Experimental test on a novel dual fluidised bed biomass gasifier for synthetic fuel production

Fuel 90 (2011) 1340–1349

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Experimental test on a novel dual fluidised bed biomass gasifier
for synthetic fuel production
K. Göransson ⇑, U. Söderlind, W. Zhang
Department of Natural Sciences, Engineering and Mathematics, Mid Sweden University, SE-871 88 Härnösand, Sweden

a r t i c l e

i n f o

Article history:
Received 15 September 2009
Received in revised form 22 December 2010
Accepted 29 December 2010
Available online 12 January 2011
Keywords:
Allothermal gasification
Biofuel
Biomass
Gasifier and syngas

a b s t r a c t
This article presents a preliminary test on the 150 kWth allothermal biomass gasifier at Mid Sweden University (MIUN) in Härnösand, Sweden. The MIUN gasifier is a combination of a fluidised bed gasifier and a
CFB riser as a combustor with a design suitable for in-built tar/CH4 catalytic reforming. The test was carried out by two steps: (1) fluid-dynamic study; (2) measurements of gas composition and tar. A novel
solid circulation measurement system which works at high bed temperatures is developed in the presented work. The results show the dependency of bed material circulation rate on the superficial gas

velocity in the combustor, the bed material inventory and the aeration of solids flow between the bottoms of the gasifier and the combustor. A strong influence of circulation rate on the temperature difference between the combustor and the gasifier was identified. The syngas analysis showed that, as steam/
biomass (S/B) ratio increases, CH4 content decreases and H2/CO ratio increases. Furthermore the total tar
content decreases with increasing steam/biomass ratio and increasing temperature. The biomass gasification technology at MIUN is simple, cheap, reliable, and can obtain a syngas of high CO + H2 concentration with sufficient high ratio of H2 to CO, which may be suitable for synthesis of methane, DME, FT-fuels
or alcohol fuels. The measurement results of MIUN gasifier have been compared with other gasifiers. The
main differences can be observed in the H2 and the CO content, as well as the tar content. These can be
explained by differences in the feed systems, operating temperature, S/B ratio or bed material catalytic
effect, etc.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction
Synthetic fuel production from biomass is an important issue
from the viewpoint of climatic conventions and energy shortage
crises. Synthetic fuels such as methane, DME, FT-fuels, and alcohol
fuels, as the second generation bio-automotive fuels, can be produced via gasification and synthesis based on various forest and
agricultural biomass residues. For biomass, a S&M (small or medium) scale bio-automotive fuel plant is preferable as biomass feedstock is widely sparse, inhomogeneous in size and shape, difficult
to be pulverized, and has relatively low density, low heating value,
and high moisture content [1]. Direct gasification with air produces
a fuel gas of heating value 4–7 MJ/Nm3 – not suitable for synthesis
of bio-automotive fuels. Pure oxygen gasification generates a fuel
heating value of 10–12 MJ/Nm3, but an oxygen plant is needed,
which can be economic only for large-scale bio-automotive fuel
production plants [2]. However, indirect (or allothermal) gasification, in a dual fluidised bed gasifier (DFBG) with steam as the
gasification agent, produces a syngas of 12–20 MJ/Nm3 heating va⇑ Corresponding author. Tel.: +46 70 289 26 30, +46 611 862 13; fax: +46 611 861
60.
E-mail address: (K. Göransson).
0016-2361/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fuel.2010.12.035

lue. Thus, DFBG turns out to be a promising biomass gasification
technology for synthetic fuel production.

There are a number of DFBGs or similar designs worldwide: the
well known Güssing gasifier in Austria, e.g. [3], MILENA gasifier in
The Netherlands [3–5], Trisaia gasifier in Italy by ENEA’s research
center [3,6], Battelle Columbus Laboratories (BCL) gasifier in USA
[3] (now called The Rentech-SilvaGas Process [7]), and the CAPE
FICFB Gasifier in New Zealand by the University of Canterbury [8].
A DFBG of 2–4 MWth was recently built by Göteborg Energi at Chalmers University of Technology in Sweden [9]. In Japan there is a
DFBG in Yokohama, by Xu et al. [10,11] and in China there are
some examples in Beijing by the Chinese Academy of Sciences
[3] and in Hangzhou by Fang and co-workers [12]. DFBGs can be
designed with different combinations of the BFB and the CFB. So
far, the most attractive design is supposed to have biomass gasification in the BFB and char combustion in the CFB. Examples of this
design are the Trisia gasifier, the Güssing gasifier, the CAPE FICFB
gasifier, the MIUN gasifier, etc. The principle of the Chalmers gasifier is similar to these gasifiers as well, as a BFB gasifier is integrated into the loop of an existing 12 MWth research CFB-boiler
[13]. The MILENA gasification process uses the riser for gasification
and the BFB for combustion [4], and the Rentech-SilvaGas Process
consists of two CFBs interconnected with each other [7].


