.54
0.61
0.53
0.51
0.45
1.04
0.7
0.3
0.6
1.0

2.28
2.32
2.41
2.53
2.40
2.06
2.5
2.0
2.0
2.0

nd
nd
nd
nd
nd
0.49
2.0
0.6
nd
nd

12.95
12.62
14.77
14.58
14.15
12.89
11.5
15.2
11.9
12.1

0.31
0.33
0.22
0.20
0.29
0.14
0.103
0.19
0.113
0.100

Data from a 12 kg/h pilot-scale vacuum multihearth reactor. b Data from a laboratory-scale vacuum batch retort reactor.

Figure 5. Pore volume and surface profiles of pyrolytic char and

activated carbon measured using the N2 adsorption method.

carbons. The decrease of the volatile matter content with

increasing temperature supports this supposition. Roy
et al.27 reported a hydrogen content of 0.70-0.97 wt %
in a vacuum process, while Cunliffe and Williams47
reported one at 0.7-0.8 wt % in a fixed bed, which are
both about half that in the present experiments. The
high hydrogen content or H/C ratio of pyrolytic char
(especially below 500 C) in this paper can be attributed
to the carbonaceous deposits formed by the condensation
and dealkylation of hydrocarbons due to a relatively long
residence time of vapor.
Within ranges of 500-550 C, the nitrogen and sulfur
contents of the chars were 0.45-0.61 and 2.28-2.53 wt
%, respectively. They both appear to not be influenced
by the pyrolysis temperature. Similar sulfur content
data were reported by Roy et al.27 (2.0-2.5 wt %),
Helleur et al.34 (2.0 wt %), and Cunliffe and Williams47
(2.0 wt %). Higher nitrogen contents were also found
by Roy et al.27 (0.70-1.04 wt %) and Cunliffe and
Williams47 (0.6-1.0 wt %).
3.4.3. Pore Characteristics of Pyrolytic Char.
Common methods to measure pore distributions include
the nitrogen adsorption method and the mercury intrusion method. The former is limited to a maximum pore
size of 200 nm, while the latter has a measuring range
of 3.6-104 nm. Both methods were used to give the fullscale analyses of pore sizes of pyrolytic char.
(1) N2 Adsorption Method. The 2.5-5.1 mm size
fraction of the pyrolytic char derived at 550 C was
crushed to about 200 m for the N2 adsorption test as
well as the BET determination. Figure 5 illustrates the
pore volume and surface profile variation with pore

diameters for pyrolytic char. The total pore volume of

char or carbons includes the macro- (>50 nm), meso(2-50 nm), and micropore volumes (<2 nm), among
which the micropore volume plays the more important
role in determining the absorption ability of chars or
carbons.9 The pore sizes should be only slightly larger
than the sizes of the molecules of absorbed materials.
Therefore, the effective adsorptive pore sizes of chars
or activated carbons are generally less than 2 nm. As
seen in Figure 5, the pore volume of pyrolytic char was
much smaller than that of the commercial activated
carbon by at least 1 order of magnitude, especially in
the micropore volume. The integrated pore volume of
pyrolytic char was 0.053 cm3/g (cubic centimeters per
gram), while that of the commercial activated carbon
was 0.5 cm3/g. Therefore, the tire-derived char had less
microporosity than the commercial activated carbon.
Figure 5 also shows the specific surfaces of pyrolytic
char as a function of the pore sizes. The mesopores had
a relatively large fraction in pyrolytic chars. The surface
area of chars can also be measured using the BET
isotherm. The BET surface area of the 2.5-5.1 mm size
pyrolytic char was 89.1 m2/g. Some investigators have
reported similar tire-derived char BET surface areas
using different pyrolysis reactors; for instance, the BET
surface area of the pyrolytic char from the vacuum
moving bed by Roy et al.5 was 95 m2/g, that from the
fluidized bed by Kaminsky et al.4 was 76-85 m2/g, that
from the ablative process by Hellur et al.34 was 51-67
m2/g, that from the batch rotary kiln by San Miguel et
al.7 was 75-85 m2/g, and that from the batch fixed bed
by Cunliffe and Williams47 was 61-68 m2/g. Generally,
the BET surface area of pyrolytic char was approximately 1/10 that of the commercial activated carbon.
(2) Mercury Intrusion Method. The meso- and
macropores in the pyrolytic char measured using the
mercury intrusion porosimeter can be regarded as
auxiliary information for the char pore characteristics.
Figure 6 compares the pore volume profiles between the
pyrolytic char and commercial carbon as a function of
the pore diameter. The contributions of macropore
volumes (>50 nm) to the total volume for both materials
were negligible. Moreover, excluding the micropores, the
mesopores (2-50 nm) in the char were similar to those
of the commercial activated carbon. The pore volume
profile density reached a maximum value at a pore
diameter of 25 nm (rj 25 nm). Finally, although the
lower limit of the mercury intrusion test is 3.6 nm, the
data in Figure 6 show that the micropores in pyrolytic
char were greatly less developed than those in the
commercial activated carbon. Pyrolytic char would be
more useful for aqueous adsorption of large molecular

Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5143

Figure 6. Pore volume profiles for pyrolytic char and commercial

activated carbon measured using the mercury intrusion method.

Figure 7. Burnoff and BET surface area achieved in carbon

dioxide for the 2.5-5.1 mm fraction of a 550 C tire-derived char
in relation to the activation temperature.

weight species instead of small molecular weight species, because of the rich mesopores in char.
3.4.4. Activation Characteristics of Pyrolytic
Chars. Because the original pyrolytic char can only be
used as an adsorbent of large molecular species, activation of the char to produce a relatively high grade
activated carbon with both high BET surface area and
micropore volume may represent a more economically
attractive option. Figure 7 shows the burnoff degree and
BET surface area achieved in carbon dioxide using the
2.5-5.1 mm fraction of 550 C pyrolytic char in relation
to the activation temperature. As the activation temperature increased from 850 to 950 C, the burnoff
increased from 21.3% to 51.3% nearly in a linear
fashion. The BET surface area also linearly increased
from 125 to 306 m2/g. The impact of the activation
temperature can be attributed to the Arrhenius law, as
discussed in detail by Cunliffe and Williams.47
Figure 8 shows the influence of the burnoff degree
on the BET surface area of the activated carbon
produced by activation of 550 C tire-derived char. The
maximum BET surface area attained was 306 m2/g at
the 51.3% burnoff (950 C and 4 h), while that of the
original pyrolytic char was only 89.1 m2/g. High surface
areas for activated tire-derived char similar to these

Figure 8. Influence of burnoff on the BET surface area of

activated char produced from the activation of a 550 C tire-derived
char.

values have also been reported in the literature. For

example, Teng et al.8 reported that activation of tirederived char with carbon dioxide using TG-DTG resulted in a maximum surface area of 370 m2/g with 50%
burnoff. Merchant and Petrich9 demonstrated conversion of the tire-derived char (using a muffle furnace) to
activated carbon with a surface area of 500 m2/g at 850
C in a nitrogen flow containing 40 mol % steam.
Mirmiran et al.48 reported that the activation of 550 C
pyrolytic char by carbon dioxide under 900 C resulted
in a maximum surface area of 500 m2/g with 80%
burnoff. Cunliffe and Williams47 studied activation of
the fixed-bed pyrolytic char and reported that, for 50%
burnoff, carbon activated by steam had a surface area
of 510 m2/g, while carbon activated by carbon dioxide
had a surface area of only 420 m2/g. The activation of
the tire-derived char (using an ablative pyrolyzer) with
carbon dioxide by Helleur et al.34 obtained a surface area
of 240-270 m2/g at about 40% burnoff. The high ash
content of the pyrolytic char in the current work
partially limited the enlargement of the BET surface
area during activation. Therefore, acid demineralization
of the char after activation is needed to further improve
the surface area of the activated char.
In addition, the approximately linear relationship
between the surface area and the carbon burnoff in
Figure 8 has been observed in many previous studies.
Cunliffe and Williams47 found an initial relatively slow
increase in the BET surface area with increasing
burnoff, followed by a linear increase up to a maximum
surface area at 65% burnoff, after which the surface
area then decreased. Merchant and Petrich9 not only
illustrated the linear relationship between the BET
surface area and burnoff but also studied the variation
of the micropore volume with burnoff. They concluded
that the increasing BET surface area with burnoff below
40% can be attributed to the increasing micropore
volume, while the increase with burnoff in excess of 40%
was attributed to the increasing meso- and macropore
volumes. Furthermore, Teng et al.8 and Mirmiran et
al.48 also pointed out the linearity between the surface
area and carbon burnoff for various scale reactors.
Figure 9 compares the pore volume profiles for
activated tire-derived char (pyrolyzed at 550 C and

5144 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004

Figure 9. Pore volume profiles for activated tire-derived char

(pyrolyzed at 550 C and activated at 950 C) and commercial
activated carbon.

activated at 950 C) and commercial activated carbon.

