Reactivity tests of the wateregas shift reaction on fresh and used fluidized bed materials from industrial DFB biomass gasifiers

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

Available online at www.sciencedirect.com

/>
Reactivity tests of the wateregas shift reaction on
fresh and used fluidized bed materials from
industrial DFB biomass gasifiers
Stefan Kern*, Christoph Pfeifer, Hermann Hofbauer
Vienna University of Technology, Institute of Chemical Engineering, Getreidemarkt 9/166, 1060 Vienna, Austria

article info

abstract

Article history:

The dual fluidized bed gasification process, offers various advantages for biomass gasifi-

Received 29 May 2012

cation as well as the utilization of other solid feedstocks. In order to improve the knowl-

Received in revised form

edge of the reactions in fluidized bed gasifier, different types of bed material used in the

28 January 2013

gasifier were tested in a micro-reactivity test rig. It has been previously observed that

Accepted 4 February 2013

during long-term operation, the surface of the bed material used (calcined olivine) un-

Available online 26 February 2013

dergoes a modification that improves catalytic activity. The main reaction of interest is the
wateregas shift reaction. Olivine taken from long-term operation at the 8 MW biomass

Wateregas shift

gasifier at Gu¨ssing (Austria), fresh olivine as a reference, and calcite, which is commonly
used for enhancing in-bed catalytic tar reduction, were tested using the micro-reactivity

Micro reactor

test rig. Tests were carried out at temperatures of 800, 850, and 900  C and space veloc-

Gasification

ities of 40,000 to 50,000 hÀ1 were applied. CO conversions of up to 61.5% were achieved for

Biomass

calcite. Used olivine showed a similar behavior, representing a large improvement

Olivine

compared to fresh olivine, which had CO conversion rates of less than 20%.

Keywords:

ª 2013 Elsevier Ltd. All rights reserved.

Calcite

1.

Introduction

The worldwide demand for energy is constantly growing and
the larger part of this growth is currently covered by oil and
gas. Developing countries in particular, exhibit high rates of
growth, connected to their expanding economies [1]. Consequently, finding new, more effective and wide-ranging applications for low grade and cheap biogenic and fossil fuels is
essential. Gasification of biomass or solid fuels multiplies the
field of its conventional application.
The dual fluidized bed steam gasification system (DFB) is a
key technology for the production of a high calorific product as
pure steam is used as gasification agent [2,3]. The heat for the
allothermal gasification process is provided by a separate

combustion reactor. The DFB process has already proven its
reliability at industrial sized plants [4].
The utilization of the syngas produced by this process is not
limited to heat and power production in a gas engine or a boiler.
There is also a huge potential for the production of liquid or
gaseous fuels from syngas [5] by FischereTropsch synthesis,
mixed alcohols synthesis or the production of synthetic natural gas. These processes can use syngas made by the gasification of solid feedstock as a source, but each process requires
a particular syngas composition, in terms of its H2:CO ratio, to
sustain optimal operation. However, the composition of the

syngas, especially the H2:CO ratio, should be as close as
possible to the required ratio to maximize product yield and
process performance. The possibilities for changing syngas

* Corresponding author. Tel.: þ43 1 58801 166382; fax: þ43 1 58801 16699.
E-mail addresses: , (S. Kern).
0961-9534/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved.
/>

228

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

compositions are limited. On the one hand, the process parameters, such as gasification temperature or steam-to-fuel
ratio, can only be varied within a limited range, and the net
effect on the gas composition is marginal [6]. The more effective way is the utilization of a catalytic active bed material in
the gasifier, which promotes reactions that force hydrogen
production, since the H2:CO ratio is normally too low.
The main gasification reactions are shown in Table 1. These
reactions are considered as equilibrium reactions, with variable states of equilibrium depending on gas concentrations,
temperature and pressure. However, in a real fluidized bed
gasifier a complete state of equilibrium will not be reached.
In the gasification reactor, these reactions can take place at
the same time and location, and some reactions can be forced
by operating parameters and by the utilization of catalytic bed
material. In the pilot and industrial DFB plants, the bed material used is calcined olivine [7,8], since it shows significant
tar reduction compared to silica sand [6] and is perceived as a
natural catalyst. Kirnbauer and Hofbauer [9] reported modification of the olivine during long-term operation in industrial
plants as two different calcium-rich layers are built around
the olivine particle. The inner layer consists mainly of calcium

