Gasification of Natural and Waste Biomass in a Pilot Scale Fluidized Bed Reactor

Abstract

Three commercially available biomass fuels, made of natural and waste wood, were fed in a pilot scale bubbling fluidized bed gasifier having an internal diameter of 0.381 m and a maximum feeding capacity of 100 kg/h. The experimental runs were carried out at about 850°C and under values of the equivalence ratio between 0.20 and 0.30. The fluidized bed was generally made of natural olivine even though some runs utilized beds of dolomite or quartz sand. Measurements taken during each run include the gas composition, the content of tar in the syngas, the mass flow rate and composition of entrained fines collected at the cyclone and the characterization of bed material. The results indicate that the air gasification process is technically feasible with all the biomass tested. The olivine as tar removal bed catalyst provides for different results with waste and natural biomass fuels.

Full text

PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by:
[EBSCOHost EJS Content Distribution - Superceded by 916427733]
On:
23 June 2010
Access details:
Access Details: [subscription number 911724993]
Publisher
Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK
Combustion Science and Technology
Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713456315
Gasification of Natural and Waste Biomass in a Pilot Scale Fluidized Bed
Reactor
U. Arenaab; L. Zaccarielloa; M. L. Mastelloneab
a Department of Environmental Sciences, Second University of Naples, Caserta, Italy b AMRA s.c. a r.l.,
Napoli, Italy
Online publication date: 08 June 2010
To cite this Article Arena, U. , Zaccariello, L. and Mastellone, M. L.(2010) 'Gasification of Natural and Waste Biomass in a
Pilot Scale Fluidized Bed Reactor', Combustion Science and Technology, 182: 4, 625 — 639
To link to this Article: DOI: 10.1080/00102200903467689
URL: http://dx.doi.org/10.1080/00102200903467689
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.

GASIFICATION OF NATURAL AND WASTE BIOMASS
IN A PILOT SCALE FLUIDIZED BED REACTOR
U. Arena,
1,2
L. Zaccariello,
1
and M. L. Mastellone
1,2
1
Department of Environmental Sciences, Second University of Naples,
Caserta, Italy
2
AMRA s.c. a r.l., Napoli, Italy
Three commercially available biomass fuels, made of natural and waste wood, were fed in a
pilot scale bubbling fluidized bed gasifier having an internal diameter of 0.381 m and a
maximum feeding capacity of 100 kg/h. The experimental runs were carried out at about
850C and under values of the equivalence ratio between 0.20 and 0.30. The fluidized bed
was generally made of natural olivine even though some runs utilized beds of dolomite or
quartz sand. Measurements taken during each run include the gas composition, the content
of tar in the syngas, the mass flow rate and composition of entrained fines collected at
the cyclone and the characterization of bed material. The results indicate that the air
gasification process is technically feasible with all the biomass tested. The olivine as tar
removal bed catalyst provides for different results with waste and natural biomass fuels.
Keywords: Biomass; Fluidized bed; Gasification; Tar; Waste
INTRODUCTION
A sustainable energy future requires a reliable combination of renewable
energy sources and advanced energy technologies. Biomass is a renewable energy
source that in the last 15 years received a great and worldwide interest, mainly
related to the reduction of greenhouse gases allowed by its utilization. On the other
hand, the environmental impact of solid fuel combustion processes can be worsened
by the heterogeneous composition of waste biomass leading to unacceptable
environmental and energetic performances (Baratieri et al., 2008; Devi et al.,
2003). A number of novel technologies utilizing gasification processes are now avail-
able to obtain fuel upgrade paths by producing syngas from biomass. This approach
provides a wide range of products, extending from clean fuel gas and electricity to
bulk chemicals, such as ammonia and methanol (Juniper, 2001; Malkow, 2004).
Among all biomass gasification technologies, fluidization is the most promising
for a series of attracting reasons. In particular, the great operating flexibility of
(bubbling and circulating) fluidized bed reactors makes it possible to utilize different
fluidizing agents, reactor temperatures, and gas residence times to add reagents along
Received 23 March 2009; revised 12 October 2009; accepted 12 October 2009.
Address correspondence to Umberto Arena, Department of Environmental Sciences, Second
University of Naples, Via Antonio Vivaldi, 43, 81100 Caserta, Italy. E-mail: umberto.arena@unina2.it
Combust. Sci. and Tech., 182: 625–639, 2010
Copyright #Taylor & Francis Group, LLC
ISSN: 0010-2202 print=1563-521X online
DOI: 10.1080/00102200903467689
625
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

