Hydrogen-rich Gas Production with a Ni-catalyst

in a Dual Fluidized Bed Biomass Gasifier

C. Pfeifera, R.Raucha, H.Hofbauera, D.wierczyskib, C.Coursonb,
A.Kiennemannb
a
Institute of Chemical Engineering, Vienna University of Technology,
Getreidemarkt 9/166, A-1060 Vienna, Austria
b
LMSPC-ECPM, UMR 7515 - 25, rue Becquerel - 67087 Strasbourg
Cedex 2 France

ABSTRACT: To improve the gas quality (low tar content and high hydrogen volume
fraction) of biomass steam gasification in a dual fluidized bed gasifier a nickel-based
catalyst has been developed. This catalyst was tested as bed material in a 100 kWth
pilot plant. Gas composition and tar content were measured after the reactor and
compared with data from gasification tests without a catalytic bed material. Tar and
ammonia reduction, changes in the gas composition as well as water conversion, gas
yield, chemical efficiencies and lower heating values are reported. The catalysts were
characterized before and after the experiments. The catalyst shows strong metal-
support interactions which of nickel on the surface which provide resistance against
attrition. Furthermore the catalyst is highly resistant and stable for repeated high
temperature processing in oxidizing and reducing atmosphere during circulation
between the fluidized beds. The performance of the catalyst over 30 h showed no loss
of reactivity for cracking tars and no deactivation by common catalyst poisons (sulfur,
chlorine, and alkali metals). The presented data represent important results for the
further development of catalytic gas cleaning in fluidized bed gasification processes.

INTRODUCTION

Biomass as energy source is required by climatic conventions and in the European

Union White Paper. Electricity generation is possible in new applications via biomass
gasification. This technology has a big potential due to its great flexibility and the high
electrical as well as the overall efficiency. A dual fluidized bed steam gasifier has been
developed during the last decade1. The basic idea of this concept is to divide the
reactor in two zones, a gasification zone fluidized with steam and a combustion zone
fluidized with air. Between these two zones a circulation loop of the bed material is
created but the flue gas remains separated from the product gas. This results in a nearly
nitrogen-free, hydrogen-rich product gas. With the aid of a catalytic active bed material
the hydrogen content can even be increased.
The key problem of gasification technologies is the tar removal from the product gas.
The use of an active bed material as a primary catalyst is the best solution in contrast to
more expensive use of secondary catalytic reactor downstream of the gasifier. The
choice of an appropriate in-bed catalyst is then crucial for the optimization of
gasification technology.
Natural olivine (iron and magnesium orthosilicate) has been chosen as bed material
because of its hardness and its higher catalytic activity compared to silica in biomass
steam gasification2. Ni-olivine catalyst has been developed3,4,5,6 to enhance olivine
performances in steam biomass gasification by methane and tar reforming leading to
hydrogen production and gas upgrading. The main advantage of this catalyst is its
attrition resistance due to strong metal support interactions permitting its direct use in
the fluidized bed.
The main limitation of nickel catalyst use for hot gas conditioning of biomass
gasification product gases is its sensitivity to deactivation, which leads to limited
catalyst lifetimes. Ni catalyst deactivation is caused by several reasons. Sulfur,
chlorine, and alkali metals present in gasification product gases act as catalyst poisons.
Coke formation on the catalyst surface can be substantial when tar levels in product
gases are high. Loss of catalyst activity is apparently due to fouling by buildup of
carbon which blocks access to the catalyst pores7. Coke can be removed by
regenerating the catalyst; however, repeated high temperature processing of nickel
catalysts can lead to sintering, phase transformations, and volatilization of nickel.
Continued disposal of spent toxic Ni catalysts is not economical and poses an
environmental hazard because of the toxicity of nickel. The catalyst can be regenerated
by transporting the catalyst to the combustion zone and burning off the carbon.1. A
critical gap identified for catalytic tar reforming technology in biomass gasification
processes is then the need for extended lifetime studies of promising commercial or
novel catalysts8.
There is a need for understanding of catalyst evolution during gasification in real
conditions in the dual fluidized bed steam gasifier. Therefore Ni-olivine catalyst as bed
material was tested in the fluidized bed followed by the characterization of the used
catalyst. Similarly gasification with olivine alone, followed by its characterization
serves as a reference for the Ni-olivine catalyst.
The aim of the present work was a comparison of activity and selectivity of olivine and
Ni-olivine catalysts and the evaluation of the catalyst stability. The effects of catalyst
poisons like sulfur, chlorine, alkali metals and continued catalyst regeneration
(circulation between gasification (850C) and combustion zones (900C)) in fluidized
bed biomass gasification conditions during the extended-term test (30h) are especially
studied.