silicates, whereas the outer layer, which is in contact with the
gas, char and fuel particles, has a similar composition to the
ash from the feedstock in combination with calcium-rich
additives like calcite.
In a previous study [10], used olivine from the 8 MW
demonstration plant at Gu¨ssing was employed in the 100 kW
pilot plant at the Vienna University of Technology. The tar
contents measured in the used olivine were five times lower
than those of fresh olivine. Moreover, a significantly lower
energy demand for gasification was observed during utilization of the used olivine particles in the gasifier. As also a
higher amount of H2 and CO2 was formed while a lower
amount of CO was found in the product gas, it leads to the
assumption that the enhancement of the slightly exothermic
wateregas shift reaction was the main reason for this effect.
Consequently, the influence of the solids in the gasifier on
the wateregas shift reaction has been studied in a microreactivity test rig in this study. In the test rig reactor, a stoichiometric mixture of carbon monoxide and steam was
passed through a fixed bed of the particles to be tested at
temperatures of 800, 850 and 900  C. These temperatures are

comparatively high [12], as the wateregas shift reaction is
favored at low temperatures (Figs. 1 and 2), but gasification
conditions are usually set in this high temperature range,
especially for fluidized bed gasifiers. In wateregas shift reactors the common commercial catalysts are iron oxide based
and copper based [13], whereas in gasifiers a natural, inexpensive bed material is favored. The thermodynamic equilibrium composition of the wateregas shift reaction is plotted
in Fig. 1. The CO conversion for equilibrium conditions and
stoichiometric input of CO and H2O is shown in Fig. 2.
Thermodynamic equilibrium of a reaction is reached when
Gibb’s energy function (G) achieves a minimum.
G¼

XK

n $DG0i
i¼1 i

XK
XK
þ RT i¼1 ni $ln yi þ RT i¼1 ni $ln P

is the Gibb’s free energy for species i at
In Equation (1),
standard conditions, ni is the number of moles of species i, yi is
the mole fraction of species i, R is the universal gas constant, T
is the absolute temperature and P is the pressure. Moe [15]
indicated the equilibrium for the wateregas shift reaction
with a suitable approach, as shown in Equation (2).


4577:8
À 4:33
Keq ¼ exp
T

Wateregas (i)
Wateregas (ii)
Boudouard
Methanation
Oxidation (i)
Oxidation (ii)
Wateregas

shift
Methane
reforming

Chemical equation

DHR,850
(kJ molÀ1)

Equation

C þ H2O 4 CO þ H2
C þ 2H2O 4 CO2 þ 2H2
C þ CO2 4 2CO
C þ 2H2 4 CH4
C þ O2 4 CO2
C þ 0.5O2 4 CO
CO þ H2O 4 CO2 þ H2

þ135.7
þ102.1
þ169.4
À89.8
À394.9
À112.7
À33.6

(1)
(2)
(3)

(4)
(5)
(6)
(7)

CH4 þ H2O 4 CO þ 3H2

þ225.5

(8)

(2)

The tests documented in this paper give an insight into the
capability of inorganic bed material particles, fresh olivine,
used olivine and calcite, in the dual fluidized bed steam
gasifier for the promotion of the wateregas shift reaction and
will be the basis of further research concerning gasesolid
contact in dual fluidized bed gasifiers.

2.

Materials and methods

2.1.