As mentioned earlier, the pore volumes in the original
pyrolytic char were much different from those in the
commercial activated carbon by at least 1 order of magnitude. As shown in Figure 9, the pore volume profile
for the activated tire-derived char is quite similar to that
for the commercial activated carbon, especially for the
mesopore volume. However, the micropore volume in the
activated tire-derived char is still less than that in the
commercial activated carbon. This suggests that the
pyrolytic or activated tire-derived char should be used
as a mesoporous carbon in some special cases, e.g., the
aqueous adsorption of large molecular species.
Aqueous adsorption using tire-derived char has also
been investigated. The adsorption of large-molecule
methyl blue by the activated or original tire-derived char
was both about 230 mg/g3, which is a little higher than
that of the commercial activated carbon. However, the
adsorption of the iodine by tire-derived char was only
1/ that of the commercial activated carbon. Further
5
research on the adsorption of specific large-molecule
pollutants would be of interest. For example, the molecular sizes of mercury and dioxins, mainly produced
from waste incinerators, are 0.45 0.45 and 1.0 0.3
nm, respectively.49 The effective pore diameter for the
adsorption of dioxins is about 2-10 nm, which belongs
to the mesopore structure, while that for mercury is
0.5-5 nm, which belongs to the transition zone between
micropores and mesopores. Thus, the original inactivated tire-derived char theoretically has the potential
for the adsorption of dioxins, and slightly activated tirederived char may be used as an adsorbent for mercury.
Further work is needed to develop these applications.
4. Conclusions
(1) A rotary kiln is a suitable alternative for a
pyrolysis reactor of scrap tires because the solid residence
time can be flexibly adjusted to meet the requirement
of complete decomposition. Two crucial experiences
should be noted in the operation of a pilot-scale rotary
kiln pyrolyzer. First, the selection of the solid residence
time, measured with difficulty during pyrolysis, can be
optimized as follows: conducting similar pyrolysis tests
by increasing the rotating rate in a step of 0.5 rpm,

judging whether the difference between char yields in

the subsequent tests is less than 1%, and finally
determining the solid residence time of complete pyrolysis. Second, the kiln length with respect to diameter
plays a crucial role in determining the vapor residence
time. The latter directly determines the oil yield as well
as its valuable light compositions such as BTX. The
optimum time/temperature profiles achieved by the
multiple heating zones with different temperatures
along the kiln length should been attempted to promote
the production of volatile aromatics.
(2) As the temperature increased from 450 to 650 C,
the char yield remained essentially constant at a mean
of 39.8 wt %. Compared with other typical processes
such as vacuum, fluidized-bed, ablative, and two-stage,
the char yield is independent of reactor types with
different heating rates. The oil yield reached a maximum value of 45.1 wt % at 500 C and then decreased.
The oil yield is less than that of Kawakami et al. using
a similar rotary kiln.38 The long vapor residence time
in the high-temperature zone caused not only the low
yield but also richer single-ring aromatics than other
typical processes. The maximum concentrations of
benzene, toluene, xylene, styrene, and limonene in the
oil were 2.09, 7.24, 2.13, and 5.44 wt %, respectively.
The absorbance peaks at 3100-3000, 1604 (1495), and
900-675 cm-1 using FT-IR analysis confirmed the
formation of rich aromatics. The increasing temperature
resulted in the increment of the aromatics composition,
especially the high concentration of PAHs.
(3) The TBP distillation test of the tire-derived oil had
originally been conducted. It had a 39.2-42.3 wt %
naphtha fraction (ibp 200 C), a 32.4-33.2 wt %
medium fraction that can be used as diesel oil (200350 C), and a 25.5-28.5 wt % heavy fraction (>350 C).
The content of the naphtha fraction was greatly higher
than that by Kawakami et al. (24.5 wt %).38
(4) The pore structure of pyrolytic char was mostly
composed of mesopores instead of micropores as measured by both N2 adsorption and mercury intrusion
analyses. It is most suitable for aqueous adsorption of
large molecular weight species.
(5) Pyrolytic char after carbon dioxide activation had
a relatively high BET surface area of 306 m2/g at 51.3%
burnoff. There is an approximately linear relationship
between the surface area and carbon burnoff. Further
studies on the adsorption of dioxins and heavy metals
are currently underway.
Acknowledgment
This work was funded by National Natural Science
Fundation of China (Grant 50076037). We sincerely
thank Prof. Paul Williams, University of Leeds, for the
inspiration on the research method by their publications. The authors thank the the staff of Department
of Chemistry, Zhejiang University, for their help in the
analysis of oil properties. We also acknowledge the help
of Dr. R.-D. Li, Dr. J.-T. Huang, Dr. D.-H. Yan, Y.-L.
Gao, and S.-E. Wen for their cooperation in the experiments.
Literature Cited
(1) Cunliffe, A. M.; Walliams, P. T. Composition of Oils Derived
from the Batch Pyrolysis of Tyres. J. Anal. Appl. Pyrolysis 1998,
44, 131.
(2) De Marco Rodriguez, I.; Laresgoiti, M. F.; Cabrero, M. A.;
Torres, A.; Chmon, M. J.; Caballero, B. Pyrolysis of scrap tyres.
Fuel Process Technol. 2001, 72, 9.

Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5145
(3) Li, S.-Q. Pyrolysis of solid wastes for clean fuels and
chemicals: Lab- and pilot-scale studies. Ph.D. Dissertation, Zhejiang University, Hangzhou, China, 2002.
(4) Kaminsky, W.; Mennerich, D. Pyrolysis of synthetic tire
rubber in a fluidised-bed reactor to yield 1,3-butadiene, styrene
and carbon black. J. Anal. Appl. Pyrolysis 2001, 58-59, 803.
(5) Roy, C.; Chaala, A.; Darmstadt, H. The vacuum pyrolysis
of used tires End-uses for oil and carbon black products. J. Anal.
Appl. Pyrolysis 1999, 51, 201.
(6) Bilitewaki, B.; Hardtle, G.; Marek, K. Usage of carbon black
and activated carbon in relation to input and technical aspects of
the pyrolysis process. In Pyrolysis and gasification; Ferrero, G.
L., Maniatis, K., Buekens, A., Eds.; Elsevier Applied Science:
London, U.K., 1989.
(7) San Miguel, G.; Fowler, G. D.; Sollars, C. J. Pyrolysis of
tire rubber: porosity and adsorption characteristics of pyrolytic
chars. Ind. Eng. Chem. Res. 1998, 37, 2430.
(8) Teng, H.; Serio, M. A.; Wojtowicz, M. A.; Bassilakis, R.;
Solomon, P. R. Reprocessing of used tires into activated carbon
and other products. Ind. Eng. Chem. Res. 1995, 34, 3102.
(9) Merchant, A. A.; Petrich, M. A. Pyrolysis of scrap tires and
conversion of chars to activated carbon. AIChE J. 1993, 39, 1370.
(10) Collins, L. W.; Downs, W. R.; Gibson, L. K.; Moore, G. W.
An evalation of discarded tires as source of fuel. Thermochim. Acta
1974, 10, 153.
(11) De Marco Rodriguez, I.; Laresgoiti, M. F.; Cabrero, M. A.;
Torres, A.; Chmon, M. J.; Caballero, B. Pyrolysis of scrap tyres.
Fuel Process Technol. 2001, 72, 9.
(12) Williams, P. T.; Bottrill, R. P.; Cunliffe, A. M. Combustion
of tyre pyrolysis oil. Trans. Inst. Chem. Eng. 1998, 76B, 291.
(13) Williams, P. T.; Besler, S.; Taylor, D. T. The batch pyrolysis
of tyre wastesfuel properties of the derived oils and overall plant
economics. Proc. Inst. Mech. Eng. 1993, 207, 55.
(14) Benallal, B.; Roy, C.; Pakdel, H.; Chabot, S.; Poirier, M.
A. Characterisation of pyrolytic light naphtha from vacuum
pyrolysis of used tyres. Comparison with petroleum naphtha. Fuel
1995, 74, 1589.