CATALYSTS
Natural olivine

Natural olivine originates from an Austrian mine and its composition (30.5 wt.%
of Mg, 7.1 wt.% of Fe and 19.6 wt.% of Si), obtained by atomic adsorption, leads to
the mean formula (Mg0.95Fe0.05 )2SiO4 + 0.04 Fe2O3. Iron excess in the form of free iron
oxide corresponds to ~4 wt.%. The material already contains small amounts of nickel
as well as Ca, Al and Cr (lower than 0.2 wt.% of each). The material was delivered
after calcination at 1600C during 4 hours, crushing and sieving to obtain the particle
sizes between 400 and 600 m. The specific surface area of olivine is very low
(<1m2/g). Some interesting physical properties of olivine are given in Table 1.
 Bulk density 1700-1900 kg/m
Material density 3025 kg/m
Hardness 6.5-7 Mohs scale
Melting point approx. 1760 C
pH 8.9-9.5
Particle size 400-600 m
BET-surface area <1 m/g
Table 1 Physical properties of olivine

Ni-olivine catalyst

Large amount of Ni-olivine (100 kg) was prepared by wet impregnation of natural
olivine with an excess of nickel nitrate solution. A glass lined steel rotary dryer was
used for the preparation. Nickel nitrate (Ni(NO3)2,6H2O) was dissolved by addition of
water and salt into the reactor. Natural olivine was added after the nitrate dissolution.
The sample was mixed by the rotation of the reactor and then heated to 100C. Then,
the water excess was evaporated under vacuum. The sample was dried overnight in an
oven at 120 C and calcined at 1100 C during four hours after a temperature increase
slope of 3 C.min-1. Characterization of representative samples of the catalyst by
elemental analysis, TPR and XRD showed good homogeneity of the large scale
preparation. The nominal percentage of Ni was equal to 3.7 wt %.
The foreseen price for this Ni-olivine, when produced in large scale for the use at the
demonstration plant in Gssing, is about twice of the price for olivine.

CHARACTERIZATION METHODS

The crystalline phases contained in the samples and the structural modifications
occurring after gasification test were examined by powder X-ray diffraction (XRD) on
a Siemens D500TT diffractometer using Cu K radiation.
Scanning electron microscopy (SEM) was performed on a JEOL 6700 F microscope.
The surface composition has been obtained by X-ray microanalysis. The values given
are mean values from at least 4 analysies taken on different grains (with a window of
200 m) and correspond to about 1 micron depth of the surface.
The active phase for hydrocarbon reforming is the metallic nickel. So, the reducibility
of the catalysts under hydrogen gives valuable information. This has been followed by
temperature programmed reduction (TPR) performed on 200 mg of catalyst placed in a
U-shaped quartz tube (6.6mm ID). The reductive gas mixture (H2 = 0.12 L.h-1 and
Ar = 3 L.h-1) passed through the reactor heated from room temperature to 950C with a
slope of 15C.min-1 then maintained at 950C until the end of H2 consumption showed
by the baseline return. A thermal conductivity detector (TCD) was used for
quantitative determination of hydrogen consumption.
The 57Fe Mssbauer spectra were recorded at liquid nitrogen temperature with 57Co
source dispersed in rhodium matrix. From the obtained spectra, the isomeric shifts
were determined in comparison with metallic iron standard at room temperature. To
identify the different forms of iron present in the sample, the spectra were fitted with a
computer program MossFit.