the reactor freeboard or riser and to operate with or without a specific catalyst
(Arena and Mastellone, 2005; Arena and Mastellone, 2006).
The greatest technical challenge to overcome for the successful development
of commercial advanced biomass gasification technologies is that of an improved
syngas cleaning able to meet defined specifications. During gasification, tars, heavy
metals, halogens, and alkaline compounds are released within the gas and can cause
environmental and operational troubles. The continual buildup of condensable
organic compounds present in the producer gas (usually referred to as tars) is
the main problem because it causes troubles in the process equipments as well
as the devices for end-use application of the syngas (Arena et al., 2009). Because
the markets for biomass gasifiers without gas cleaning are rather limited, the key
to achieving economically and environmentally efficient energy recovery from
natural and waste biomass gasification is overcoming problems associated with the
formation and release of these contaminants (Babu, 2006; Milne et al., 1998). The
approaches for reduction of tar formation and for tar removal from obtained syngas
can substantially be divided in treatments inside the gasifier (primary methods) and
hot gas cleaning after the gasifier (secondary methods). Secondary methods (thermal
or catalytic tar cracking and mechanical methods, such as the use of cyclones,
ceramic filters, fabric or electrostatic filters, and scrubbers) are reported to be very
effective in several cases, even though they are not always economically viable and
are often particularly complex when a considerably low tar content is required.
Primary methods (adequate selection of main operating parameters, use of a proper
bed additive or catalyst, specific gasifier design modifications) are gaining increasing
attention for biomass gasification because they may strongly reduce the need for
downstream cleanup. It is likely that an adequate combination of primary and
secondary treatments may optimize the gasifier performance and allows the
production of a syngas with minimum tar content.
The aim of this study was to evaluate the influence of some operating
parameters (gasifying agent, equivalence ratio, fluidizing velocity) and of different
bed materials on the performance of the gasification process of three biomass fuels
fed in a fluidized bed gasifier of pilot scale.
PILOT SCALE FLUIDIZED BED GASIFIER
The pilot scale bubbling fluidized bed gasifier (BFBG) has a feedstock capacity
between 30 and 100 kg=h, depending on the type of fuel, and a maximum thermal
output of about 500 kW. Its main design and operating features are schematically
listed in Table 1. The pilot scale plant is composed of three main sections: the feeding
system, the fluidized bed gasifier, and the syngas treatment unit. The feeding system
can be divided in the blast feeding (measuring, mixing, and injection of gasification
agents) and the fuel feeding (measuring and injection of solid feedstock). The blast
feeding is heated up to 200C by an electric heater, sent to a mixing point with an
optional stream of steam at about 150C, and finally heated by a second electric
heater up to 600C before entering the reactor. In the experiments reported here,
air or an air–steam mixture was used as a blast agent and always injected at the
bed bottom while the fuel was always fed by means of an overbed feeding system.
The fuel and the blast flow rates were mutually adjusted so that at the fixed fluidizing
626 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

