WHEC 16 / 13-16 June 2006 – Lyon France

Reaction Mechanism of Tar Evolution in

Biomass Steam Gasification for Hydrogen Production
Shingo Katayamaa, Masahiro Suzukia, and Atsushi Tsutsumia
a
Department of Chemical System Engineering, The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo,
113-8656, Japan, tsutsumi@chemsys.t.u-tokyo.ac.jp (A. Tsutsumi)

ABSTRACT:
Reaction mechanism of tar evolution in steam gasification of biomass was investigated with a continuous
cross-flow moving bed type differential reactor, in which tar and gases can be fractionated according to
reaction time. We estimated that time profile of tar and gas evolution in the gasification of cellulose, xylan,
and lignin, and compared it with experimental product time profile of real biomass gasification. The
experimental tar evolution rate is different from estimated tar evolution rate. The estimated tar evolution rate
has a peak at 20 s. On the other hand, the experimental tar evolution rate at 20 s is little, and tar at initial
stage includes more water-soluble and water-insoluble compounds. It can be concluded that in the real
biomass steam gasification the evolution of tar from cellulose and lignin component was found to be
precipitated by that from hemicellulose component.
KEYWORDS: Biomass gasification, Tar, Continuous Cross-flow moving bed type Differential Reactor

Introduction
The steam gasification of biomass is a promising
technology for thermochemical hydrogen production Volatile tar
kv
from biomass. Since biomass is a renewable and clean ki
energy resource, biomass hydrogen is essential for the Cellulose Active cellulose
sustainable hydrogen society. In addition, among kc
biomass conversion processes, gasification is one of the
Char + Gases
most efficient ones [1]. On the other hand, biomass
gasification has problems such as tar production. Tar is Fig. 1 Broido-Shafizadeh model
a complex mixture of condensable hydrocarbons such
as aromatics. In biomass gasification process, a large
amount of tar is produced, causing some trouble such
as pipeline plugging, defluidization, and the reduction of K1 K2 K3
thermal efficiency [2]. Thus, it is essential to elucidate Virgin Active Tar Gas
the reaction mechanism of tar production in gasification Cellulose K4
for improvement in controllability and thermal efficiency Lignin
of biomass gasifiers. or
x Char + (1-x) Gas
So far, many researchers have conducted the Hemicellulose
experimental studies about pyrolysis of biomass and its
Fig. 2 Miller-Bellan model
components [3-30]. Especially, cellulose has most been
studied. Broido- Shafizadeh model (Fig. 1) was the
initially proposed mechanism for the pyrolysis of
cellulose [3-5]. This model consists of three coupled first order reactions. The first stage of pyrolysis involves
no weight loss and leads to form intermediate called active cellulose. This is followed by a pair of competing
reactions which produces either volatiles or char and gases with weight loss. This concept of active cellulose
was inherited in many pyrolysis or gasification models of cellulose proposed later [6-11, 13,14]. Moreover,
extensive efforts for intermediate and tar evolution had been reported [11-14, 16-20]. Boutin et al. [17-20]
observed that the active cellulose is liquid at the reaction temperature but solid at ambient temperature. The
active cellulose is completely soluble in water (anhydro-oligosaccharides) and its calculated mean lifetime is
0.02 s at the reaction temperature [18]. Wooten et al. [16] found that final carbohydrate converts to aromatic
carbons but not to aliphatic carbons when subjected to prolonged heating.
With regard to the pyrolysis of real biomass, many researchers has been studied and proposed reaction
models [23-27]. Most of those assume that cellulose, hemicellulose, and lignin are pyrolyzed independently.
Müller-Hagedorn et al. [23] proposed a parallel reaction model of 5 components (cellulose 1, cellulose 2,
hemicellulose 1, hemicellulose 2, and lignin) and evaluated formal kinetic parameters. Miller and Bellan [24,
25] considered that kinetic scheme of cellulose pyrolysis is also adopted to hemicellulose and lignin (Fig. 2).
They mentioned that pyrolysis of biomass can be explained based on a superposition of the three
components.