DESCRIPTION OF THE PILOT PLANT

Fig. 1 shows a simplified flow sheet of the 100 kWth pilot plant used in this study. The
system is a dual fluidized bed steam gasifier. The gasifier is divided into two zones, a
gasification zone and a combustion zone. The gasification zone has at the bottom of the
bed an inner diameter of 100 mm. The fluidised bed is expanded to a square with
285 mm side length and has a total height of about 1 m. The gasification zone is a
bubbling fluidised bed whereas the fluidisation velocity about five times higher than
the minimum fluidisation velocity. Biomass is fed into the gasification zone and
gasified with steam. The gas produced in this zone is therefore nearly free of nitrogen
and has a high hydrogen content. The bed material, together with some charcoal and
ungasified biomass, circulates to the combustion zone. This zone is fluidized with air
and the charcoal is burned. Light fuel oil as additional fuel is used to raise the
temperature of the bed material leaving the combustion zone to the desired
temperature. In this way the gasification temperature is controlled. The exothermic
reaction in the combustion zone provides the energy needed for the endothermic
gasification with steam. The flue gas is removed from the bed material without coming
in contact with the product gas. To accomplish this, the gasification zone is separated
from the combustion zone by the use of two siphons which are fluidized with steam too
(Fig. 1).
The biomass is fed into the gasification zone by two screws. The first one is equipped
with a frequency converter to control the amount of fuel. To keep the biomass in close
contact with the hot bed material, the biomass is brought in by the second screw
directly into the fluidized bed. This is important especially in the case of catalytically
active bed materials.

Fig. 1 Simplified flow sheet of the pilot plant

parameter value units

 C 49 mass% wf
H 6.52 mass% wf
N 0.12 mass% wf
S < 0.05 mass% wf
O 44.31 mass% wf
Ash 0.26 mass% wf
LHV 17120 kJ/kg
HHV 18620 kJ/kg
Table 2 Results of the fuel analysis
Wood pellets with a mass flow of 25 kgwet/h were used as feedstock for all
experiments. The water content of the wood pellets was 9 wt%. The water-free (wf)
composition of the fuel is reported in Table 2.

Tar Sampling and Analysis

The dust and tar contents as well as the gas composition are measured after the product
gas heat exchanger. The tar measurement is carried out similarly to the tar protocol
gravimetric tars9. A small amount of product gas is sampled isokinetically for a
certain period of time. Dust and heavy tars (TB>200 C) are deposited in a filter
cartridge filled with glass wool. Low-boiling tars (TB<200 C) are washed out with
toluene using washing flasks, which are operated at -20 C. The tars in the filter
cartridge are washed out with isopropanol using a soxhlet extractor and the tars are
added to the amount of tars of the toluene phase. Samples of the toluene from the
washing flasks as well as of the isopropanol are taken for the characterization of the
tars using a gas-phase chromatograph coupled with a mass spectrometer (GC-MS).
Along with the tar content, the particle, coke and moisture contents in the product gas
are measured. A gas meter and a thermocouple are used to determine the dry product
gas stream.
The gas composition of the product gas is analyzed with an offline gas-phase
chromatograph as well as with an online gas-phase chromatograph both coupled with a
thermal conductivity detector (TCD) to get reliable figures.
The experiments were carried out at 850 C (100 % Olivine) and at 838 C (100 % Ni-
olivine) and a steam-to-fuel ratio of 0.63. Both tests were carried out for about 30
hours under gasification conditions. The gasification temperature is defined as the
temperature of the fluidized bed at the height of the biomass feeding point and the
steam-to-fuel ratio is defined as the total mass flow of water in the gasifier (fluidization
agent and biomass moisture) divided by the dry biomass mass flow.

