The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
Pyrolysis of Lignin in a Laboratory Fluidized Bed
Reactor
Isak Lindén1*, Atte Aho2, Nikolai DeMartini1, Anders Brink1, Dmitry Murzin2, Mikko Hupa1, Jyri-Pekka
Mikkola2,3
1
2
Laboratory of Inorganic Chemistry, Process Chemistry Centre, Åbo Akademi University
Biskopsgatan 8
FI-20500 Turku/Åbo
Finland
isak.linden@abo.fi
Telephone: +35822154933
Fax: +35822154962
Laboratory of Industrial Chemistry and Reaction Engineering, Process Chemistry Centre, Åbo Akademi
University
Biskopsgatan 8
FI-20500 Turku/Åbo
Finland
3
Technical Chemistry, Department of Chemistry, Chemical-Biological Center, Umeå Universitet
SE-901 87 Umeå
Sweden
* corresponding author
ABSTRACT
In this paper lignin pyrolysis was performed at temperatures between 400 and 550 °C
with analysis of the bio-oil and determination of yields of oil, char and gases. For
comparison, woody biomass was also pyrolysed at 450°C. A tubular fluidized bed reactor
purged with N2-gas was used for pyrolysis, while the pyrolysis gases were cooled in
several steps to -20º, and the remaining aerosols were removed with a particulate filter.
Compared to pyrolysis of biomass, lignin gave a high char yield, around 50%, while the
oil yield was much lower, around 25%. The pyrolysis behavior of lignin posed special
problems, as lignin is prone to agglomeration which can cause loss of fluidization. The
decomposition of lignin was also studied with a thermogravimetric analyser. At 500 °C
about 50 % of the lignin remained as char, while 37% remained at 988 °C.
Keywords: Pyrolysis, lignin
1. INTRODUCTION
Lignin is a major constituent of biomass. It is a cross-linked, highly branched threedimensional polymer with no exact structure. Lignin is synthesized in plants from three
-1-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
monomeric units, p-coumaryl, coniferyl, and sinapyl alcohol [1]. Fast pyrolysis of
biomass for the production bio-oil has attracted ever more interest in recent years, and is
covered for example by Mohan et al. in a review article [2]. Several literature review
articles on pyrolysis of lignin have also been published during the recent years [3-5].
Fast pyrolysis of lignin has recently been studied in an international collaboration
between 14 laboratories in eight different countries. The results showed that pyrolysis of
lignin produces a heavy tarry bio-oil in lower quantities than in pyrolysis of
lignocellulosic biomass, while the char yield is much higher, around 50%. Lignin was
also prone to cause plugging and agglomeration problems [6]. Sharma et al. reported in a
study on pyrolysis of lignin that the char yield was 62% at 400°C, and decreased to 40%
at 750°C [7]. Lignin has also been found to decompose much slower than other biomass
components, over a broader temperature range, mainly around 200-500°C [8]. The
degradation products of lignin pyrolysis have been studied by Alén et al. [9]
In this paper we characterize lignin fast pyrolysis at temperatures of 400-550°C. We
present a way of successfully pretreating the lignin so that it can be fed with a screw
feeder into a reactor, and determine the yields of char, bio-oil and light gas (CO and
CO2). We also analyse the composition of the bio-oil. For comparison, we have also
pyrolysed woody biomass with the same laboratory setup.
2.
EXPERIMENTAL
2.1
Materials
Lignin and woody biomass was used as raw materials for the pyrolysis experiments. The
lignin for the experiments was supplied by SEKAB, Sweden. It had been produced by
enzymatic hydrolysis of softwood. The woody biomass used was pine, pinus silvestris,
which had been ground into sawdust in a laboratory mill.
2.2
Fluidized bed reactor setup
The reactor setup used in the pyrolysis experiments can be seen in Figures 1 and 2,
respectively. It consisted of a lignin feeder, a fluidized bed reactor, a set of coolers and
particle filters. The setup has previously been used for catalytic pyrolysis of woody
biomass [10].