1/8
 WHEC 16 / 13-16 June 2006 – Lyon France

However, in the conventional reactors for kinetic study of biomass gasification such as a thermobalance
reactor, a drop tube reactor, a fixed bed reactor, and a fluidized bed reactor, it is impossible to investigate the
time profile of tar and gas evolution during biomass gasification with differential method of kinetic analysis in
continuous feeding condition. Thus, we have developed a continuous cross-flow moving bed type differential
reactor (CCDR), in which biomass sample is continuously fed, and the products (tar, gas and char) can be
fractionated from each compartment according to the reaction time. By using this apparatus, we previously
researched the mechanism of pyrolysis of cellulose, and found that the dehydration of nascent char takes
place after the devolatilization of tar is completed [8].
In the present study, the time profile of tar and gas evolution was investigated in steam gasification of
biomass and its components: cellulose, xylan as hemicellulose, and lignin. The reaction temperature was
kept at 673 K. The fractionated volatile matter and char were separated immediately and quenched. In this
experimental condition, no secondary reaction between volatile matter and char takes place because of low
temperature and short residence time. Based on the experimental results, we discuss on the reaction
mechanism of tar evolution in biomass steam gasification.

Experimental
Apparatus
Figure 3 shows the schematic diagram of a continuous cross-flow moving bed type differential reactor
(CCDR). The reactor consists of a quartz glass half tube covered with a quartz glass plate and a belt
conveyor system. The reactor is divided into six compartments (W90 mm×D80 mm×H40 mm) where gas
flows are independent. The reactor is heated using an infrared gold image furnace (Ulvac Riko, Co. Ltd.).
The heating zone of the furnace is divided into a preheat zone and three reaction zones. The temperatures
of each zone are measured by K-type thermocouples and are controlled to be constant at 673 K
independently. Biomass sample is fed out of a feeder onto the conveyor belt made of siliglass (SiO2 > 96%),
which carries the sample across the six compartments, delivering produced char to a char sampling system.
The initial time is defined as the time when sample is fed into the preheat zone. Tar and gases produced in
each compartment are sampled with a carrier gas and fractionated according to reaction time. At the end of
belt conveyor, char is collected using the char sampling system. By changing the belt speed, the residence
time of each compartment can be varied in the range of 2.4 - 32 s.
A bowl feeder (model PEF-90AL; Sanki Co.) included in an acrylic feeder box was used to feed biomass

Fig. 3 Experimental apparutus : continuous cross-flow

2/8 moving bed type differential reactor
 WHEC 16 / 13-16 June 2006 – Lyon France

sample continuously with an Ar carrier gas into the preheat zone through a SUS316 tube (6.35 mm, o.d.).
The feed rate was calibrated by weighing fed sample within 1 min before and after experiment, and calculate
the average feed rate using both weight of fed sample. Steam was generated in a steam generator and was
fed into the reactor with an Ar carrier gas. Tar evolved in each compartment went through heated sampling
lines, and then was collected separately in six tar traps that were cooled by ice. Water was also eliminated
in CaCl2 columns. Although the six sampling lines were gathered in a gas sampler (EMT 4SC6MWE; Valco
Instruments Co.), only one of the lines was connected to a micro gas chromatograph (M-200H; Hewlett
Packard Co.), and the line to the micro GC was changeable. The flow rate of the effluent gas was measured
with a mass flow meter. By using the micro GC we analyzed H2, O2, N2, CH4 and CO with an MS-5A column
and Ar carrier gas; CO2, C2H4 and C2H6 with a Pora Plot Q column and Ar carrier gas.