EXPERIMENTAL RESULTS

RESULTS FROM THE PILOT PLANT EXPERIMENTS

Chemical engineering results

The usage of Ni-catalysts in combination with biomass gasification processes

decreases the tar content of the product gas and increases the hydrogen volumetric
content10,11. The methane content decreases because of steam reforming which is a
heavily endothermic reaction12. For this reason the gasification temperature could not
be raised without changing the settings for the operation of the pilot plant.
The main results of the experiments are shown in Table 3. Previous publications
describe different mixtures of olivine and Ni-olivine13, 14 whereas the matter in hand
shows a comparison of only olivine resp. Ni-olivine. Because of some changes in the
gasifier design the experiments with different olivine-catalyst mixtures are not
comparable with those described here.

olivine Ni-olivine
bed temperatue, C 850 838
steam-to-fuel ratio, kgH2O/kgdry fuel 0.63 0.63
dry gas composition, vol%
H2 38.9 43.9
CO 29.1 27.2
CO2 17.5 18.8
CH4 11.4 8.3
C2H4 2.0 1.3
lower heating value of the product
13.8 12.4
gas, MJ/Nm
gas production, Nm/kg* 0.95 0.99
water conversion, % 4.4 7.2
*
Volume of dry gas per kg dry wood
Table 3 Results with olivine and with nickel catalyst

The water conversion is the difference of the total mass flow of water in and the total
mass flow of water out divided by the dry biomass mass flow. The water conversion is
an important value used to estimate the efficiency of steam gasification systems. As
expected, the water conversion increases when using a bed material with a higher
catalytic potential. But as can be seen in Table 3 the water conversion is quite low (4.4
resp. 7.2 %) which means that the main part of the steam leaves the gasifier unreacted.
Nevertheless the conversion rates are increased parallel with the excess of water found
in previous work13, 14. An optimum has to be found between the wanted gas quality and
the heat loss due to unreacted steam. From results of previous work a steam-to-fuel
ratio of about 0.6 represents a procedural optimum.

An important parameter for gasification processes is the chemical efficiency which is

defined according the following equation:

LHV pg *V&pg
chem = [%]
LHV fuel * m& fuel + LHVadd _ fuel * m& add _ fuel

The chemical efficiency varies not significant comparing olivine and Ni-olivine as bed
material and lies in the same range (about 70%) as resulted at the demonstration plant
in Gssing15. The chemical efficiency is heavily dependent on the moisture of the
biomass feed and the heat loss over the reactor.

Fig. 2 shows the gas composition using Ni-olivine as bed material over the
experimental time of about 30 hours. It can be observed that the catalyst has initially a
high reactivity. The methane volumetric content could be lowered from 11.4 compared
to natural olivine to 7.2 vol%. After reaching the so called steady state reactivity the
methane fraction increased up to 8.3 vol% (see Table 3). The catalyst reached the
steady state reactivity after about 20 hours under gasification conditions. The decrease
of reactivity at the beginning is mainly due to attrition of nickel loosely deposited on
the olivine surface. This point is discussed in details in the characterization chapter.

50.0
14.0
45.0

40.0 12.0
concentration [vol% dry gas]

35.0
10.0
H2

LHV [MJ/Nm]
30.0
8.0
CO
25.0 CO2
CH4
20.0 6.0
C2H4
15.0 LHV
4.0
10.0
2.0
5.0

0.0 0.0
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00
time [h]

Fig. 2 Product gas composition over time (gasification

temperature: 838 C, steam-to fuel ratio: 0.63, Ni-olivine)

Fig. 3 shows the difference in the molar flow rate (in [moles/kgdry biomass]) for the main
gaseous components in the product gas at steady state conditions compared to natural
olivine. To close the mass and energy balances the mass flow of ungasified carbon to
the combustion zone has to be increased. This flow is not displayed in Fig. 3.
4
molar flow rate change [mole/kg biomass dry]