The lignin feeder was a Schenk AccuRate MOD102M screw feeder. The feeder, along
with its connection to the reactor, was hermetically sealed so that no air would leak into
the reactor. The fluidized bed reactor was made of Sandvik Sanicro 31 HT alloy. It
consisted of a lower part, where the fluidization gas was heated, and an upper fluidization
part where the lignin was pyrolysed. The length of the pre-heating and pyrolysis parts
where 267 and 102 mm, respectively. The reactor had an inner diameter of 34 mm with a
wall thickness of 4 mm. The reactor was inside a Carbolite split tube furnace model VST
-2-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
12/400, which was controlled by a 3216P1
model controller. The reactor was thus
heated from the outside.
2 l/min
N2
-20°C
Filters
The pyrolysis vapours were transported from
the reactor through a steel pipe which was
Vent
2
heated to 400ºC. The steel pipe led the
vapours into a series of five coolers made of
Gas
Pyrex glass. The first cooler was a jacketed
1
analysis
-78°C
one, while the second and fifth coolers were
cooled with an internal spiral. In the third
1.3 l/min N
and fourth coolers, the pyrolysis vapours
Figure 1: A schematic of the fluidized bed setup. travelled through a spiral inside the coolers.
All five coolers were filled with ethylene
The pyrolysis reactor consists of a preheating
zone (1) and a fluidized bed reactor (2)
glycol cooled with a Lauda Ecoline RE106
to -20°C. After this set of coolers, the
remaining aerosols in the gas stream were captured by two particulate filters, which
consisted of layers of cotton wool in glass tubes. Finally, the pyrolysis vapours were led
into two coolers made of Pyrex glass and cooled by a mixture of -78°C dry ice and
acetone. After this step a side stream of the remaining pyrolysis gases were fed to the CO
and CO2 analysers.
Feeder
2
The gas flows of nitrogen were controlled by Brooks® 5850S mass flow controllers. The
temperature in the pyrolysis reactor was measured by a K-type thermocouple, whose
signal was logged on a computer.
Figure 2: A photograph of the reactor setup
2.3
Gas analysis
A side stream of the gases which remained uncondensed after the cooling were analysed
by a Ten-3-Gas analyser for CO and CO2. The concentrations in volumetric percentage
-3-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
(Xi) were recorded by a computer. The molar flows ( n&i ) of CO and CO2 were calculated
by solving the equations:
X CO =
X CO2 =
n& CO
n& CO
+ n& CO2 + n& N 2
n& CO2
n& CO + n& CO2 + n& N 2
The total masses of CO and CO2 were calculated by integrating the area below the mass
flow curves, which had been calculated by multiplying the molar flows with the molar
weights of CO and CO2. The yield was calculated by dividing the mass of the gases with
the amount of original dry mass of the lignin or biomass used in the experiment.
2.4
Bio-oil analysis
The water content of the bio-oil was analysed by means of Karl Fischer titration. 10 mg
of the bio-oil was titrated in methanol. It was assumed that all water condensed in the first
cooler.
The bio-oil was also analysed by gas chromatography and the compounds were
determined by mass spectra (GC-MS). The analysis procedure has been described before
by Aho et al. [11]. Prior to analysis, the bio-oil was diluted in methanol. The column used
in the GC-MS was an Agilent DB-Petro 50 m column, with an inner diameter of 0.2 mm
and a film thickness of 0.5 µm. The heating of the column began with an isothermal
period at 40°C for 10 minutes. The temperature was then raised to 200°C with four
different successive heating rates, 0.90°C/min between 40°C and 75°C, 1.10°C/min
between 75°C and 120°C, and finally 10°C/min between 120°C and 200°C. At the final
temperature, 200°C, there was an isothermal period for 20 minutes. The inlet pressure of
the column was 135 kPa and the scanning range was 10-300 amu. The applied solvent
delay time was 6 min.