Procedure
In this study, these are used as samples: Table 1 Elemental compositions of samples (wt %)
Avicel microcrystalline cellulose (Merck Co. Ltd., < C H N O S
160 μm, Avicel), xylan from birch wood (Sigma-
Cellulose 43.5 6.68 0.14 48.91 0.77
Aldrich Co.), kraft lignin (Kanto Chemical Co. Ltd.,
Lignin 53.2 4.6 0.12 22.2 2.4
6.8-160 μm) and real biomass (Chilean Xylan 40.16 6.12 0.14 52.96 0.62
eucalyptus). The elemental compositions of these Real Biomass 48.87 6.05 0.19 44.17 0.72
samples are shown in Table 1. The sample was
dried at 333 K in an oven for 12 h prior to the experiment. The sample was put in the feeder and the feeder
box was purged with Ar. The reactor was also purged with Ar until the concentration of O2 declined below
1% and the flow rate of the carrier gas at each compartment was set to 400 cm3 min-1. The steam generator,
steam feeding lines and sampling lines were heated at 523, 453 and 423 K, respectively, to avoid the
condensation of steam and tar. After heating the reactor up to 673 K, steam (40 vol. %) and cellulose were
fed for 60 min. The calibration of the sample feed rate from the feeder was carried out at room temperature.
The sample feed rate was kept at 30-119 mg min-1 in this experiment. The belt speed was 750 mm min-1
and the reaction time of biomass sample was 38.4 s. Thus, the corresponding residence time of biomass
sample in each compartment was 6.4 s.

Analysis of tar
After the experiment, distilled water was introduced into the reactor and sampling lines. Deposited tar was
collected in the tar traps until the color of the solution became transparent. Acetone was also used as a
solvent. Molecular weight of tar was also measured by a gel filtration chromatograph (pump: Waters 600E
Multi Solvent Delivery System, column: Shodex AsahipakGS-220HQ, detector: Waters 2414 differential
refractometer) while KCl was added to the solution in order to reduce the intramolecular ionic repulsion
between hydroxyl groups. After evaporating solvent in the solution, the composition of dried tar was
analyzed with a CHNS elemental analyzer (2400II CHNS/O; Perkin Elmer). A small amount of the dried tar
was milled with KBr (0.1 g) and measured by FT-IR (FTIR-230; JASCO). The 4000-400 or 4600-400 cm-1
wave number rage with 16 scans was investigated.
In each compartment of CCDR the evolution rates of gas
and tar were defined as the follows: 100
Mc
Carbon Conversion Rate [%]

evolution rate (carbon molar basis ) =

M ⋅τ 80

MG
evolution rate ( weight basis ) = 60
W ⋅τ
where M c , M , τ , M G , and W represent a molar 40
amount of carbon in product in one compartment [mol], total
molar carbon in fed biomass [mol], residence time in one
compartment [s], molar amount of gases in product in one 20
compartment [mol], and total amount of fed biomass [g],
respectively. 0
lignin

superposition
cellulose

xylan

biomass

Results and discussion

Trend of gas and tar evolution in gasification of each
sample
Figure 4 shows the conversion rate to the gas, tar and
char in steam gasification of cellulose, lignin, xylan and real uncollected tar
biomass. The estimated biomass carbon conversion as a gas char
superposition of cellulose (50 %), lignin (27 %), and xylan Fig. 4 Carbon conversion rate of each sample
(23 %) is also shown in Fig. 4. This ratio of three main

3/8
 WHEC 16 / 13-16 June 2006 – Lyon France

components of biomass is estimated with elemental analysis of biomass by the measure reported by
Hasegawa et al. [31]. Figures 5-8 show the time profile of evolution rate of CO, CO2 (weight basis) and tar
(carbon molar basis) during steam gasification of these samples at 673 K for cellulose. Figure 9 shows the
evolution rate estimated as a superposition of cellulose, lignin, and xylan. In these figures, abscissa
represents the elapsed time after the sample is fed onto the quartz belt. Since the gas evolution rate of H2,
CH4, C2H4, and C2H6 was smaller than 2% of that of CO2 in any case, these gases are not represented in the
figures.
In case of cellulose (Fig. 4), most of volatile was observed to be water soluble tar. It cam be seen in Fig. 5
that the evolution of tar increased rapidly with reaction time to achieve a peak at 22.4 s. In addition, the
evolution of gas was found to exhibit a peak at 28.8 s. These results are attributed to the existence of
intermediates [8].
0.30 0.012
In case of lignin, about 70 % of sample