K. Göransson et al. / Fuel 90 (2011) 1340–1349

1341

Nomenclature
BFB
CFB
db
DFB
DFBG
FB

GC

bubbling fluidised bed
circulating fluidised bed
dry basis
dual fluidised bed
dual fluidised bed gasifier
fluidised bed
Gas Chromatograph

In general, the biomass gasification process occurs through
three steps: (1) pyrolysis, which produces volatile matter and char
residue; (2) secondary reactions, involving the volatile products;
(3) gasification reactions of the remaining carbonaceous residue
with steam and carbon dioxide [14]. The pyrolysis of biomass results in volatiles and char that subsequently participate in a series
of complex and competing reactions. The main conversion of the
biomass to syngas takes place within the bed, but some conversion
to syngas occurs in the freeboard section. In DFBGs, biomass is gasified in a bed fluidised with steam. The composition of the syngas is
mainly dependent on the type of gasification agent used. Steam
encourages the shift reactions with carbon and carbon monoxide
to increase the hydrogen content. Shift reactions and steam
reforming of methane reaction are common reactions used to predict the composition of syngas.
Synthetic fuel production requires a high quality syngas of high
CO + H2 concentration with sufficient high ratio of H2 to CO and
low tar content. Steam-to-biomass (S/B) ratios, gasification temperature and bed material circulation rate, are important parameters in
determining the syngas composition and tar content. The required
heat for biomass steam gasification is maintained by the char
combustion and the bed material circulation. The bed material
circulation rate can control the temperature balance, and is an
important parameter to be considered in the gasifier design and

operation. The operation of the gasifier needs to be optimised, and
hence, the operational behaviour of the gasifier needs to be studied.
Attempt to measure the solid circulation rate have been performed both in old proven ways like ‘‘bucket-and-stopwatch’’
and in new imaginative ways, e.g. by using signals from an optical
mouse to PC [15] or a fibreglass spiral with a rotation electronics
[16]. Unfortunately, most of the existing techniques for measuring
solid circulation rate are limited for various reasons. A novel solid
circulation measurement system is developed in the presented
work, which is called here as Pressure-induced Measurement of
Circulation (PIMC), a technique that also works at high bed temperatures under gasification conditions.
The 150 kW allothermal biomass MIUN gasifier has been built
up in 2008 for the research on synthetic fuel production. This paper
presents a preliminary test of the operational behaviour of the
MIUN gasifier. The test was carried out by two steps: (1) fluid-dynamic study; (2) measurements of gas composition and tar. The
experimental results of MIUN gasifier have been compared with
other DFBGs, which leads to a discussion on DFBG design and operation. This paper shows a successful development of a pilot-scale
DFBG with internal reforming of tar and methane for synthesis
gas production.
2. Experimental facilities
The BTL (biomass to liquids) system at MIUN is sketched in
Fig. 1. It has been presented elsewhere [17]. A description focusing
on the gasifier is given below.

Gs
MIUN
PIMC
S/B
Uc
Ug

solid circulation rate (kg/m2 s)
Mid Sweden University
Pressure-induced Measurement of Circulation
steam/biomass ratio (kg/kg dry biomass)
superficial gas velocity in the combustor (m/s)
superficial gas velocity in the gasifier (m/s)

2.1. The gasifier and test conditions
The gasifier (see Fig. 2) consists of an endothermic steam fluidised bed gasifier and an exothermal CFB riser combustor, and has a
biomass treatment capacity of 150 kWth, i.e. approx. 30 kg wood
pellet feed per hour.
The heat carrier between the reactors is silica sand of about
150 lm diameter. The bed material circulation is controlled with
the gas velocity in the combustor, the total solids inventory and
aeration in the tube connecting the bottoms of the gasifier and
the combustor. The gas flow through the gasifier can also enhance the aeration in the abovementioned tube, so that the
bed material circulation increases with gas velocity in the
gasifier.
The PLC-operated (Process Logic Controller) feeding system is
designed for constant biomass feeding. The biomass is fed into a
pneumatic oscillatory vane feeder to achieve high precision. A
coaxial left- and right-handed thread screw feeder thereafter
makes a steady stream of biomass into the final screw feeder without plugging the pipe. The final screw feeder is rapidly feeding biomass into the gasifier to avoid reaction in the screw feeder and
particle congestion. The feeding rate can be controlled by frequency-controlled motors of screw feeder. Nearest to the gasifier,
the final screw feeder is cooled by water flow through a cylindrical
jacket.
The gas distributors in both the gasifier and the combustor are
own-produced sintered metal plates. The fluidisation agent in the
gasifier is steam (12 bar and 150 °C) and the synthesis gas is
drawn off from the top of the gasifier. The residual biomass char

is then transferred by bed material into the combustor through
the lower pressure lock. In the combustor, the fluidisation agent
is air, which results in an oxidation of the char and produces heat
at a temperature of 950–1050 °C. The hot bed material separates
from the flue gas in a cyclone to be recycled into the gasifier
through the upper pressure lock (loop seal pot), which prevents
gas leakage between the separate environments in the gasifier
and the combustor. The aeration medium in the upper pressure
lock is steam.
The steam gasifier is surrounded by electrical heaters (total effect of 20 kW) and insulated. There are no lining in the reactors.
The gasifier and the combustor have a height of 2.5 and 3.1 m,
and inner diameters (i.d.) of 300 and 90 mm, respectively.
Temperatures and pressures, at a number of points from
the distributor to the top of the gasifier and the combustor,
at the upper pressure lock and at the cyclone, as well as all
gas flows, are registered through a computer data collection
system.
The test on the DFBG was carried out at the temperatures 750,
800 and 850 °C, in the fluidised bed gasifier. The steam supply into
the gasifier was held at 4 kg/h. The biomass feedstock was wood
pellets from SCA BioNorr AB, and the fuel analysis is given in
Table 1.


1342

K. Göransson et al. / Fuel 90 (2011) 1340–1349

Fig. 1. The BTL system at MIUN with (A) biomass feeding system, (B) gasifier, (C) syngas cleaning system, and (D) catalysis reactor.