2.5
Experimental procedure
Thermogravimetric analysis was performed on the lignin sample. The lignin, and the
woody biomass sample, was also pyrolysed in the fluidized bed. In the case of
thermogravimetric analysis, the untreated lignin was heated from 30°C to 988°C, with a
20°C/min heating rate in a flow of 100 ml/min nitrogen. At 100°C, the sample was dried
for 15 minutes.
In the case of pyrolysis in the fluidized bed, the untreated lignin in the form of a powder
was not suited for feeding into the reactor. Initial experimental runs indicated that
untreated lignin, which was in the form a fine powder, tended to plug the feeding screw
and cause damage to the feeder, especially since the lignin might partially pyrolyse in the
screw. The lignin was therefore pretreated by suspending it in water and then drying it at
105ºC for 24 hours. This caused the dust particles to agglomerate and form larger
-4-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
particles. The lignin was then screened, and a size fraction of 355-500 µm was used for
the experiments. The same fraction, 355-500 µm, was also used for the biomass sample,
which was in the form of sawdust.
The pyrolysis was carried out in the fluidized bed reactor. The fluidized bed consisted of
40 g of quartz sand, screened to a fraction of 100-150 µm. About 15 g of lignin or
biomass was fed. Fluidization was obtained by feeding 1.3 l/min of nitrogen through the
bottom of the reactor. A nitrogen flow of 2 l/min was also fed through the feeder to
ensure easier feeding. The pyrolysis gases were condensed and captured in the first set of
coolers, in the particulate filters and in the last set of coolers. A side stream of
noncondensable gases was fed to the CO and CO2 analysers. Pyrolysis experiments were
conducted at 400, 450, 500 and 550ºC.
When all biomass had been fed to the reactor and the concentrations of CO and CO2
diminished to zero the heating was turned off. After the experiment the amounts of char
and bio-oil were gravimetrically determined. All the char remained in the bed, while the
bio-oil remained on the surfaces of the coolers and in the filters.
3. RESULTS & DISCUSSION
100 %
0.3
90 %
0.25
Weight [wt-%]
80 %
0.2
70 %
0.15
60 %
0.1
50 %
40 %
0.05
30 %
0
dX/dT [wt-%/°C]
Thermogravimetric analysis illustrated that decomposition starts at about 200°C. At
500°C, about 50% of the lignin remained as char and at 988°C 37% of it remained as
char. The thermogravimetric curve from 150-980°C can be seen in Figure 3. The mass
release rate is highest around 360°C.
950
900
850
800
750
700
650
600
550
500
450
400
350
300
250
200
150
Temperature [°C]
Figure 3: Thermogravimetric weight percentage and mass release rate curves (dX/dT) for the
untreated lignin, at a heating rate of 20°C/min.
The yields of char, oil and gas from the experiments in the fluidized bed reactor are
shown in Figure 5. At 400 and 450°C the char yield was high, about 50%, and the oil
yield low – about 25% – compared to pyrolysis of woody biomass at the same
-5-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
temperature. The CO and CO2 yields were comparable
for both materials. Similar results were observed for
lignin at 500°C. At 550°C, the char yield was lower
while gas yield was very high. The large gas yield was
could be expected since the lignin had agglomerated
and formed larger structures which blocked the flow of
the fluidization gas. This caused the CO and CO2 flows
to seem larger than they actually were, as the flow of
nitrogen was actually lower than the value used in the Figure 4. Agglomeration of char after
calculations. The agglomeration of char at 550°C is pyrolysis at 550ºC.
visualized in Figure 4. The CO and CO2 yields as well
as the CO/CO2 ratio can be seen in Figure 6. The CO/CO2 ratio rose as the temperature
rose.
The mass balance closure was 88-95% for the experiments at 450°C and 500°C. Only
insignificant amounts of condensed vapours were captured in the last coolers, which were
cooled by -78°C dry ice in acetone. The combination of cooling to -20°C and particulate
filters was sufficient to achieve a high mass balance closure. At 550°C, the mass balance
closure exceeded 100% due to the incorrect values for CO and CO2.