Micro-reactivity test rig

The flow sheet of the micro-reactivity test rig used in these
investigations is shown in Fig. 3. The heart of the testing rig is

a glass reactor where the catalyst can be placed, in this case
the different types of olivine or the calcite. The reactor has an
inner diameter of 10 mm and is electrically heated. Thus, the
operating temperature of the reactor can be controlled, up to
900  C. Gas mixtures from up to six different sources can be

Table 1 e Equilibrium reactions in biomass gasification
[11].
Name of
reaction

(1)

G0i

Fig. 1 e Thermodynamic equilibrium composition of the
wateregas shift reaction [14].


b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

2.3.

Fig. 2 e CO conversion for equilibrium conditions [14].

dosed and mixed by thermal mass flow controllers. A steam
generator with a constant flow device is also installed for
adding steam. After mixing of the gaseous species at the inlet
of the reactor, the gas mixture passes through the catalytic
active material in the heated zone and leaves the reactor

through a Liebig cooler, for water condensation, and into the
gas analyzer.

2.2.

Analytics

The measurement device for determining the permanent gas
components at the reactor outlet is an extractive gas analyzer
(model NGA2000) made by Rosemount.

229

Particle characterization

The materials tested in this experimental campaign are fresh
olivine, used olivine from the Gu¨ssing CHP, and calcite. Table 2
shows characteristic values for the particle sizes of the
materials used in this investigation. The results of the X-ray
fluorescence analyses (XRF) of these materials are given in
Table 3.
As can be seen in Table 3, the main differences between
fresh olivine and used olivine are significantly increased
contents of calcium and potassium in the latter. A detailed
view on the structure of the olivine particles is provided by
Figs. 4 and 5. In Fig. 5, the outer layer of the particle, which is
created during long-term operation in the gasifier, is visible.
A more comprehensive study is given by Kirnbauer and Hofbauer [9] where also the results of the EDX (energy dispersive
x-ray spectroscopy) analysis of the layers are listed. There can
be found that the calcium content increases massively in the

layers compared to the particle inside.

3.

Results and discussion

The influence of the used materials on the wateregas shift
reaction was investigated for each material at temperatures of
800, 850, and 900  C. Apart from the temperature, each test was
undertaken using the same operating conditions. To provide
stoichiometric conditions, an equimolar amount of carbon
monoxide (CO) and steam (H2O) was fed into the reactor. This
ratio of CO to H2O was chosen as the focus in this work was only
the wateregas shift reaction. In the industrial sized DFB

Fig. 3 e Flow sheet of the micro-reactivity test rig.


230

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

Table 2 e Characteristic values for the particle sizes of the
used materials.
Unit
dp10
dp50
dp90

mm

mm
mm

Fresh olivine

Used olivine

Calcite

184
439
694

196
456
716

350
860
978

gasification plants as located in Gu¨ssing, there is no equimolar
ratio of the educts for the wateregas shift reaction present as
there the steam-to-carbon ratio varies between 1.2 and
1.6 kg kgÀ1 at a fuel input power of 8000 kW which yields a dry
volume fraction of CO in the product gas of 20.1% [10]. For the
tests here, the chosen volumetric flow rate of CO was
30 dm3 hÀ1 (all gas volumes measured at standard conditions of
101.3 kPa and 273.15 K), which results in a corresponding mass
flow rate of H2O of 24.1 ghÀ1. The desired carrier flow stream for