Table 1
Fuel analysis of wood pellets from SCA
BioNorr AB [18].
Elementary analysis

% db

Ash cont.
Sulphur, S
Chlorine, Cl
Carbon, C
Hydrogen, H
Nitrogen, N
Oxygen, O (calc.)
Moisture
Durability of pellets
Bulk density
Calorific value as rec.
Calorific value db

0.4
<0.01
<0.01
50.9
5.9
<0.1
42.7
6.5%
98.1%
667 kg/m3

18.849 MJ/kg
20.159 MJ/kg

The steam/biomass ratios in weight were 0.3, 0.6 and 0.9 calculated according to Eq. (1). The biomass input was 95, 39 and 27 kW
respectively. The surface of bed material in the gasifier was 0.7 m
above the distributor, and about 0.2 m above the biomass feeding
point.

steam=biomass ratio ¼

Fig. 2. The MIUN dual fluidised bed gasifier.

_ H2 O in biomass
_ steam þ m
m
_ dry biomass
m

ð1Þ

_ steam represents the mass flow of steam (kg/s), m
_ H2 O
where m
represents the mass flow of water in the biomass (kg/s),
_ dry biomass represents the flow of dry biomass (kg/s).
m


K. Göransson et al. / Fuel 90 (2011) 1340–1349

2.2. The solid circulation measurement system,
According to Bernoulli’s equation, the total pressure (pt) of a flow
is the sum of the static pressure (ps), the hydrostatic pressure (ph)
and the dynamic pressure (pd) in the flow. The dynamic pressure is
associated with the gas velocity, and represents the kinetic energy
of the flow. A common form of Bernoulli’s equation is as follows:

pt ¼ ps þ ph þ pd
ph ¼ qgh pd ¼ q

u2
2

where q represents the density of the fluid (kg/m3), g represents the
gravitational acceleration (m/s2), h represents the elevation above
the reference plane (m), u represents the velocity of the fluid (m/s).
As shown in Fig. 3, the solid circulation rate (Gs) is measured by
the solid circulation measurement system PIMC, with two pressure
transducers. Purge gas is used to create a dynamic pressure (pd).
The positive side on each transducer is connected to a purge gas
nozzle inside the downcomer under the combustor cyclone (see
Fig. 3). Transducer A is connected to purge gas nozzle A, and transducer B is connected to purge gas nozzle B. A glass tube, that is
holding a certain height of a water column, is connected to the
transducer’s negative side. The water column imposes a constant
hydrostatic pressure so that the transducers indicate a negative
pressure in normal condition.

1343

By turning off the fluidisation of the upper pressure lock, the

circulation in the gasifier is interrupted and the solids accumulate
in the upper pressure lock. Hence, the level of bed material will
start to rise in the downcomer. When the bed material reaches
and blocks the nozzle B and then the nozzle A in sequence, the
purge gas entering the downcomer is stopped. That creates a dynamic pressure in nozzle B and then in nozzle A in sequence on
the positive side of the transducer. As a consequence, the pressure
registered by the transducers will turn to positive in sequence.
Thus, the time delay of the pressure changes between the two nozzles is registered and used to indicate the time used for bed material to fill the specific volume in the downcomer. The distance
between the two nozzle locations inside the downcomer (40 mm
i.d.) is 200 mm.
Hence, the solids circulation rate can be calculated as below.

Gs ¼

md
t  Ar

where Gs represents the solid circulation rate (kg/m2 s), md represents the mass of the bed material in the downcomer between nozzle B and nozzle A (kg), Ar represents the cross-section area of the
riser (m2), t represents the time used for bed material to fill the
downcomer between nozzle B, and nozzle A (s).
2.3. Measurements of gas composition
The major syngas stream from the gasifier is led to a special gas
burner for complete combustion of the syngas. A minor stream of
the syngas passes through the gas sampling and analysis system.
To avoid tars condensation in the pipes, electrical heaters outside
the pipes hold the temperature of the pipes at about 400 °C.
An overview picture of the syngas sampling and analysis system
can be seen in Fig. 4. The sampling procedure was repeated three
times for each experimental run.
For the measurements of the gas composition, the syngas is

drawn by a vacuum pump through a gas conditioning step and
sampled manually in a gas sampling bag (Cali-5-bond) and analysed off-line in a parallel FID and TCD detection GC-system.
2.4. Tars measurement

Fig. 3. Pressure-induced Measurement of Circulation (PIMC), the solids circulation
rate measurement system developed at MIUN.

Tars can be divided into heavy tars and light tars. Heavy tars
comprise high molecular weight hydrocarbon compounds that
have a boiling point higher than 350 °C. Light tars comprise the
range in molecular weight from $78 to $300, which are volatile
and semi-volatile aromatics and phenolics. The tar sampling takes
place at the same time as the syngas sampling, as seen in Fig. 4. The
total tar and light tar are measured in two different ways with
gravimetrical analysis and SPA method respectively.
The total tars are gravimetrically analysed. First the product gas
passes a high temperature filter usually kept at 380 °C during sampling, and cooled down to 40 °C while passing through two tar capture glass fibre filters in series. A known volume (4 l) of gas
measured by a flow meter was drawn by a vacuum pump through
the sampling during about 45 s. After sampling the glass fibre filters were frozen in aluminium foil envelopes, which later were
placed in a clean round flask containing isopropanol. After about
18 h the glass fibre filters and the parts of the foil were washed
out with isopropanol. The solution containing the tar sample together with some particles and filter fibres is filtered through a
glass fibre funnel to a new round flask that has been weighted.
The round flask is inserted into an evaporation condenser (rotation
60 rpm in water bath 45 °C). When the tar sticks on the inside of
the round flask the evaporation is finished. The weight difference,
i.e. the tar content, can finally be calculated.