[mol-C in tar / mol-C in sample s-1]

tar
converted to char, and the amount of tar was 0.25 0.010
very small as shown in Fig. 4. It was see from

[mmol g-sample-1 s-1]

Gas Evolution Rate

Tar Evolution Rate

Fig. 6 that the initial stage exhibits large 0.20 0.008
evolution rates of tar and gas. With increasing
time the evolution rates of tar and gas 0.15 0.006
decreased. Most of tar was water insoluble. This
was attributed to the composition of lignin which 0.10 0.004
mainly consists of hydrophobic aromatic CO
components. 0.05 CO2 0.002
For xylan, the amount of gas was observed to
be larger than that of cellulose and lignin (Fig. 4). 0.00 0.000
0 10 20 30 40
In Fig. 7 the time profile of gas evolution for
Time [s]
xylan gasification was found to exhibit two
peaks at the initial stage and 28.8 s. This result
corresponds to the report by Müller-Hagedorn et Fig. 5 Time profile of gas and tar evolution (cellulose)
0.30 0.012 0.30 0.012
[mol-C in tar / mol-C in sample s-1]

[mol-C in tar / mol-C in sample s-1]

0.25 0.010 0.25 0.010
[mmol g-sample-1 s-1]

[mmol g-sample-1 s-1]

Gas Evolution Rate

Gas Evolution Rate

Tar Evolution Rate

Tar Evolution Rate

0.20 0.008 0.20 CO2 0.008

0.15 0.006 0.15 0.006

0.10 CO 0.004 0.10 CO 0.004

2
CO tar
0.05 tar 0.002 0.05 0.002

0.00 0.000 0.00 0.000

0 10 20 30 40 0 10 20 30 40
Time [s]
Time [s]

Fig. 6 Time profile of gas and tar evolution (lignin) Fig. 7 Time profile of gas and tar evolution (xylan)

0.30 0.012 0.30 0.012

[mol-C in tar / mol-C in sample s-1]
[mol-C in tar / mol-C in sample s-1]

0.25 0.010 0.25 0.010

[mmol g-sample-1 s-1]
[mmol g-sample-1 s-1]

Gas Evolution Rate

Gas Evolution Rate

Tar Evolution Rate

Tar Evolution Rate

0.20 0.008 0.20 0.008

CO
0.15 0.006 0.15 2 0.006
tar tar
CO2
0.10 0.004 0.10 0.004

0.05 0.002 0.05 CO 0.002

CO
0.00 0.000 0.00 0.000
0 10 20 30 40 0 10 20 30 40

Time [s] Time [s]

Fig. 8 Time profile of gas and tar evolution Fig. 9 Time profile of gas and tar evolution
(biomass : experimental result) (biomass : estimated as superposition of cellulose
(50 %), lignin (27 %) , and xylan (23 %))

4/8
 WHEC 16 / 13-16 June 2006 – Lyon France

0.05 0.05

The Amount of Tar [g / g-sample]

The Amount of Tar [g / g-sample] Water Insoluble Water Insoluble
0.04 Water Soluble 0.04 Water Soluble

0.03 0.03

0.02 0.02

0.01 0.01

0 0
3.2 9.6 16.0 22.4 28.8 35.2 3.2 9.6 16.0 22.4 28.8 35.2
Time [s] Time [s]
(a) (b)
Fig. 10 Amount of tar in biomass gasification (a) experiment (b) estimated as the superposition of
cellulose, lignin, and xylan

al. [23] which shows hemicellulose has two

decomposition steps. The dominant tar evolution Tar from cellulose
takes place at initial stage. The amount of tar in xylan
gasification was smaller than that in cellulose
gasification and larger than that in lignin gasification.
In case of biomass (Fig. 8), a similar trend for tar
and gas evolution was observed with xylan
gasification. The tar and gas evolution was found to
exhibit two peaks at the initial stage and 28.8-35.2 s. Tar from xylan
As can be seen in Fig. 9, the estimated time profile of
gas evolution was similar to that of experimental data.
The gas evolution exhibits two peaks at the initial
Transmittance [a. u.]

stage and 28.8 s.