the steam generator was nitrogen (N2), with a volumetric flow
rate of 3$10À2 m3 hÀ1. These input flow rates supply an input
concentration of CO of 50% based on dry gas volume (CO and
N2) into the reactor. The temperature was increased from
800  C to 900  C in 50  C steps. For each test about 10 g of particles were used, which represented a fixed bed height of between 8 and 10 cm. This inventory resulted in a space velocity
of 40,000 to 50,000 hÀ1 during the experiments. The particles
were present in the glass reactor as fixed bed and the gas flow
was downward. A test run with an empty reactor at each
operating point was carried out for reference reasons.
The most important values in such tests are obviously the
outlet gas composition, the CO conversion and the hydrogen
selectivity. Fig. 6 shows the gas composition at the reactor
outlet following the empty reactor run for each tested temperature. It can be seen that only a very small amount of the
initial CO and H2O was converted to CO2 and H2. Nevertheless,
the H2 content rose from 0.90 to 1.58% (based on dry gas volume) as the temperature in the reactor increased. After filling
the reactor with 10 g of fresh olivine, the wateregas shift reaction was obviously enhanced, compared to the empty
reactor. With these particles, H2 production yielded a dry
volume concentration of 3.6% at 800  C, increasing to 5.7% at
900  C (Fig. 7).

Fig. 4 e SEM image of fresh olivine [9].

The assumption that the calcium and potassium-rich layer
around the used olivine particles forces the wateregas shift
reaction is supported by Fig. 8, which demonstrates that H2
production was much higher, compared to fresh olivine,
where no such layer is available. The H2 concentration in the
gas after the reactor was more than four times higher than
with fresh olivine. Especially at lower temperatures, this gap
increased. The results for calcite are shown in Fig. 9. This

material was tested for comparison reasons, since it is typically used in industrial plants as a tar reducing additive. Given
the high CaO content, the conversion of CO and H2O to CO2
and H2 reached the highest measured values. The dry based H2
content ranged between 21.7% at 800  C and 22.26% at 900  C.
The complete (wet) composition of the gas streams after the
reactor are summarized in Table 4. It has to be kept in mind
that, as mentioned before, nitrogen was used as a carrier gas,
representing half of the volume of the dry gas flow into the
reactor. CH4 was also detected by the gas analyzer but in
contrast to conventionally used Fe-based catalysts that are
operated at stoichiometric CO to H2O values, there was no CH4
(<0.01%) found in the gas at the reactor outlet. This leads to
the assumption that also the formation of solid carbon, that

Table 3 e Used materials mass fraction composition %.
Metal oxide
Na2O
MgO
Al2O3
SiO2
P2O5
SO3
K2O
CaO
Cr2O3
MnO
Fe2O3
NiO
Cl
others

Fresh olivine

Used olivine

Calcite

0.43
46.76
0.40
39.84
0.03
0.06
0.32
0.90
0.28
0.15
10.32
0.31
0.10
0.11

1.67
40.52
0.44
33.60
0.19
0.08
3.89
10.71

0.38
0.24
7.45
0.33
0.27
0.23

0.94
0.73
0.42
1.38
0.03
0.06
0.11
95.44
0.05
0.00
0.44
0.03
0.22
0.15

Fig. 5 e SEM image of used olivine (after long-term
gasification operation in the Gu¨ssing DFB plant) [9].


231

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

Fig. 6 e Outlet gas composition, empty reactor.

Fig. 9 e Outlet gas composition, reactor filled with calcite.

the bed material and the H2 selectivity (SH2 ). The activity of the
bed material is shown as CO conversion (XCO) in Equation (3).
XCO ¼

½COŠin À ½COŠout
$100%
½COŠin

(3)

Hydrogen selectivity puts the actual H2 yield into perspective, compared with the theoretical maximum H2 yield, based
on thermodynamic equilibrium [14] at the corresponding
temperature and CO/H2O ratio, as Equation (4) shows.
SH2 ¼

Fig. 7 e Outlet gas composition, reactor filled with fresh
olivine.

can often be found at such tests, was not observed here. After
the tests no carbon was found. This can be caused by the
relatively high temperatures for wateregas shift catalysis at
the tests done here.
An appropriate way to demonstrate and compare the performance of the used bed material is to examine the activity of

½H2 Šyield
½H2 Šmax

$100%

(4)

The CO conversion (XCO) is shown in Fig. 10 and the H2
selectivity (SH2 ) of the tested materials is plotted in Fig. 11. In
Fig. 10, two areas of CO conversion can be seen at values that
range from 2.1% for olivine at 800  C to 61.5% for calcite at
900  C. Calcite and used olivine are much more active with
regard to CO conversion than fresh olivine and, obviously, the
empty reactor. However, the trend lines for fresh and used
olivine are similar at different conversion levels. Their catalytic activity increases considerably between 800 and 850  C.
The CO conversions of calcite, and in the empty reactor, result
in straight lines that, in comparison to olivine, rise only
slightly with increasing temperature.