3 H2

1 CO2

CO C2H4

-1
CH4

-2

Fig. 3 Difference in the molar flow rate; catalyst olivine

Generally, it can be stated that the catalyst showed no further deactivation after
reaching steady state condition. This performance is also known from literature.7
Attrition is an important value for in-bed catalysts especially in circulating fluidized
bed processes. A comparison of the attrition of pure olivine and the catalyst showed no
significant difference. The attrition rates were about 0.025 kg/kg of fuel (dry). The
attrition rate could be lowered by sieving the entrained fines and bringing back the
larger fractions to the process. Investigations at the demonstration plant in Gssing
showed that the recirculation of the sieved bed material could lower the loss of bed
material down to 5 %16. Nevertheless frequent disposal of spent nickel catalyst
becomes an economic problem as well as an environmental burden. At present the ash
from the demonstration plant in Gssing has to be deposed anyway because the particle
size is too low to bring the ash out on the fields. But from the environmental point of
view this problem must be solved. Thus investigations about this topic are in process.

Two tar analysis methods were performed a gravimetric analysis and a gc-ms. The
total values of these two methods are different. The reason for this is that one of the
steps to determine the gravimetric tars is the evaporation of the toluene from the
sample. During this intermediate step the main part of the lighter hydrocarbons (e.g.
indene, naphtaline) disappears which is the reason for the gap between the gravimetric
tars and the so called gc-ms tars (Fig. 4 and Fig. 5).
The conversion is calculated with both the gravimetric tars and the tar amount
calculated from the results of the GC-MS measurements. The tar reduction was always
more than 90 % (Fig. 4).

5.0
100% olivine
4.5 100% Ni-olivine

4.0
tar content [g/Nm dry gas]

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0
0 5 10 15 20 25 30 35
time [h]

Fig. 4 Gravimetric tars over gasification time

What can be clearly seen in Fig. 5 is that the catalyst is more able to crack naphthalene
than indene. Secondary indene and naphthalene are intermediate products from the
cracking reactions of the heavy tars. The catalyst showed no loss of activity in
reforming the tars which can be seen in the amount of gravimetric tars (Fig. 4) as well
as on the tar composition over the gasification time (Fig. 6).
 Fig. 5 Comparison of the tar composition

100%

95%

90%

85%
tar composition [%]

rest
80% Acenaphthylene
Fluorene
75% Phenol
Styrene
70%
Naphthaline
65% Indene

60%

55%

50%
2.6 6.3 6.8 10.8 11.3 14.6 15.5 19.6 21.0 25.2 27.9 28.9
time [h]

Fig. 6 Tar composition using Ni-olivine over

gasification time

The catalyst is also able to reduce the ammonia content in the product gas from
820 ppm down to about 500 ppm that stands for a conversion of ammonia of more than
40 %.

Characterization of catalyst before and after test

X-Ray diffraction was performed on the samples to verify stability of the structure and
to follow the evolution of crystalline phases present in the catalyst during the test. The
diffractograms for the olivine and Ni-olivine samples before and after test are shown in
Fig. 7.
 Fe2O3
o Fe2O3

x MgSiO3

* NiO

o
x
o o
fresh olivine

used olivine
*
*
fresh Ni/olivine

used Ni/olivine
27 29 31 33 35 37 39 41 43 45 47 49
2 theta (degrees)