1344

K. Göransson et al. / Fuel 90 (2011) 1340–1349

Fig. 4. An overview picture of sampling and analysis of syngas and tars, with (A) gasifier, (B) syngas burner, (C) syngas sampling, (D) total tars sampling (E) light tars sampling
(SPA method), (F) ventilation. (The syngas sampling with (1) impinger bottles, (2) cooler, (3) drying filters, (4) vacuum pump and (5) the outlet to GC-system; the total tar
sampling with (6) hot gas filter, (7) glass fibre filters, (8) vacuum pump, (9) flowmeter and (10) total flowmeter; the SPA tar sampling with (11) Amino phase cartridge and
(12) syringe 100 ml.).


1345

K. Göransson et al. / Fuel 90 (2011) 1340–1349
18,0

Solid Circulation Rate (kg/m2s)

The light tars were analysed according to the Solid-Phase
Adsorption (SPA) method. The samples were taken by a septum
port of an electrically heated T-connection (400 °C) inserted in
the sampling line, in the same time as the total tars sampling as
shown in Fig. 4. A known amount of gas is extracted by using a
100 ml syringe filled at one minute. The tar vapours are trapped
on amino propyl-bonded silica sorbent packed in a small cartridge
which was later sent to Royal Institute of Technology at Stockholm
for analysis. Light tars can pass through a nonpolar GC-column
while heavy tars cannot. The SPA method is used for compounds
which are enough volatile to be separated on a GC-column, and
is calibrated for 18 aromatics and 10 phenolics.

16,0

Bed inventory: 160 kg

14,0

Bed inventory: 145 kg

12,0

Bed inventory: 130 kg

10,0
8,0
6,0
4,0
2,0
0,0
2

2,2

2,4

2,6

2,8

3

3,2

3,4

Gas Velocity in Combustor (m/s)

3. Results and discussion

Fig. 6. The solid circulation rate as a function of bed inventory and the gas velocity
in the combustor.

To maintain a sufficient heat transfer between the gasifier and
the combustor, the solid circulation rate, Gs, must be held at a certain high level. Gs is mainly determined by the gas velocity in the
combustor, Uc, and the bed material inventory. Gs is also affected
by the gas velocity in the gasifier, Ug, since it is serving as an aeration of the lower pressure lock. The lower pressure lock is connecting the gasifier and the combustor.
Fig. 5 shows the effect of Ug on Gs, while Uc was kept constant at 3.5 m/s. Gs increased gradually with Ug. The results in
Fig. 6 show the dependency of solids circulation rate on the
superficial gas velocity in the combustor and the bed material
inventory. The bed inventory strongly influences Gs due to the
change of solids concentration in the upper part of the
combustor.
A circulation rate of 20 kg bed material per 1 kg biomass is required in normal gasifier operation condition. At the full load of
the 150 kW gasifier, i.e. approx. 30 kg wood pellets feed per hour,
a solid circulation rate of 600 kg bed material per hour is needed,
which corresponds to a bed material circulation rate of 26 kg/m2 s.
The temperature difference between the gasifier and the combustor is strongly dependent on Gs as seen in Fig. 7. A high Gs gives
a small temperature difference and vice versa. Gs values that exceeded 10 kg/m2 s is required for a sufficiently low temperature
difference.

200

Temperature Difference (oC)

3.1. Bed material circulation

180
160
140
120
100
80
60
4,0

5,0

6,0

7,0

8,0

9,0

10,0

12,0

13,0

Fig. 7. The temperature difference between the gasifier and the combustor as a

function of the solids circulation rate.

Oxidation:

C þ O2 ! CO2

DH À 394 kJ=mol

1
C þ O2 ! CO DH À 111 kJ=mol
2

ðAÞ
ðBÞ

Boudouard:

C þ CO2 ! 2CO DH þ 172 kJ=mol

3.2. Syngas composition

11,0

Solid Circulation Rate (kg/m2s)

ðCÞ

Water–gas (steam oxidation):
The pyrolysis of biomass is the first step of biomass gasification,
which results in volatiles and char. The volatile and char subsequently participate in a series of complex and competing reactions,

as given below [14,19,20].

ð1stÞ C þ H2 O ! CO þ H2

DH þ 131 kJ=mol

ð2ndÞ C þ 2H2 O ! CO2 þ 2H2

DH þ 97 kJ=mol

ðDÞ
ðEÞ

Methanation:

C þ 2H2 ! CH4
Solid Circulation Rate (kg/m2s)

20
18
16

ðFÞ

Water–gas shift:
Bed inventory: 145 kg
Uc: 3,5 m/s

CO þ H2 O ! CO2 þ H2

DH À 41 kJ=mol

ðGÞ

Steam reformation:

14
12

CH4 þ H2 O ! CO þ 3H2

10
8
6
4
2
0
0,00

DH À 75 kJ=mol

0,02

0,04

0,06

0,08

0,10

0,12

0,14

Gas Velocity in Gasifier (m/s)
Fig. 5. The solid circulation rate as a function of gas velocity in the FB gasifier.

DH þ 206 kJ=mol

ðHÞ

According to Franco et al. [14], the produced gas composition in
steam gasification at temperatures in the range of 730–830 °C is
dependent on the water–gas shift reaction (G), together with the
reforming and cracking reactions.
At temperatures higher than 830 °C, the water–gas shift reaction becomes less important as the Boudouard reaction (C) and
the water–gas reactions (D) and (E) become dominant.
The syngas measurement results from the MIUN gasifier are
summarized in Table 2 with the averaged values of syngas components


1346

K. Göransson et al. / Fuel 90 (2011) 1340–1349

Table 2
Syngas composition (vol.% dry.) as a function of the steam/biomass (S/B) ratio at 750,
800 and 850 °C.
Temp.