In spite of similar trend for gas evolution, tar
evolution was found to exhibit a different manor. In Tar from biomass (3.2 s)
the initial stage of the large real biomass gasification,
tar evolution rate was large and reduced with
increasing time to a minimum at 22.4 s. On the other
hand, tar evolution was found to be estimated to
increase with time to reach a maximum at 22.4 s.
Figures 10 (a) and 10 (b) show the evolution time
profiles of water-insoluble and water soluble tar for Tar from biomass (28.8 s)
the experimental result and the estimation by means
of the superposition of cellulose, lignin and xylan,
respectively. In the experimental result, both water-
soluble and water-insoluble tar evolved at initial
stage, and then the ratio of water-insoluble tar
decreased with reaction time. On the other hand, in
the estimation of tar evolution most of tar was water-
soluble because of small amount of tar from lignin 4000 3000 2000 1000 400
and xylan, and most of tar evolved from cellulose Wave number [cm-1]
was water soluble. These results suggest that, in
biomass gasification, the amount of water-insoluble Fig. 11 FT-IR spectra of produced water-soluble
tar which mainly evolves from lignin increases and tar from cellulose, xylan, and biomass (reaction
the evolution of water soluble tar which evolves from time : 3.2 s and 28.8 s)
cellulose occurs earlier than in the cellulose
gasification.
Analysis of tar
Figure 11 shows FT-IR spectra of water-soluble tar from cellulose, xylan, and biomass (3.2 s and 28.8 s).
No significant change in FT-IR spectra of tar evolved from cellulose and xylan was observed with reaction
time. Tar from cellulose has some peaks at 950 – 800 cm-1. These are considered to be assigned to α1-6
bond of levoglucosan, the main component of water-soluble tar from cellulose. Tar from xylan has a strong

5/8
 WHEC 16 / 13-16 June 2006 – Lyon France

peak at about 1400 cm-1. The tar from biomass

gasification at 3.2 s after the beginning of reaction Lignin (sample)
also has a comparatively strong peak at 1400 cm-1.
Tar derived from biomass at 28.8 s has some
distinct peaks at 950 – 800 cm-1. It is concluded

Transmittance [a. u.]

that at initial stage in biomass gasification tar is
mainly from xylan and at about 30 s later, tar
derived from cellulose increases.
The FT-IR spectra of lignin and water-insoluble Tar from biomass (3.2 s)
tar from biomass (3.2 s and 28.8 s) were shown in
Fig. 12. The spectrum of lignin shows peaks
assigned to aromatic ring at 1600, 1510, 1420 cm-1.
Tar from biomass at 3.2 s also has these peaks.
Figure 13 shows the molecular weight
distribution of water-insoluble tar from biomass Tar from biomass (28.8 s)
(reaction time: 3.2 s and 28.8 s) measured by the
gel filtration chromatography (GFC). Hasegasa et
al. reported that the molecular weight distribution of
tar derived from cellulose has peaks at about 150
and 300, and the MW of tar from lignin is
distributed at about 1000 - 4000. The MW
distribution of tar at 3.2 s of reaction time has a
broad peak at 1000 – 3000. This suggests that the 4000 3000 2000 1000 400
water-insoluble tar at initial stage of biomass
gasification contains the component derived from
Wave number [cm ] -1