Fig. 7 XRD of olivine and Ni-olivine before and after test in FICFB gasifier

XRD of fresh olivine shows that the main lines are close to those of the Mg2SiO4
forsterite (34-0189 JCPDS file). The secondary crystalline phases observed are the
MgSiO3 phase (19-0606 JCPDS file) visualized by its 100-intensity line at 28.2 (2)
and Fe2O3 and Fe2O3. The presence of those secondary phases is due to a
modification (oxidation) of the (Mg,Fe)2SiO4 phase at high temperature during initial
calcination of olivine. After gasification test crystalline phase of olivine is preserved
but the intensities of the strongest lines of the secondary phases MgSiO3 and Fe2O3
decrease.
Similar trends can be also observed for Ni-olivine catalyst. The main lines of natural
olivine close to those of the Mg2SiO4 forsterite are maintained indicating that the
support structure is not changed. Small decrease of the MgSiO3 peak can be seen and
Fe2O3 disappears after the test. The two lines characteristic for the cubic NiO phase
(1-1239 JCPDS file), observed for the fresh catalyst indicate the presence of large size
aggregates. They diminish significantly after gasification test. This indicates a loss of
nickel containing particles by attrition in fluidized bed or nickel insertion into the
olivine structure.
 Fig. 8. SEM of olivine and Ni-olivine before and after tests in the gasifier.

The scanning electron micrographs of Fig. 8 show the changes in the surface of olivine
and Ni-olivine after the test in the pilot plant. The surface of the fresh olivine support is
shown in Fig. 8a. Smooth grain surface can be observed (probably due to prior
calcination at 1600C) with few iron oxide and olivine phase particles present on the
surface. The surface of used olivine presented in Fig. 8c is changed and seems to be
covered with a kind of deposit. Those changes of surface morphology can be related to
the changes of the surface composition obtained by X-ray microanalysis shown in
Table 4. Principal elements present on the surface of fresh olivine are Mg (26%), Si
(13.5%), O (54.5% ) and Fe (4.9%) coming from the olivine structure. In the sample of
used olivine lower percentages of Mg (17.3%) and Si (7.9%) and higher Fe (6.9%)
contributions can be found. The diminishing of Mg and Si contribution can be
explained by the apparition additionally of 7% of calcium and small contributions of P
(1%), K (0.4%) and Na (0.2%), elements present in biomass.
For fresh Ni-olivine shown in Fig. 8b the deposit of NiO grains cover homogeneously
nearly all surface of olivine. The NiO deposit can be also observed in the used Ni-
olivine sample but only a part of the surface is covered in this case.

Atomic %
Mg Si O Fe Ni Ca P K Na
fresh olivine 26,2 13,5 54,5 4,9 0,7 0,1 0,0 0,0 0,0
used olivine 17,3 7,9 58,8 6,9 0,0 7,4 1,0 0,4 0,2
fresh Ni/olivine 9,4 1,9 50,8 3,6 34,1 0,1 0,0 0,1 0,0
used Ni/olivine 15,6 5,8 53,6 8,8 10,3 5,2 0,5 0,2 0,0
Table 4. Surface composition of the olivine and Ni-olivine before and
after tests in the gasifier. (Determined by X-ray microanalysis coupled
with SEM)
X-ray microanalysis of fresh Ni-olivine shows lower surface concentrations of Mg, Si
and Fe due to the fact that the surface of olivine is covered with NiO deposit (34% of
Ni) observed in Fig. 8b. Comparison with the used Ni-olivine shows that Ni content is
about 3 times lower (10%) than in the fresh sample. These results could explain the
partial loss of nickel loosely deposited on the olivine surface and are in agreement with
XRD results. Very low concentration of P and alkaline metals (K and Na) as well as no
presence of S nor Cl on the surface of used Ni-olivine were observed.
In both olivine and Ni-olivine used in the biomass gasification we can observe the
apparition of Ca on the surface in a concentration comparable with this of Fe (7 and
5 % respectively). Without any further investigations it is difficult to evaluate the real
influence of this element. However calcium oxide has been already proven to be an
active catalyst for biomass gasification and it has been reported in the literature as an
additive having beneficial effect for enhancement of carbon gasification and increasing
coke deposition resistance.17

Different parts of the catalyst were also tested under reductive conditions (Fig. 9).
Reduction profile for fresh olivine shows only one reduction peak with a maximum at
660C. This broad peak is attributed to the reduction of the free iron oxide associated
to the olivine structure and formed by oxidation of Fe2+ present in olivine structure
during the initial calcination. The quantity of reducible iron oxide diminishes in the
used olivine sample. The decrease of the iron quantity is not in accordance with the
results of surface analysis obtained by X-ray microanalysis where the contrary effect is
observed. However, the difference could be due to the fact that TPR refers to reduction
of larger surface of the sample including pore surface and total reducible iron oxide
distribution can be different than in the external surface layer of 1 micron analyzed by
X-ray microanalysis.