velocity the desired equivalence ratio (ER) was obtained (where ER is defined
as the ratio between the oxygen content of air supply and that required for the
stoichiometric complete combustion of the fuel effectively fed to the reactor).
The gasification section is composed of a cylindrical BFBG reactor, which is heated
up to the reaction temperature by the sensible heat of preheated blast gases and by a
set of three external electrical furnaces. The gas generated in the reactor is sent to
the syngas treatment section composed of a high-efficiency cyclone, a simple wet
scrubber (for removal of tars, residual fly ashes, and acid gases) and a flare. A more
accurate description of the plant is provided by Arena et al. (2008a).
BIOMASS FUELS, BED MATERIALS, AND EXPERIMENTAL PROCEDURES
The reactions occurring during the gasification process and, in particular,
during the pyrolysis step, as well as the type of products obtainable from these
reactions, are strongly correlated to the chemical structure (in addition to the
chemical composition) of the biomass. Any biomass can be characterized by at
least three main components—cellulose, hemicelluloses, and lignin—to which it is
necessary to add the presence of extractives (e.g., low molecular weight compounds
such as phenols, waxes) and inorganics (e.g., salts of Ca, Cl, Fe, K, Mg, Na).
The fuels utilized in this study come from the Italian consortium for wood
recovery and are different for source and, as a consequence, for composition. In
particular, RIL1 is a natural wood utilized to prepare fuel pellets for domestic heat-
ing: its quality is high because it cannot be obtained by contaminated wood-based
materials. On the contrary, RIL2 and RIL3 are wood wastes that cannot be utilized
to produce fuel for domestic heating because they come from potentially contami-
nated waste: RIL2 is made of sawdust from wood packaging industry and RIL3 is
obtained as a recycled product from furniture and from door and window frames.
Their physical and chemical properties are reported in Table 2, together with the
relative amounts of cellulose, hemicelluloses, and lignin, whereas the chemical
composition of inorganic fraction is reported in Table 3. The natural biomass
RIL1 consists of more than 22% of lignin (the typical range is 22–30%) and of about
Table 1 Main design and operating features of the bubbling fluidized bed gasifier
Geometrical parameters ID ¼0.381 m; total height ¼5.90 m; reactive zone height ¼4.64 m;
wall thickness ¼12.7 mm
Feedstock capacity 30–100 kg=h (depending on the type of fuel)
Thermal output about 500 kW
Typical bed amount 135–215 kg
Feeding equipments inbed (water cooled) and overbed (air cooled) screw feeders
Gasifying agents air, oxygen, steam, carbon dioxide (alone or as mixture)
Range of operating temperatures 700–950C
Range of fluidizing velocities 0.3–1 m=s
Flue gas treatments cyclone, scrubber, flare
Safety equipments water seal, safety valves, rupture disks, alarms, nitrogen line for
safety inerting
Main process variables reactor temperature, bed height, fluidizing velocity, blast flow rate,
equivalence ratio
GASIFICATION OF NATURAL AND WASTE BIOMASS 627
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