(15) Pakdel, H.; Roy, C. Simultaneous gas chromatographicFourier transform infrared spectroscopic-mass spectrometric analysis of synthetic fuel derived from used tire pyrolysis oil, naphtha
fraction. J. Chromatogr. A 1994, 683, 203.
(16) Pakdel, H.; Pantea, D. M.; Roy, C. Production of dllimonene by vacuum pyrolysis of used tires. J. Anal. Appl.
Pyrolysis 2001, 57, 91.
(17) Pakdel, H.; Roy, C.; Aubin, H.; Jean, G.; Coulombe, S.
Formation of DL-limonene in used tire vacuum pyrolysis oils.
Environ. Sci. Technol. 1992, 25, 1646.
(18) Cypres, R.; Bettens, B. Production of Benzoles and Active
Carbon from Waste Rubber and Plastic Materials by Means of
Pyrolysis with Simultaneous Post-cracking. In Pyrolysis and
gasification; Ferrero, G. L., Maniatis, K., Buekens, A., Eds.;
Elsevier Applied Science: London, U.K., 1989.
(19) Kaminsky, W.; Shin, H. Pyrolysis of plastic waste and
scrap tyres using a fluidised bed process. In Thermal Conversion
of Solid Waste and Biomass; Jones, J. L., Radding, S. B., Eds.;
ACS Symposium Series 130; American Chemical Society: Washington, DC, 1980; p 423.
(20) Kaminsky, W.; Kim, J.-s. Pyrolysis of mixed platics into
aromatics. J. Anal. Appl. Pyrolysis 1999, 51, 127.
(21) Williams, P. T.; Taylor, D. T. Aromatisation of tyre
pyrolysis oil to yield polycylic aromatic hydrocarbons. Fuel 1993,
72, 1469.
(22) Williams, P. T.; Bottrill, R. P. Sulfur-polycylic aromatic
hydrocarbons in tyre pyrolysis. Fuel 1995, 74, 736.
(23) Williams, P. T.; Besler, S.; Taylor, D. T. The pyrolysis of
scrap automotive tyres: the influence of temperature and heating
rate on product combustion. Fuel 1990, 69, 1474.
(24) Williams, P. T.; Besler, S.; Taylor, D. T.; Bottrill, R. P.
Pyrolysis of automotive tyre waste. J. Inst. Energy 1995, 68, 11.
(25) Roy, C.; Labrecque, B.; de Caumia, B. Recycling of scrap
tyres to oil and carbon black by vacuum pyrolysis. Resour.,
Conserv. Recycl. 1990, 4, 203.
(26) Roy, C.; Unsworth, J. Pilot-scale plant demostration of used
tyres vacumm pyrolysis. In Pyrolysis and gasification; Ferrero, G.
L., Maniatis, K., Buekens, A., Eds.; Elsevier Applied Science:
London, U.K., 1989.
(27) Roy, C.; Rastegar, A.; Kaliaguine, S.; Darmstadt, H.
Physicochemical properties of carbon blacks from vacuum pyrolysis
of used tires. Plast. Rubber Compos. Process. Appl. 1995, 23, 21.