750 °C

H2
CO
CO2
N2
O2
CH4
C2H4
C2H6
C3Hx
C4Hx
S/B (kg/kg)

45.8
40.9
6.1
1.6
0.2
13.4
3.2
–
4.1
–
0.3

800 °C
51.1
35.3

4.8
2.3
0.5
11.4
3.0
–
1.9
–
0.6

62.1
28.4
4.9
2.6
0.5
9.3
2.1
–
0.1
–
0.9

41.5
40.8
7.5
5.5
1.6
13.4
3.7
–

2.0
–
0.3

850 °C
41.9
37.1
9.3
3.2
0.8
12.3
3.4
–
1.7
–
0.6

50.9
29.6
12.6
2.0
0.5
9.5
2.6
–
2.5
–
0.9

40.8

42.1
9.5
0.5
0.2
13.6
4.0
–
3.7
–
0.3

45.6
36.4
9.8
0.9
0.2
11.3
3.3
–
1.5
–
0.6

51.3
36.9
11.2
0.2
0
11.0
3.4

–
2.9
–
0.9

16,0

5,0

14,0

4,5

3.3. Tar content
Tars in the syngas not only results in a part of energy lost from
syngas before synthesis, but also leads to many knotty problems to
downstream processes, blocking of filters and pipeline passages,
dirty working environment, heavy waste treatment, poisoning of
catalyst in synthesis reactors and so on.
A good gasifier should provide syngas with tar content under 1–
2 g/m3 to facilitate the downstream syngas cleaning. After syngas
cleaning, tar content should be under 10 mg/m3 for engine and turbine applications, and under 1 ppmv for most synthesis reactors.
However, as seen in Fig. 9, the content of tar from MIUN gasifier
with sand bed material ranges from 10 to 65 g/Nm3, although the
tar content decreases as S/B and temperature increase. Increasing
S/B means that the conversion of the H2O in the gasifier gradually
decreases, and hence a greater proportion of steam bypass the gasifier unused [25].

70

3,5

850°C

10,0
8,0
6,0

800°C

3,0

750°C
750°C

2,5

800°C

2,0

850°C

1,5
4,0
1,0
2,0
0,0
0,2

0,5

Total tar content (g/m n3 )

4,0
12,0

H2/CO ratio

Concentration of CH4
(vol% dry gas)

for each experimental condition. In the temperature range of 750–
850 °C, the H2 yield increases with S/B, while slightly decreases
with temperature. The CO and CH4 yields decrease with S/B, but
are fairly stable within the given temperature range.
The decreasing trend of the H2 yield with temperature may be
governed by the exothermic water–gas shift reaction as a higher
temperature pushes the reaction to the left hand at the expense
of H2. Those endothermic reactions including the Boudouard, the
water–gas and the steam reforming reactions are in favour of H2
yield, but have not been dominating reaction network under the
present low temperature experimental conditions. A higher S/B ratio means a higher steam partial pressure that pushes the water–
gas shift reaction to the right hand for more H2 production at the
expense of CO. Thus, the CO yield decreases with S/B. All reactions,
including Boudoyard, water–gas, water–gas shift and steam reformation reactions, suggest an increase of the CO yield with temperature. But the temperature effect have not been significant within
the given temperature range. A higher gasification temperature
will be in favour of the H2 and CO yields since the endothermic gasification reactions (C), (D), (E), and (H) are enhanced.
At low S/B ratio the carbon is only partially gasified. A higher S/B
ratio results in greater carbon conversion as a result of the enhanced water–gas shift reaction and steam reformation reactions.

Thus the methane and carbon monoxide contents decrease, and
the hydrogen and carbon dioxide contents increase. When carbon
conversion is completed, an increase of the S/B ratio just results
in dilution of the syngas [21]. Within a range of S/B ratios 0.3–
0.9, as seen in Table 2 and Fig. 8, the methane and carbon monoxide contents decrease whereas the hydrogen and carbon dioxide
contents increase with increasing S/B ratios. This result is in good
agreement with the Refs. [22,23].

When the syngas is used for synthesis of transportation fuel
such as FT-diesel, DME, and alcohols, a high CO + H2 concentration
(>80%) with a high H2/CO ratio in the syngas is required to ensure
downstream synthesis smoothly. The stoichiometric H2/CO ratio is
2 for the synthesis reaction of Fischer–Tropsch and Methanol, 2.5
for DME and 3 for Methanation (Bio-SNG). The stoichiometric ratio
is the optimal H2/CO ratio to obtain a maximum yield of the product. In practise the ratio differs from the optimal ratio, but a certain
high H2/CO is still necessary [24].
Table 2 shows the CO + H2 concentration ranges from 80 to 90%,
the higher concentration at higher S/B ratio, and the H2/CO ratios
from $1.0 to $2.2, the higher ratio at higher S/B ratio as seen in
Fig. 8. The MIUN gasifier can obtain a syngas with sufficient high
ratio of H2 to CO at high S/B ratio. That is insensitive to the temperatures within the given temperature range.
H2/CO adjustment can be achieved with water–gas shift reaction; either by an additional step in a separate reactor, or in terms
of DFBGs, by using the so-called Absorption Enhanced Reforming
(AER) process. The AER process uses calcium carbonate as bed
material for in situ CO2 capture during gasification. The continuous
CO2 removal during the gasification enhances the reforming reactions and the water–gas shift reaction towards H2 production [24].
The CH4 content is not sensitive to the temperature change
within the given range (750–850 °C), but decreases significantly
with increasing S/B, as seen in Fig. 8 as a result of the enhanced
steam reformation of methane. The CH4 range from 9.5% to 13.5%

in the syngas represent 1/3 of energy contained in the syngas,
and would be lost in once through synthesis of transportation fuels
or chemicals. Such CH4 content must be reduced by thermal cracking and catalytic reforming to CO and H2 before syngas entering
the synthesis reactors.