65% of carbohydrates, cellulose, and hemicelluloses (the typical range is 60–70%);
the waste biomass fuels RIL2 and RIL3 have a reduced content of lignin (about
18%), together with 60% or less of carbohydrates. It should be also noted that the
carbon–oxygen ratio moves from about 1 for the natural biomass to about 2 for
Table 3 Chemical composition of inorganic fraction of the biomass tested (expressed with reference to the
dry fuel or, in the case of chlorides, to the fuel as received)
Element (mg=kg
ds
) RIL1 RIL2 RIL3
Arsenic <0.1 0.35 0.21
Aluminium 150 142 903
Antimony 2 0.62 98.6
Cadmium 0.05 0.11 0.76
Calcium 1750 3382 15288
Cobalt 0.17 0.26 1.03
Chromo 0.95 2.23 5.99
Iron 290 428 622
Magnesium 465 657 1321
Manganese 15 63.4 39.4
Mercury 0.01 0.12 <0.1
Nickel 0.35 1.52 3.8
Lead 1.2 4.44 35.0
Potassium 330 137 1483
Copper 1.5 <0.1 3.85
Sodium 110 186 3461
Tin 90 41.3 50
Thallium <0.25 <0.25 <0.25
Vanadium <0.1 0.2 61.2
Clorides (mg=kg
ar
) 3650 — 803
Table 2 Main chemical and physical features of the biomass fuels tested
Factor RIL1 RIL2 RIL3
Ultimate analysis (%)
C45.3 57.4 50.2
H5.6 7.3 7.0
N0.5 0.7 2.3
S000
O(by difference) 38.4 26.4 25.9
moisture 9 7 11.3
ash 1.2 1.2 3.3
C=O ratio 1.2 2.2 1.9
Cellulose (g=100 g) 45.1 43.2 38.8
Hemicelluloses (g=100 g) 19.6 19.3 20.4
Lignin (g=100 g) 22.3 18.5 18.0
LHV (kJ=kg
ar
)15700 20300 18000
Size (diameter and height; mm) 5, 15 (5–25) 5, 10 (10–20) 5, 15 (10–20)
Bulk density (loose – packed; kg=m
3
) 575 (550–600) 725 (650–800) 560 (530–590)
Note. Cellulose data are as obtained by the value of acid detergent fiber (ADF) less that of acid deter-
gent lignin (ADL). Hemicellulose data are as obtained by the value of neutral detergent fiber (NDF) less
that of ADF. Lignin data are as obtained by the value of ADL.
628 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

the two waste biomasses: this suggests a larger presence of additives in RIL2 and
RIL3, such as glues, resins, and foams (that, particularly for RIL3, is further con-
firmed by the high content of nitrogen).
Three types of bed materials were used during the experimental runs, olivine,
quartz sand, and dolomite. Quartz sand is the typical bed material of fluidized
bed combustor and gasifier; olivine and dolomite have been chosen for their good
performance as tar-removal bed additives, as reported in several studies on gasifi-
cation of biomass (Corella et al., 1999; Devi et al., 2005a, 2005b; Devi et al.,
2003) and polyolefinic plastic wastes (Arena et al., 2009; Mastellone and Arena,
2008). Their main chemical and physical properties are reported in Table 4. The uti-
lized olivine (a magnesium–iron silicate mineral), (Mg,Fe
2
)SiO
4
, comes from Austria
and is composed of different oxides: MgO ¼48–50%, SiO
2
¼39–42%, Fe
2
O
3
¼
8–10.5%, (Al
2
O
3
þCr
2
O
3
þMg
3
O
4
¼0.8), CaO <0.4%.
Each run has a startup of about 3 hr, during which the electric heaters rise the
reactor temperature up to about 800C, while the bed is fluidized at the desired value
of fluidizing velocity, U. At this point, the values of fuel and air flow rate are
adjusted in order to obtain the desired value of ER. The reactor is operated under
these conditions until all the values of gas composition remain constant for about
1 hr. The gas and solids sampling procedures are then activated and measurements
of pressure, temperature, blast flow rates, syngas composition, and fines collected
at the cyclone are taken.
In all the runs gas composition downstream of the syngas treatment section
was on-line measured by means of two systems: the first utilizes IR analyzers (Horiba
VA-3115 for CO, CO
2
,andO
2
; Horiba VA-3001 for CH
4
; and Teledyne Anal. Instr.-
2000 for H
2
), the other uses a couple of Agilent 3000 micro-gas-chromatographs
equipped with different columns for detection of lighter and heavier hydrocarbons
as well as carbon monoxide and dioxide, hydrogen, nitrogen and water. This double
system allows a high reliability of measured gas composition. Gas was also sampled
in other three points (at two levels of 1.5 and 2.3 m along the reactor and just after
the gasifier exit) and collected in Tedlar bags to be sent to off-line measurements.
Because the tar content is one of the major concerns, two different methods of tar
evaluation have been used. The first conservatively imputes to the tar amount
(assumed to be composed of all organic compounds with a molecular weight larger
than benzene, excluding soot and char) the whole carbon loading which, as a result
of a mass balance on atomic species with an overall error <5%, cannot be attributed
either to the produced gas or to the solids collected at the cyclone or present inside
the bed. The second method utilizes the TA 120 Tar Analyzer recently developed
by the Institute of Process Engineering and Power Plant Technology of the
Table 4 Main physical and hydrodynamic features of the bed materials tested
Bed material Olivine Quartz sand Dolomite
Size range (mm) 200 400 200 400 300 800
Sauter mean diameter (mm) 298 205 360
Particle density (kg=m
3
) 2900 2600 2900
Minimum fluidization velocity at 850C(m=s) 0.030 0.013 0.044
Terminal velocity at 850C(m=s) 2.0 1.0 2.8
GASIFICATION OF NATURAL AND WASTE BIOMASS 629
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