lignin.
Fig. 12 FT-IR spectra of lignin and produced
Scheme of biomass gasification
From the analyses of tar, the evolution water-insoluble tar from biomass (reaction time :
mechanism of tar and gas in biomass gasification is 3.2 s and 28.8 s)
summarized schematically in Fig. 14. Horizontal
axis represents elapsed time from the beginning of MW
gasification, and vertical axis shows the amount of 1000 - 3000
produced gas and tar.
In cellulose gasification, tar and gas evolution
rate increases with reaction time. The evolutions of
tar and gas achieve a peak at about 20 s and 30 s, 3.2 s
respectively. On the other hand, in biomass
gasification, evolution of water-soluble tar showed
peak early in the reaction. Thus, it can be
considered that cellulose component of biomass
releases tar earlier than pure cellulose. 28.8 s
Pure lignin releases gas at initial stage of
gasification. With little tar evolution, most of lignin
converted into char. In biomass gasification,
however, at initial stage water-insoluble tar was
produced much more than expected from the Large Small
gasification of lignin alone. As this tar contained MW Retention time MW
aromatic components derived from lignin, it is
considered that the evolution of tar from lignin Fig. 13 Molecular Weight distribution of water-
component of biomass was enhanced comparing insoluble tar measured by GFC
with pure lignin gasification.
In case of xylan, the representative of hemicellulose, the evolution of gas shows two peaks at initial stage
and 30 s later. Tar evolution occurs at the beginning of the reaction, and the amount is more than lignin but
much less than cellulose.
As mentioned above, tar evolution of biomass gasification cannot be estimated by the superposition of
cellulose, lignin, and hemicellulose. The increase in tar derived from lignin component in real biomass
gasification is likely to be due to the effect of reaction of hemicellulose. Omori et al. [32] indicated that
holocelluose (hemicellulose and α-hemilose) acts as hydrogen donor. Sagehashi et al. [33] reported that in
biomass (Japanese cedar) gasification production of phenol and guaiacol surpassed the amounts predicted
from the result of cellulose, xylan, and lignin. They considered that holocelluose which acts as hydrogen
donor accelerates the pyrolysis of biomass. On the other hand, Müller-Hagedorn et al. [23] reported that
inorganic salts promote pyrolysis reaction. In addition, some researchers have reported the catalytic effect of

6/8
 WHEC 16 / 13-16 June 2006 – Lyon France

alkali and alkaline earth metallic (AAEM)

species for biomass gasification [34, 35]. Thus, Gas
it is also likely that the inorganic salts like Gas
AAEM influences the tar evolution. Further
Tar
investigation is needed to reveal the factor
which influences the evolution of tar in biomass Tar
gasification. (water soluble)
(in biomass, a part of tar is released earlier)
Conclusions
By using a continuous cross-flow moving bed Cellulose
type differential reactor (CCDR), which can
separate the volatiles according to reaction
time, steam gasification of cellulose, lignin, Gas
xylan, and real biomass was investigated at Gas
673 K. The fractionated tar was measured by
weight and analyzed by FT-IR spectrometry, Tar Tar
gel filtration chromatography, and elemental (water insoluble, including aromatics)
analysis. It is proven that in real biomass (in biomass, the amount of tar increases)
gasification the evolution rate of gas is similar
to that of the predicted as superposition of Lignin
cellulose, lignin, and xylan. On the other hand,
the experimental tar evolution rate of real
biomass is different from estimated tar Gas
evolution rate from each gasification result of
the three components. The estimated tar
Gas
evolution rate has a peak at 20 s, but
experimental tar evolution rate has two peaks Tar
at the initial stage and 30 s. Tar derived from Tar
real biomass at initial stage includes more (water soluble and insoluble, not including aromatics)
water-soluble and water-insoluble compounds. Hemicellulose
From the analyses of tar, it is indicated that
the composition of tar varies with reaction time. Gas
Tar at initial stage of gasification of real
Gas
biomass includes both water-soluble tar
derived from cellulose and hemicellulose, and Tar
water-insoluble tar from lignin. In addition, the
main component of tar at 30 s of reaction time Tar from cellulose
is water-soluble tar from cellulose. Tar from hemicellulose
Tar from lignin
Acknowledgement Biomass
This study was financially supported by a
Fig. 14 Schematic diagrams for evolution of tar
“Core Research for Evolutional science and
gas and tar in biomass gasification at 673 K
Technology (CREST)” grant from the Japan
Science and Technology Agency (JST) and New Energy and Industrial Technology Development
Organization (NEDO). The authors appreciate this financial support.