110 %
100 %
90 %
CO2
80 %
CO
70 %
Water
60 %
Bio-oil
50 %
40 %
Char
30 %
20 %
10 %
0%
450ºC (woody
biomass)
400ºC (lignin)
450ºC (lignin)
500ºC (lignin)
550ºC (lignin)
Figure 5: The yields of different phases as weight percentage of original dry mass for pyrolysis of
woody biomass and hydrolysis lignin.
-6-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
20 %
0,7
18 %
CO
CO2
0,6
CO/CO2
16 %
0,5
14 %
12 %
0,4
10 %
0,3
8%
6%
0,2
4%
0,1
2%
0%
0
450ºC (woody
biomass)
400ºC (lignin)
450ºC (lignin)
500ºC (lignin)
550ºC (lignin)
Figure 6: Yields of CO and CO2 as weight percentage of original dry mass, and CO/CO2 ratios for
pyrolysis of woody biomass and hydrolysis lignin.
SEM-images of the untreated lignin and char from pyrolysis at 450°C can be seen in
Figure 8. In Figure 9, the untreated and pyrolysed samples can be seen in magnification.
The pyrolysed char sample is significantly sintered compared to the original, untreated
lignin.
A
B
Figure 8: A) untreated lignin, B) char from pyrolysis at 450ºC. The smaller pieces in the image
are sand while the larger are lignin char.
-7-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
A
B
Figure 9: SEM-images at 5000x magnification. A) untreated lignin, B) char from pyrolysis at
450ºC.
The compounds identified by GC-MS can be seen in Figure 10. All identified compounds
are aromatic and most are phenolic, having methoxy side chains.
HO
CH3
HO
HO
OH
benzene-1,2-diol
H3C
CH3
OH
2-methylbenzene-1,3-diol
CH3
H3C
CH3
CH3
H3C
OH
OH
4-methylbenzene-1,2-diol
OH
4-(hydroxymethyl)phenol
CH3
OH
3,4-dimethylphenol
OH
4-ethyl-3-methylphenol
OH
2-ethyl-4,5-dimethylphenol
OH
2-methyl-6-(prop-2-en-1-yl)phenol
CH3
CH3
OH
2-methoxyphenol
O
CH3
O
OH CH3
2-methoxy-3-methylphenol
H3C
OH
2-methoxy-4-methylphenol
O
CH3
O
O
OH
4-hydroxy-2-methoxybenzaldehyde
O
CH3
O
O
H3C
CH3
H2C
H3C
OH
4-hydroxy-3-methoxybenzaldehyde
-8-
OH
4-methoxy-3-methylphenol
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
CH2
CH3
CH3
H2C
O
CH3
CH3
O
OH
2-methoxy-4-[(1E)-prop-1-en-1-yl]phenol
OH
4-ethenyl-2-methoxyphenol
OH
2-methoxy-4-(prop-2-en-1-yl)phenol
OH
H3C
O
H3C
CH3
O
1,4-dimethoxybenzene
CH3
O
OH
2-methoxy-4-propylphenol
H3C
O
O
O
CH3
CH3
O
H3C
1,2-dimethoxy-4-methylbenzene
2,6-dimethoxyphenol
H3C
O
O
CH3
H3C
O
O
O
CH3
H3C O
HO
2-ethoxyphenol
methyl 3,4-dimethoxybenzoate
HO
O
O
H3C
OH
H3C
O CH3
1-(4-hydroxy-3-methoxyphenyl)ethanone
H3C O
OH
O
(4-hydroxy-3-methoxyphenyl)acetic acid
CH3
O
OH
O
methyl 4-hydroxy-3-methoxybenzoate
H3C
O
OH
O
H3C
O CH3
HO
2-(2-methoxyphenyl)ethanol
O
methyl (4-hydroxy-3-methoxyphenyl)acetate
Figure 10: Identified compounds in the bio-oil
4. CONCLUSIONS
Pyrolysis of lignin produced a high amount of char compared to pyrolysis of woody
biomass, around 50% of the original fuel, and lower quantities of bio-oil, about 25%. The
yield of gas was comparable to that of woody biomass. The char yield was slightly lower
at higher temperatures. A number of aromatic compounds which were mostly phenolic
could be identified with GC-MS in the bio-oil.