Table 4 e Complete gas characterization after reactor, wet
gas volume fraction.
Material

Temperature

C

CO
(%)

CO2
(%)

H2
(%)

N2
(%)

H2O
(%)

Empty
reactor

800
850
900
800
850
900
800
850
900
800
850
900

31.97
31.63
31.52
38.98

28.97
28.03
24.92
21.47
21.18
17.45
17.40
16.93

0.51
0.85
0.93
3.25
3.17
4.65
11.52
15.74
16.52
17.82
17.88
18.34

0.60
0.97
1.06
2.85
3.15
4.10
12.30
16.86

17.69
19.06
19.11
19.70

33.41
33.40
33.39
35.07
33.39
34.84
33.55
33.59
33.65
33.55
33.56
33.52

33.51
33.15
33.11
19.85
31.33
28.38
17.71
12.35
10.95
12.11
12.04
11.51

Fresh
olivine
Used
olivine
Calcite

Fig. 8 e Outlet gas composition, reactor filled with used
olivine.


232

b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

Fig. 10 e CO conversion.

the combination with the high potassium content is assumed
to be responsible for catalyzing the desired reaction. Mudge
et al. [17] summarized the order of reactivity for alkaline
metals suitable for catalyzing biomass gasification and
reforming reactions where potassium carbonate to be the
most reactive one. The major catalytic effect of fresh olivine is
dominated by magnesium and Fe2O3 in the particles. Delgado
et al. [18] determined ranks of activity for mainly reforming
reactions at gasification in which dolomite was the most
active followed by magnesite and calcite. CaO is, in addition to
MgO, one of the major components of dolomite, SiO2, Fe2O3
and Al2O3 [19].
For the tests done here, the CaO content of calcite is more

than nine-times higher than that of used olivine. Nevertheless, the catalytic activity of used olivine almost achieves the
performance of calcite. This could be caused by the combination of calcium and potassium, as potassium is also
enriched to a level 10-times higher than in fresh olivine. This
is an extremely promising result, since the modification of
fresh olivine to olivine coated with a layer rich in calcium and
potassium automatically occurs during operation in the
gasifier, with the composition of the particle surface determined by the composition of the ash of the fuel used for
gasification.

Acknowledgments

Fig. 11 e Hydrogen selectivity.

This study was carried out within the framework of the
Fecundus project, funded by the Research Fund for Coal and
Steel of the European Union (CONTRACT N RFCR-CT-2010
kova´ from VSB00009). Thanks are given to Ms Iva Macha´c
TU Ostrava, Czech Republic, for her great help for the experimental part of this work.

references
Hydrogen selectivity exhibited similar behavior and
reached values of up to 95.3% for calcite at 900  C. The highest
H2 selectivity for fresh olivine was at 900  C, corresponding to
a value of only 24.5%. In this case, used olivine again showed
better performance, since a selectivity of 85.1% was detected
at 900  C, which is already as high as for calcite at 800  C.

4.