References:
1. Bridgewater, A. V. The technical and economic feasibility of biomass gasification for power generation.
Fuel, 74, 631-653, 1995.
2. Devi, L.; Ptasinski, K. J.; Janssen, F. J. J. G. A review of the primary measures for tar elimination in
biomass gasification processes. Biomass and Bioenergy, 24, 125-140, 2003.
3. Shafizadeh, F; Lai, Y.Z. Thermal Degradation of 1,6-Anhydro-p-β-glucopyranose, J. Org. Chem. 37, 278-
284, 1972.
4. Bradbury, A. G. W.; Sakai, Y.; Shafizadeh, F. A kinetic model for pyrolysis of cellulose. J. Applied
Polymer Science, 23, 3271-3280, 1979.
5. Shafizadeh, F. J. Introduction to pyrolysis of biomass. J. Anal. Appl. Pyrolysis, 3, 283-305, 1982.
6. Fushimi, C.; Araki, K.; Yamaguchi, Y.; Tsutsumi, A. Effect of Heating Rate on Steam Gasification of
Biomass. 1. Reactivity of Char. Ind. Eng. Chem. Res., 42, 3922-3928, 2003.
7. Fushimi, C.; Araki, K.; Yamaguchi, Y.; Tsutsumi, A. Effect of Heating Rate on Steam Gasification of
Biomass. 2. Thermogravimetric-Mass Spectrometric (TG-MS) Analysis of Gas Evolution. Ind. Eng. Chem.
Res., 42, 3929-3936, 2003.