The mass balance coverage of the laboratory setup was around 90-95%. The combination
of five spiral coolers cooled by -20°C glycol and two particulate filters could capture
most condensable material in the pyrolysis vapours. Further cooling by a -78°C mixture
of dry ice and acetone only condensed insignificant amounts of pyrolysis vapours and did
not improve the mass balance closure.
-9-
The Swedish and Finnish National
Committees of the International Flame
Research Foundation – IFRF
Lignin tended to agglomerate and form larger structures during pyrolysis, especially at
higher temperatures (550°C). This caused problems in fluidization and resulted in
plugging in the reactor. Plugging in the feeding unit was avoided by pretreating the lignin
and feeding a size fraction of 355-500 µm instead of the untreated lignin which was in the
form of a fine dust.
5. REFERENCES
[1] P. Stenius, Papermaking science and technology. Book 3, Forest products chemistry,
Helsinki: Fapet, 2000, ISBN 9525216039
[2] D. Mohan, C.U. Pittman, Jr., P.H. Steele, Pyrolysis of Wood/Biomass for Bio-oil: A
Critical Review, Energy & Fuels 20 (2006) 848-889
[3] C. Amen-Chen, H. Pakdel, C. Roy, Production of monomeric phenols by
thermochemical conversion of biomass, Biosource Technology 79 (2001) 277-299
[4] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, The Catalytic
Valorization of Lignin for the Production of Renewable Chemicals, Chemical Reviews
110 (2010) 3552–3599
[5] M. Brebu, C. Vasile, Thermal degradation of lignin – a review, Cellulose Chemistry
and Technology 44 (9) (2010) 353-363
[6] D.J. Nowakowski, A.V. Bridgwater, D.C. Elliott, D. Meier, P. de Wild, Lignin fast
pyrolysis: Results from an international collaboration, Journal of Analytical and Applied
Pyrolysis 88 (2010) 53–72
[7] R.K. Sharma, J.B. Wooten, V.L. Baliga, X. Lin, W.G. Chan, M. R. Hajaligol,
Characterization of chars from pyrolysis of lignin, Fuel 83 (2004) 1469–1482
[8] H. Yang, R. Yan, H. Chen, C. Zheng, D.H. Lee, D.T. Liang, In-Depth Investigation of
Biomass Pyrolysis Based on Three Major Components: Hemicellulose, Cellulose and
Lignin, Energy & Fuels 20 (2006), 388-393
[9] R. Alén, E. Kuoppala, P. Oesch, Formation of the main degradation compound groups
from wood and its components during pyrolysis, Journal of Analytical and Applied
Pyrolysis, 36 (1996) 137-148
[10] A. Aho, N. Kumar, T. Salmi, B. Holmbom, P. Backman, M. Hupa, D. Yu. Murzin,
Catalytic pyrolysis of woody biomass, Biofuels 1(2) (2010) 261-273
[11] A. Aho, N. Kumar, A.V. Lashkul, K. Eränen, M. Ziolek, P. Decyk, T. Salmi, B.
Holmbom, M. Hupa, D. Yu. Murzin, Catalytic upgrading of woody biomass derived
pyrolysis vapours over iron modified zeolites in a dual-fluidized bed reactor, Fuel 89
(2010) 1992–2000
6. ACKNOWLEDGEMENTS
This work was part of the activities at the Åbo Akademi University Process Chemistry
Centre within the Finnish centre of Excellence Programme (2000-2011) by the Academy
of Finland. SEKAB is acknowledged for providing samples of lignin. Peter Backman is
acknowledged for performing thermogravimetric analysis and Linus Silvander for SEM
analysis.
-10-