2000
620C
fresh Ni/olivine
1500 used Ni/olivine
Intensity (a.u.)

fresh olivine
used olivine
1000
660C

500

0
300 400 500 600 700 800 900 950
1000 1100
T (C) isothermal

Fig. 9. TPR for olivine and Ni-olivine before and after

test in FICFB gasifier
For fresh Ni-olivine the wide reduction profile includes a free iron oxide reduction
peak (max. at 620C) followed by reduction of nickel species in different degree of
interactions with the support (reduction between 700 and 900C) and linked nickel
oxide reduction peak (max. at 910C). As it was shown previously6 after calcination at
1100C, nickel oxide forms grafts with the support and these strong nickelolivine
interactions were confirmed by high reduction temperature (910C). In this case, linked
nickel oxide reduction is compatible with gasification temperature.
TPR of used Ni-olivine shows a small decrease of the iron oxide reduction peak and an
important decrease of the high temperature shoulder (700-900C) attributed to the
reduction of different Ni species. The diminishing of the nickel species reduction peak
for the used catalyst confirms the results observed by XRD and X-ray surface analysis
and can be attributed to the loss of Ni due to attrition.

The particle size distribution was found nearly the same for the fresh and used Ni-
olivine catalyst with a mean particle size 574 and 578 m respectively, confirming the
attrition resistance (which could be related with the olivine hardness) of catalyst
particles.

It seems that the only cause of catalyst deactivation could be the loss of the active
phase by attrition of nickel loosely deposited on the olivine surface. This could explain
the initial drop of activity. The followed stable performance indicates that grafted Ni is
resistant for attrition in the fluidized bed.
CONCLUSIONS

Ni-olivine preparation has been successfully scaled up from laboratory scale to

industrial scale (100kg). Large amount of NiO/olivine catalyst was prepared with high
homogeneity. The results of catalyst characterization were in good agreement with
those obtained for catalysts prepared previously in small amount (20g), confirming the
presence of strong Niolivine interactions guarantying efficient performance of the
catalyst in biomass gasification in the pilot plant.

This work has proven that Ni-olivine catalyst fulfils all the criteria for the use in
fluidized bed biomass gasification process. It is highly effective in the removal of tars
and methane, increases the hydrogen content and decreases ammonia content. The
strong metal support interactions (Ni in the form of grafts on the olivine surface)
prevent attrition, sintering and volatilization of the nickel. Ni-olivine catalyst proved
high stability and resistance for repeated high temperature processing in oxidizing and
reducing atmosphere during circulation between the gasification (850C) and
combustion zone (900C). This provides constant regeneration of the catalyst
excluding any deactivation as a result of carbon fouling. No deactivation by catalyst
poisons like sulfur, chlorine, and alkali metals was observed during the 30h of catalyst
testing.
Using of Ni-olivine as an inexpensive (compared to commercial catalysts) primary
catalyst in the fluidized bed is a good solution to enhance the efficiency of the process,
to simplify the gas cleaning line without an expensive secondary catalytic reactor and
to decrease the operating costs.