(28) Chaala, A.; Roy, C.; Ait-kadi, A. Rheological properties of

bitumen modified with pyrolytic carbon black. Fuel 1996, 75, 1575.
(29) Kaminsky, W. Chemical recycling of mixed plastics by
pyrolysis. Adv. Polym. Technol. 1995, 14, 337.
(30) Grittner, N.; Kaminsky, W.; Obst, G. Fluid bed pyrolysis
of anhydride hardened epoxy resins and polyester-polyurethane
by the hamburg process. J. Anal. Appl. Pyrolysis 1993, 25, 293.
(31) Black, J. W.; Brown, D. B. Preliminary mass balance
testing of the continuous ablation reactor. In Biomass Thermal
Processing, Proceedings of the First Canada/European Community
R&D Contractors Meeting; Hogan, E., Robert, J., Grassi, G.,
Bridgwater, A. V., Eds.; CPL Press: 1990; p 123.
(32) Brown, D. B. Continuous ablative regenerator systems. In
Bio-oil Production and Utilisation; Bridgwater, A. V., Hogan, E.
N., Eds.; CPL Press: 1996; p 96.
(33) Bridgwater, A. V.; Peacocke, G. V. C. Fast pyrolysis
processes for biomass. Renewable Sustainable Energy Rev. 2000,
4, 1.
(34) Helleur, R.; Popovic, N.; Ikura, M.; Stanciulescu, M.; Liu,
D. Characterization and potential applications of pyrolytic char
from ablative pyrolysis of used tires. J. Anal. Appl. Pyrolysis 2001,
58-59, 813.
(35) Dufton, P. W. The value and use of scrap tyres. Report
dated 1987; Rapra Technolgy Ltd.: England, p 11.
(36) Fortuna, F.; Cornacchia, G.; Mincarini, M.; Sharma, V. K.
Pilot-scale experimental pyrolysis plant: mechanical and operational aspects. J. Anal. Appl. Pyrolysis 1997, 40-41, 403.
(37) Klose, W.; Wiest, W. Experiments and mathematical
modeling of maize pyrolysis in a roatry kiln. Fuel 1999, 78, 65.
(38) Kawakami, S.; Inoue, K.; Tanaka, H. Pyrolysis process for
scrap tyres. In Thermal Conversion of Solid Waste and Biomass;
Jones, J. L., Radding, S. B., Eds.; ACS Symposium Series 130;
American Chemical Society: Washington, DC, 1980; p 557.
(39) Li, A. M.; Li, X. D.; Li, S. Q.; Yan, J. H.; Cen, K. F.
Experimental Studies on Municipal Solid Waste Pyrolysis in a
Laboratory-scale Rotary Kiln. Energy 1999, 3, 209.
(40) Li, A. M.; Li, X. D.; Li, S. Q.; Yan, J. H.; Cen, K. F. Pyrolysis
of Solid Waste in a Rotary Kiln: Influences of Final Pyrolysis
Temperature (FPT) on the Pyrolysis Products. J. Anal. Appl.
Pyrolysis 1999, 50, 149.
(41) Li, S.-Q.; Yan, J.-H.; Li, R.-D.; Chi, Y.; Cen, K.-F. Axial
Transport and Residence Time of MSW in Rotary Kilns: Part I
Experimental. Powder Technol. 2002, 126, 217.
(42) Li, S.-Q.; Chi, Y.; Li, R.-D.; Yan, J.-H.; Cen, K.-F. Axial
Transport and Residence Time of MSW in Rotary Kilns: Part II
Theoretical and Optimal Analyses. Powder Technol. 2002, 126,
228.
(43) Chi, Y.; Li, S.-Q.; Huang, J.-T.; Yan, J.-H.; Cen, K.-F. Clean
Fuels from Pyrolysis of Scrap Tyre in Rotary Kiln: Experimental
Studies and Mechanical Design. The 3rd International Symposium
on Incineration and Flue Gas Treatment Technologies, Incineration Division, Brussels, Belgium, July 2-4, 2001; Section 7.
(44) Yang, J.; Kaliaguine, S.; Roy, C. Improved quantative
determination of elastomers in tyre rubber by kinetic simualation
of DTG Curves. Rubber Chem. Technol. 1993, 66, 213.
(45) Leung, D. Y. C.; Wang, C. L. Kinetic modeling of scrap
tyre pyrolysis. Energy Fuels 1999, 13, 421.
(46) Darmstadt, H.; Roy, C.; Kaliaguine, S. Characterization
of pyrolytic carbon-blacks from commercial tire pyrolysis plants.
Carbon 1995, 33, 1449.
(47) Cunliffe, A. M.; Walliams, P. T. Influence of process
condition on the rate of activation of chars derived from pyrolysis
of used tires. Energy Fuels 1999, 13, 166.
(48) Mirmiran, S.; Pakdel, H.; Roy, C. Nitrogen-polycyclic
aromatic hydrocarbons in tyre pyrolysis oil. J. Anal. Appl. Pyrolysis
1992, 22, 205
(49) Hasselriis, F.; Licata, A. Analysis of heavy metal emission
data from municipal waste combustion. J. Hazard. Mater. 1996,
47, 77.

Resubmitted for review January 14, 2004

Accepted April 30, 2004
IE030115M