7/8
 WHEC 16 / 13-16 June 2006 – Lyon France

8. Yamaguchi, Y.; Fushimi, C.; Tasaka, K.; Furusawa, T.; Tsutsumi, A. Kinetic Study on Pyrolysis of
Cellulose Using Novel Continuous Cross-Flow Moving Bed Type Differential Reactor. Energy and Fuels,
submitted.
9. Mok, W. S. –L.; Antal, M. J. Effect of Pressure on Biomass Pyrolysis. 2. Heat of Reaction of Cellulose
Pyrolysis. Thermochimica Acta. 68, 165-186, 1983.
10. Cooley, S.; Antal M. J. Kinetic of Cellulose Pyrolysis in the Presence of Nitric-oxide. J. Anal. Appl.
Pyrolysis. 14, 149-161, 1988.
11. Banyasz, J. L.; Li, S.; Lyons-Hart, J.; Shafer, K. H. Gas evolution and the mechanism of cellulose
pyrolysis. Fuel. 80, 1757-1763, 2001.
12. Li, S.; Lyons-Hart, J.; Banyasz, J.; Shafer, K. H. Real-time evolved gas analysis by FTIR method: an
experimental study of cellulose pyrolysis. Fuel. 80, 1809-1817, 2001.
13. Banyasz, J. L.; Li, S.; Lyons-Hart, J.; Shafer, K. H. Cellulose pyrolysis: the kinetics of
hydroxyacetaldehyde evolution. J. Anal. Appl. Pyrolysis. 26, 159-174, 1993.
14. Antal, M. J.; Friedman, H. L.; Rogers, F. E. Kinetics of cellulose pyrolysis in nitrogen and steam,
Combust. Sci. and Tech. 21, 141-152, 1980.
15. Várhegyi, G, Szabó, P. Mok, W. S. -L.; Antal, M, J. Kinetics of the thermal decomposition of cellulose in
sealed vessels at elevated pressures. Effect of the presence of water on the reaction mechanism. J.
Anal. Appl. Pyrolysis, 26, 159-174, 1993.
16. Wooten, J.B. ; Seeman, J. I. ; Hajaligol, M. R. Observation and characterization of cellulose pyrolysis
intermediates by C-13 CPMAS NMR. A new mechanistic model. Energy & Fuels. 18, 1-15, 2004.
17. Boutin, O.; Ferrer, M.; Lédé, J. Radiant flash pyrolysis of cellulose - Evidence for the formation of short
life time intermediate liquid species. J. Anal. Appl. Pyrolysis. 47, 13-31, 1998.
18. Lédé, J.; Blanchard, F.; Boutin, O. Radiant flash pyrolysis of cellulose pellets: products and mechanisms
involved in transient and steady state conditions. Fuel, 2002, 81, 1269-1279
19. Lédé, J. The Cyclone: A Multifunctional Reactor for the Fast Pyrolysis of Biomass Ind. Eng. Chem. Res.
39, 893-903, 2002
20. Boutin, O.; Ferrer, M.; Lédé, J. Flash pyrolysis of cellulose pellets submitted to a concentrated radiation:
experiments and modelling, Chem. Eng. Sci. 57. 15-25. 2002.
21. Morf, P.; Hasler, P.; Nussbaumer, T. Mechanisms and kinetics of homogeneous secondary reactions of
tar from continuous pyrolysis of wood chips. Fuel, 81, 843-853, 2002.
22. Evans, R. J.; Milne, T. A. Molecular Characterization of the Pyrolysis of Biomass. 1. Fundamentals.
Energy & Fuels, 1, 123-137, 1987.
23. Müller-Hagedorn, M.; Bockhorn, H.; Krebs, L.; Müller, U. A comparative kinetics study on the pyrolysis of
three different wood species. J. Anal. Appl. Pyrolysis. 68-69, 231-249, 2003.
24. Miller, R. S.; Bellan, J. A Generalized Biomass Pyrolysis Model Based on Superimposed Cellulose,
Hemicellulose and Lignin Kinetics. Combust. Sci. and Tech. 126, 97-137, 1997.
25. Miller, R. S.; Bellan, J. Tar Yield and Collection from the Pyrolysis of Large Biomass Particles. Combust.
Sci. and Tech. 127, 97-118, 1997.
26. Di Blasi, C.; M. Lanzetta. Intrinsic kinetics of isothermal xylan degradation in inert atmosphere. J. Anal.
Appl. Pyrolysis. 40-41, 287-303, 1997.
27. Di Blasi, C. Comparison of semi-global mechanisms for primary pyrolysis of lignocellulosic fuels. J. Anal.
Appl. Pyrolysis. 47, 43-64, 1998.
28. Manyà, J. J.; Velo, E.; Puigjanaer, L. Kinetics of Biomass Pyrolysis: a Reformulated Three-Parallel-
Reactions Model. Ind. Eng. Chem. Res., 42, 434-441, 2003.
29. Várhegyi, G.; Antal, M. J.; Jakab, E.; Szabó, P. Kinetic modeling of biomass pyrolysis. J. Anal. Appl.
Pyrolysis, 42, 73-87, 1997.
30. Svenson, J.; Pettersson, J. B. C. First Pyrolysis of the Main Components of Birch Wood. Combust. Sci.
and Tech. 176, 997-990, 2004.
31. Hasegawa, I.; Fujisawa, H.; Sunagawa, K.; Mae, K. Quantitative Prediction of Yield and Elemental
Composition during Pyrolysis of Wood Biomass. J. Jpn. Inst. Energy (in Japanese), 84, 46-52, 2005.
32. Omori, S.; Aoyama, M.; Sakakibara, A. Hydrolysis of Lignin with Dioxane-Water XIX. Reaction of β-O-4
Lignin Model Compounds in the Presence of Carbohydrates. Holzforschung. 52, 391-397, 1998.
33. Sagehashi, M.; Miyasaka, N.; Shishido, H.; Sakoda, A. Superheated steam pyrolysis of biomass
elemental components and Sugi (Japanese cedar) for fuels and chemicals. Bioresource Technology. 97,
1272-1283, 2006.
34. Sutton, D.; Kelleher, B.; Ross, J. R. H. Review of literature on catalysts for biomass gasification. Fuel
Processing Technology, 73, 155-173, 2001.
35. Sathe, C.; Hayashi, J.; Li, C. Z.; Chiba, T. Combined effects of pressure and ion-exchangeable metallic
species. Fuel, 82, 343-350, 2003.

8/8