Conclusion

The wateregas shift reaction is an important reaction occurring in combination with gasesolid contact in fluidized bed
gasification reactors with the bed materials present in the
reactor. At high temperatures, the results showed that calcium is an appropriate catalyst for enhancing the reaction.
Calcium has been identified as playing a major role in the
promotion of wateregas shift reactions if present as CaO.
Clemens et al. [16] found that for gasification with steam, the
Ca content of coal ash both promotes the wateregas shift
equilibrium and influences the composition of the product
gas. For the used materials here, the Ca content plays the
major role, especially for calcite. In the case of the used olivine

[1] IEA. World energy outlook 2010. Paris: International Energy
Agency; 2010.
[2] Kern S, Pfeifer C, Hofbauer H. Gasification of wood in a dual
fluidized bed gasifier: Influence of fuel feeding on process
performance. Chem Eng Sci 2013. />j.ces.2012.12.044i.
[3] Go¨ransson K, So¨derlind U, He J, Zhang W. Review of syngas
production via biomass DFBGs. Renew Sust Energ Rev 2011;
15:482e92.
[4] Hofbauer H, Rauch R, Bosch K, Koch R, Aichernig C. Biomass
CHP plant Gu¨ssing e a success story. In: Bridgewater AV,
editor. Pyrolysis and gasification of biomass and waste.
Newbury, Berks., UK: CPL Press; 2003. p. 527e36.
[5] Demirbas A. Progress and recent trends in biofuels. Prog
Energ Combust 2007;33(1):1e18.
[6] Koppatz S, Pfeifer C, Hofbauer H. Comparison of the
performance behaviour of silica sand and olivine in a dual
fluidised bed reactor system for steam gasification of
biomass at pilot plant scale. Chem Eng J 2011;175:468e83.


 ski D,
[7] Rauch R, Pfeifer C, Bosch K, Hofbauer H, Swierczy
n
Courson C, et al. Comparison of different olivines for
biomass steam gasification. In: Bridgewater AV,
Coocock DGB, editors. Science in thermal and chemical


b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 2 7 e2 3 3

[8]

[9]

[10]

[11]
[12]

biomass conversion. August 30th e September 2nd 2004.
Victoria, Canada. UK: CPL Press; 2006. p. 799e809.
Devi L, Craje M, Thu¨ne P, Ptasinski KJ, Janssen FJJG. Olivine as
tar removal catalyst for biomass gasifiers: catalyst
characterization. Appl Catal A 2005;294(1):68e79.
Kirnbauer F, Hofbauer H. Investigations on bed material
changes in a dual fluidized bed steam gasification plant in
Gu¨ssing, Austria. Energy Fuels 2011;25:3793e8.
Kirnbauer F, Wilk V, Kitzler H, Kern S, Hofbauer H.
The positive effects of bed material coating on tar

reduction in a dual fluidized bed gasifier. Fuel 2012;95:
553e62.
Kaltschmitt M, Hartmann H, Hofbauer H. Energie aus
Biomasse. Berlin, Heidelberg: Springer; 2009.
Haryanto A, Fernando SD, To SDF, Steele PH, Pordesimo L.
Hydrogen production through the wateregas shift
reaction: thermodynamic equilibrium versus experimental
results over supported Ni catalysts. Energy Fuels 2009;23:
3097e102.

233

[13] Mukherjee S, Hatalis MK, Kothare MV. Water gas shift reaction
in a glass microreactor. Catal Today 2007;120:107e20.
[14] HSC. Hsc chemistry 5.1. Pori, Finland: Outokumpu Research
Oy; 2002.
[15] Moe JM. Design of wateregas shift reactors. Chem Eng Prog
1962;58:33e7.
[16] Clemens AH, Damiano LF, Matheson TW. The effect of
calcium on the rate and products of steam gasification of
char from low rank coal. Fuel 1998;77:1017e20.
[17] Mudge LK, Baker EG, Mitchell DH, Brown MD. Catalytic steam
gasification of biomass for methanol and methane
production. J Solar Energy Eng 1985;107:89.
[18] Delgado J, Aznar MP, Corella J. Biomass gasification with
steam in fluidized bed: effectiveness of CaO, MgO, and
CaOÀMgO for hot raw gas cleaning. Ind Eng Chem Res 1997;
36:1535e43.
[19] Sutton D, Kelleher B, Ross JRH. Review of literature on
catalysts for biomass gasification. Fuel Process Technol 2001;

73:155e73.