60
750oC
800oC
850oC

50
40
30
20
10

0,0
0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

Steam/Biomass ratio (S/B)

0
0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

Steam/Biomass ratio (S/B)
Fig. 8. The H2/CO ratio and the concentration of methane as a function of S/B ratio
at 750, 800 and 850 °C.

Fig. 9. The total tar content as a function of S/B ratio at 750, 800 and 850 °C.

1347

K. Göransson et al. / Fuel 90 (2011) 1340–1349

3.4. Comparison of the MIUN gasifier with other indirect gasifiers

3

Light tar content (g/mn )

30
25

850
800
750

20
15
10
5
0
0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

Steam/Biomass ratio (S/B)
Fig. 10. The light tar content as a function of S/B ratio at 750, 800 and 850 °C.

3

light tar content (g/m n )
Be
nz
en
T
e
m olue
/p
-X ne
y
o- lene
Xy
le
ne

In
da
2M N In n
d
e a
1- thy pht ene
M ln h a
et ap le
hy h
n
ln tha e
ap le
ht ne
ha
le
Ac
en Bip ne
ap he
A c h t ny
en hy l
a p le n
ht e
he
Ph Flu ne
en ore
an ne
A n th r
th en
Fl rac e
uo en

ra e
nt
e
P y ne
re
P h ne
e
o- no
C l
r
m eso
-C l
r
2, p eso
5/ -C l
3, re
5- so
X
2 , y le l
4- no
Xy l
2, l e
6- no
X
3, yle l
4- no
Xy l
le
no
l

As seen in Fig. 10, the light tars content decreases with increasing S/B ratio, but unlike the total tars content, the light tars content
increases with increasing temperature – with the exception of the
case at S/B ratio 0.9, where the light tar content is lowest at the
temperature 800 °C.
The light tars content increases with temperature because the
heavy tars are cracked to light tars as the temperature increases.
This trend can be identified by Fig. 11, where the high molecular
weight compounds of the light tars like phenol are decreasing with
increasing temperature while the low molecular weight compounds like naphthalene are increasing. The exceptional case could
be attributed to the co-action of light tar reduction by thermal
cracking at higher temperatures and steam reforming.
For MIUN gasifier, the tar content is too high to be used for synthesis applications, and should be reduced further. The tar reduction can be realized by (1) optimum of gasifiers as described
above, (2) catalytic in-bed material, and (3) downstream measures.
Catalytic bed material (such as olivine, dolomite, nickel or iron) is
often used to crack down tar inside gasifiers as a primary method.
Although primary measures are of importance, to keep the thermodynamic efficiency losses to a minimum, downstream methods,
such as external reforming, thermal cracking, catalytic cracking
or mechanical separation (such as scrubber, filter, electrostatic precipitator or cyclone separator), are required when the syngas is
used for synthesis.
A scrubber is easy to run and has beneficial effect on air pollution control. The scrubber agent can be water or a scrubbing liquid
compatible with tar. In terms of tar scrubbing with organic liquid
solvents, the tars are recycled to gasifier and destructed avoiding
wastewater production. The mechanical separation technologies
are often applied in combination or together with catalytic tar removal technologies.

A comparison of the MIUN gasifier to the Güssing gasifier, the
MILENA gasifier, the CAPE gasifier, and the Chalmers gasifier, is given in Table 3, with regards to syngas composition and tar content.
The main differences in the syngas of the gasifiers can be observed
in the H2 and the CO content. The H2/CO ratios of the MIUN gasifier

and the Güssing gasifier are high (1.7 and 1.8–1.9 respectively).
The content of H2 in the syngas of the MILENA gasifier, the CAPE
gasifier and the Chalmers gasifier are quite low, which leads to
the H2/CO ratio close to 0.5–0.8. The CH4 content in the syngas of
the MILENA gasifier is slightly higher than in the other gasifiers.
The Chalmers gasifier and the MIUN gasifier have similar tar contents (20 g/Nm3). The Güssing gasifier has the lowest tar content
(4–8 g/Nm3) while the MILENA gasifier has the highest tar content
($40 g/Nm3). The CAPE gasifier has a high content of N2 in the syngas, probably caused by the nitrogen purge gas in the biomass
feeding system.
The water–gas shift reaction and the steam reforming of methane reaction are common reactions used to explain changes in
syngas composition. The most important parameter governing
the reactions are S/B as has been discussed previously. The first
reason behind the difference in H2/CO ratios is that the S/B ratio
used in the MIUN gasifier and the Güssing gasifier is higher than
that for the MILENA gasifier, the CAPE gasifier and the Chalmers
gasifier.
The second reason is attributed to the difference in the biomass
feeding location. Biomass were fed in the beds of the MIUN gasifier
and the Güssing gasifier, but on the top of the beds for the Chalmers gasifier and the CAPE gasifier. The location of the feeding
point to the gasifier influences the product distribution, since the
heating rate of the biomass depends on the location where it is
fed [25]. High heating rates produce more light gases and less char
and condensate. Improved mixing of the biomass and the bed
material, with higher temperature and longer residence times will
drive the syngas composition in the direction of equilibrium [26].
When the feed point is situated above the bed surface, the relatively low density of the biomass particles precludes good mixing
with the bed material [26].
The poor contact between the volatiles and the bed material
and extremely low temperature in the freeboard lead to a limited
water–gas shift reaction so that a certain amount of CO has not

been shifted to H2 by steam, which gives a low H2/CO ratio.
The differences in the H2/CO ratio can also be attributed to the
factors such as the type and the size of the biomass and of the bed
material (for instance, olivine influences the water shift equilibrium), and some variations of the sampling procedure.
The CH4 content in the syngas of the MILENA gasifier with a low
S/B ratio is slightly higher than in the syngas of the other gasifiers.