ACKNOWLEDGMENT

This work was carried out under EC-Project NNE5-2000-00212 (Contract ENK5-
CT2000-00314). The authors thank the European Commission for its financial support.
The work also received financial support from RENET-Austria, which is a competence
network in the framework program of Kind/Knet. This financial support is gratefully
acknowledged.
REFERENCES

(1) H. Hofbauer, R. Rauch, G. Loeffler, S. Kaiser, E. Fercher, H. Tremmel, (2002) Six

Years Experience with the FICFB-Gasification Process, Proc 12th Europ. Conf. and
Techn. Exhibition on biomass for Energy, Industry and Climate ProtectionAmsterdam
(2) S. Rapagn, N. Jand, A. Kiennemann, P.U. Foscolo, (2000) Biomass & Bioenergy,
19 (3) 187.
(3) C. Courson et al., (2001) European Patent No. PCT/FR01/01547
(4) H. Hofbauer, R. Rauch, (2001) Hydrogen-rich gas from biomass steam
gasification, Publishable Final Report, Contract JOR3CT970196
(5) C. Courson, E. Makaga, C. Petit, A. Kiennemann, (2000) Catalysis Today, 63 427
(6) C. Courson, L. Udron, C. Petit and A. Kiennemann, (2002) Science and
Technology of Advanced Materials, 3 (2002) 271.
(7) Baker, E.G.; Mudge, L.K.; Brown, M.D. (1987) Steam Gasification of Biomass
with Nickel Secondary Catalysts, Ind.Eng.Chem.Res. 26, 1335-1339
(8) D. Dayton, (2002) A Review of the Literature on Catalytic Biomass Tar
Destruction, NREL, Golden, CO, USA, Report no. NREL/TP-510-32815
(9) Neeft, J.P.A.; Knoef, H.A.M.; Zielke, U.; Sjstrm, K.; Hasler, P.; Simell, P.A.;
Dorrington, M.A.; Abatzoglou, N.; Deutch, S.; Greil, C.; Buffinga, G.J.; Brage, C.;
Soumalainen, M.(1999) Guideline for Sampling an Analysis of Tar and Particles in
Biomass Producer Gas, version 3.1; Energy project EEN5-1999-00507 (Tar protocol)
(10) Corella, J.; Orio, A.; Toledo, J.M. (1999) Biomass Gasification with Air in
Fluidized Bed: Exhaustive Tar Elimination with Commercial Steam Reforming
Catalysts, Energy & Fuels, Vol. 13, Number 3, 702-709
(11) Simell, P.; Sthlberg, P.; Solantausta, J.; Hepola, J.; Kurkela, E. (1996)
Gasification Gas Cleaning with Nickel Monolith Catalyst, Bridgwater, A.V. &
Boocock, D.G.B. (eds.). Developments in thermochemical biomass conversion. Vol. 2.
Banff, Canada, 20-24 May
(12) Ising, M. (2002) Zur katalytischen Spaltung teerartiger Kohlenwasserstoffe bei
der Wirbelschichtvergasung von Biomasse, PhD, Frauenhofer Umsicht, D-46047
Oberhausen, Germany
(13) Pfeifer, C.; Rauch, R.; Hofbauer, H. (2004) In-Bed Catalytic Tar Reduction in a
Dual Fluidized Bed Biomass Steam Gasifier, Ind.Eng.Chem.Res.; 43 (7); 1634-1640
(14) Pfeifer, C.; Rauch, R.; Hofbauer, H. (2004) Hydrogen-rich Gas Production with a
Catalytic Dual Fluidized Bed Biomass Gasifier, 2nd World Conference and Technology
Exhibition on Biomass for Energy, Industry and Climate Protection, Rome 10-14 May,
proceedings in press
(15) Hofbauer, H.; Rauch, R.; Loeffler, G.; Kaiser, S.; Fercher, E.; Tremmel, H. (2002)
Six Years Experience with the FICFB-Gasification Process, presented at the 12th
European Conference and Technology Exhibition on Biomass for Energy, Industry and
Climate Protection; Amsterdam
16 Cubero, A. (2003) Untersuchung von Additiven zur Verbesserung der Gasqualitt
eim FICFB-Vergasungsprozess, Master Thesis, Vienna University of Technology,
Vienna
(17) Zhang Z. L., Verykios X. E., Catal. Today, 21 (1994) 589.