6

750°C S/B 0,3
800°C S/B 0,3
850°C S/B 0,3

5
4
3
2
1
0

Fig. 11. The components of the light tars.


1348

K. Göransson et al. / Fuel 90 (2011) 1340–1349

Table 3
Steam gasification syngas composition (vol.% dry.).

a
b
c
d

DFB gasifier

MIUN 150 kWth

Güssing FICFB
8 MWth [23,24]

MILENA 800 kWth
[5,27,28]

CAPE FICFB
100 kWth [8,19]

Chalmers 2–4 MWth
[29,30]

H2
CO
CO2
N2
O2
CH4
C2H4
C2H6
C3Hx

Tar (g/Nm3)

50.9
29.6
12.6
2.0
0.5
9.5
2.6
–
2.5
20

36–42
19–24
20–25
<1
No info.
9–12
2–2.6
1.3–1.8
0.3–0.6
4–8

$17–21a
$36–40a
$12a
No info.
$0a
$14a

No info.
No info.
No info.
$40

21.7
28.4
17.4
16.9d
No info.
11.6
3.5
0.5
No info.
No info.

18.74–27.00
26.96–33.33
14.62–18.32
No info.
No info.
11.11–11.74
3.93–4.26
<1
<1
$20

Temp. (°C)
Feedstock Fuel feeding

800 °C
Wood pellets into the side
of the bed in the gasifier

850–900 °C
Biomass chips fed into the
gasification-reactor bed

800 °C
Wood pellets into the
riser (i.e. the gasifier)

753 °C
Wood chips/pellets on the
top of the bed of the gasifier

S/B
Bed material
H2/CO

0.9
Silica sand
1.7

$0.75 [31]
Olivine
1.8–1.9

$0.12b 0.02–0.35c
Sand

0.5

0.8
Greywacke sand
0.8

780–830 °C
Wood chips/pellets fed
from the top of the
gasifier
$0.5[30]
Silica sand
0.7–0.8

Raw gas composition according to Fig. 4 in [5].
Basic design data MILENA pilot plant [5].
Normal range for the MILENA lab-scale gasifier [28].
Nitrogen purge gas feed into the biomass feeding system ($10 l/min) and air fluidising of the siphon [19].

That is reasonably when the carbon is only partially gasified at low
S/B ratio.
The differences in the tar content can be explained based on the
influences of the bed material, the gasification temperature and
also the S/B ratio. The Güssing gasifier uses a tar cracking active
bed material (olivine) at a high (850–900 °C) gasification temperature. The MILENA gasifier uses sand and a low S/B ratio.
In a fluidised bed gasifier, the temperature in the freeboard is
lower than in the bed. At higher temperatures the endothermic
gasification reactions (such as the Boudouard reaction (C), the
water gas reactions (D) and (E) and the steam reforming reaction
(H)) are enhanced. This give rise to a higher H2 and CO contents

and a lower tar and methane contents.

H2/CO ratios of the Güssing gasifier and the MIUN gasifier are high
(1.8–1.9 and 1.7 respectively), while the MILENA gasifier, the Chalmers gasifier and the CAPE gasifier have a relatively low content of
H2 in the syngas. That can partly be explained by differences in the
feed systems and the operating temperature.
Finally, differences in the tar content can be seen. The Chalmers
gasifier and the MIUN gasifier have similar tar contents (20 g/
Nm3). The Güssing gasifier has the lowest tar content (4–8 g/
Nm3) while the MILENA gasifier has the highest tar content
($40 g/Nm3). The probable reason is differences in the bed material catalytic effect, the gasification temperature and the S/B ratio.

Acknowledgements
4. Conclusion
A preliminary test on the 150 kW allothermal biomass gasifier
at Mid Sweden University was carried out. A novel method to measure solid circulation rate is developed, which works well at high
temperature condition. The test provides basic information for
temperature control in the combustor and the gasifier by the bed
material circulation rate. The circulation rate is, in turn, determined by the superficial gas velocity in the combustor, the bed
material inventory, and the aeration of the lower pressure lock.
Measurements of gas composition as a function of temperature
and the steam/biomass ratio have been performed in the temperature range 750–850 °C. The results shows a high CO + H2 concentration (>80%) in the syngas. The H2 yield decreases while the CO2
yield increases slightly with increasing temperature. The variation
of temperature in this range, has a negligible impact on the yields
of CO and CH4.
Within a range of S/B ratios 0.3–0.9, the test shows that the
hydrogen content increases while the concentration of methane
decreases as the S/B ratios increase. The total tar content decreases
with increasing S/B ratio and increasing temperature. The light tars
content decreases with increasing S/B ratio, but unlike the total

tars content the light tars content increases with increasing temperature. Most likely some of the heavy tars are cracked to light
tars as the temperature increases.
A comparison of the MIUN gasifier with other indirect gasifiers
has been carried out. The main differences in the syngas of the gasifiers can be observed in the H2 content and the CO content. The

The authors would like to acknowledge the project support
from the EU regional structure fond, Ångpanneföreningen Foundation for Research and Development, LKAB, Länsstyrelsen Västernorrland, FOKUSERA, Härnösands Kommun, Toyota, SCA BioNorr
AB and SUNTIB AB.

References
[1] Zhang W, Forsman B, Söderlind U. Biofuels in Sweden. In: 7th World congress
on recovery, recycling and re-integration, Beijing, China; September 2005.
[2] Zhang W. Transportation fuels from biomass via gasification. In: 2nd World
conference on biomass for energy, industry and climate protection, Rome,
Italy; 10–14 May 2004.
[3] Corella J, Toledo JM, Molina G. A review on dual fluidized-bed biomass
gasifiers. Ind Eng Chem Res 2007;46:6831–9.
[4] van der Meijden CM, Veringa HJ, van der Drift A, Vreugdenhil BJ. The 800 kWth
allothermal biomass gasifier MILENA. In: 16th Biomass conference &
exhibition, Valencia, Spain; 2–6 June, 2008.
[5] van der Meijden CM, Veringa HJ, Bergman PCA, van der Drift A, Vreugdenhil BJ.
Scale-up of the MILENA biomass gasification technology. In: 17th European
biomass conference and exhibition, Hamburg, Germany; 2009.
[6] Molino A, Alvino E, Freda C, Mannarino V, Petrone MT, Blasi A, et al. Simulation
of a molten carbonate fuel cell integrated into a biomass steam gasification
pilot plant. In: FP6 bigpower SSA project biogastech workshop, Gebze, Turkey;
9–11 April 2008.
[7] Sim Weeks. Rialto renewable energy center – synthetic fuels and renewable
electric power from biomass’’ presentation in biorefinery deployment
collaborative, Durham, NC; 10/2/2009.

[8] Bull D, Gilmor I, Williamson C, Pang S. The development and operation of a
100 kW dual fluidised bed biomass gasifier for production of high quality
producer gas. In: ICPS 09 (international conference on polygeneration
strategies), Vienna, Austria; 2009.


K. Göransson et al. / Fuel 90 (2011) 1340–1349
[9] Thunman H. Co-gasification of biomass in a coal fired power plant. Presented
at biomass energy workshop of IVA-CAE project on cooperation of renewable
energy and environment, Nanning, China; 2009.
[10] Xu G, Murakami T, Suda T, Matsuzaw Y, Tani H. Two-stage dual fluidized bed
gasification: its conception and application to biomass. Fuel Process Technol
2009;90:137–44.
[11] Xu G, Murakami T, Suda T, Matsuzawa Y. Reactor siphon and its control of
particle flow rate when integrated into a circulating fluidized bed. Ind Eng
Chem Res 2005;44:9347–54.
[12] Fang M, Yu C, Shi Z, Wang Q, Luo Z, Cen K. Experimental research on solid
circulation in a twin fluidized bed system. Chem Eng J 2003;94:171–8.
[13] Seemann MC, Thunman H. The new Chalmers research-gasifier. In: ICPS 09
(international conference on polygeneration strategies), Vienna, Austria; 2009.
[14] Franco C, Pinto F, Gulyurtlu I, Cabrita I. The study of reactions influencing the
biomass seam gasification process. Fuel 2003;82:835–42.
[15] Matsuda S. Measurement of solid circulating rate in a circulating fluidized bed.
Powder Technol 2008;187:200–4.
[16] Ludlow JC, Monazam ER, Shadle LJ. Improvement of continuous solid
circulation rate measurement in a cold flow circulating fluidized bed.
Powder Technol 2008;182:379–87.
[17] Göransson K, Söderlind U, Zhang W. 16th European biomass conference &
exhibition, BTL Laboratory at Mid Sweden University, Valencia, Spain; 2–6
June, 2008.

[18] Stefan Rönnquist, SCA BioNorrAB, Härnösand, Sweden, via e-mail; June 2009.
[19] Brown J. Biomass gasification: fast internal circulating fluidised bed gasifier
characterisation and comparison; 2006.
[20] Gao N, Li A, Quan C. A novel reforming method for hydrogen production from
biomass steam gasification. Bioresour Technol 2009;100:4271–7.
[21] Rutherford J, Williamson C. Integrating advanced biomass gasifiers into the
New Zealand wood industry. New Zeal J For 2006;51(3):35.

1349

[22] Hofbauer H, Rauch R. Stoichiometric water consumption of steam gasification
by the FICFB-gasification process. Institute of Chemical Engineering, Fuel and
Environmental Technology, Vienna, Austria.
[23] Pfeifer C. Experiences and results derived from the dual fluidised bed gasifier
at Vienna University of Technology. In: International conference on polygeneration strategies, Vienna, Austria; 2009.
[24] Hofbauer H. Biomass for synthesis gas – overview. In: 17th European biomass
conference and exhibition, Hamburg, Germany; 2009.
[25] Corella J, Toledo JM, Molina G. Biomass gasification with pure steam in
fluidised bed: 12 variables that affect the effectiveness of the biomass gasifier.
Int J Oil Gas Coal Technol 2008;1(1/2).
[26] Rapagnà S, Jand N, Kiennemann A, Foscolo PU. Steam-gasification of
biomass in a fluidised-bed of olivine particles. Biomass Bioenergy 2000;
19(3):187–97.
[27] MILENA gasification process. ECN Energy research Centre of the Netherlands.
< />[28] van der Meijden CM, van der Drift A, Vreugdenhil BJ. Experimental results
from the allothermal biomass gasifier MILENA. In: 15th European biomass
conference, Berlin, Germany; 2007.
[29] Thunman H. GoBiGas and other gasification activities at Chalmers,
presentation.
<denius.fi/project_files/HighBioProjektiseminaari%20180510/Henrik%20Thunman%201805-2010.pdf>

(18.05.10).
[30] Martin Seemann. Chalmers University of Technology in Gothenburg, Sweden,
via e-mail; August 2009.
[31] Doherty W, Reynolds A, Kennedy D. Modelling and simulation of a biomass
gasification-solid oxide fuel cell combined heat and power plant using Aspen
Plus, Dublin Institute of Technology; 2009.