University of Stuttgart for quasi-continuous on-line measurement of the content of
condensable aromatic hydrocarbons, designated as tars, in the producer gas from
biomass gasification.
Some sampling of these condensable species was made at the reactor exit, for
about 30 min, by means of four in-series cold traps, which are encased in an insulat-
ing container and chilled by a mixture of ice and household salt, that reaches a
temperature of about 15C. The condensed samples are sent to off-line analyses:
they are first dissolved using dichloromethane, which is flushed in the coils until
they are completely clean, then weighed, and finally sent to a gas chromatograph
coupled with a mass spectrometer (GC-MS). This is able to recognize the tars
belonging to the classes from 2 to 5 of the tar classification system proposed by Kiel
et al. (2004).
Data obtained from on- and off-line gas measurements and from chemical
analyses of solid samples are processed to develop for each run complete mass
balances on atomic species and related energy balances. The flow rate of produced
syngas is determined by the tie component method (Felder and Rousseau,
2000) applied to the value of nitrogen content in the dry syngas, as obtained by
(on- and off-line) GC measurements. Carbon balance allows the determination of
tar content on the basis of the above recalled assumption, whereas hydrogen balance,
which takes into account also the tar hydrogen content as measured in some samples
taken during the run, allows that of water content in the syngas (Basu, 2006). The
whole mass balance must take into account the amount of carbon present in the
bed and that elutriated from the reactor as carbon fines collected by the cyclone.
The latter are periodically collected, weighed, and analyzed, so that their amount
is easily determined. The amount of carbon present in the bed is instead determined
by two methods, by analyzing one or two bed solids samples taken during each
run and establishing oxidizing conditions in the reactor at the end of each run
(i.e., after the stop of fuel feeding) and calculating the related amounts of
correspondingly produced CO and CO
2
.
RESULTS AND DISCUSSION
The basic idea of primary methods is that the values of main operating
parameters of a gasifier can be adequately selected so that the gasifier performance
is optimized, and the quality of exit gas can meet the requirements for the different
end-use applications. A great research effort in the developing and validating
of primary methods for biomass gasification has been provided in the last
10 years (Babu, 2006; Corella et al., 1999; Devi et al., 2003; Gil et al., 1999;
Narvaez et al., 1996) but so far they are not yet fully understood and implemented
commercially.
The pilot scale BFBG was fed with the three biomass fuels under various
experimental conditions in order to obtain information about the role of main
operating parameters (gasifying agent, equivalence ratio, bed material, fluidizing
velocity) in the definition of process performance parameters and, in particular, in
the formation and decomposition of tar. Table 5 lists the values of the operating
parameters chosen for the complete set of experiments together with those of the
main performance parameters.
630 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

Table 5 Operating parameters and main results of all reported experiments
Fuel Test
Bed
material
T
bed
(C)
U
(m=s) ER
S=F
(kg
steam
=kg
fuel
)
%
CO
2
%
CO
%
H
2
%
CH
4
Qsyngasðm3
N=hÞLHVsyngasðkJ=m3
NÞ
Specific
energy
(kWh=kg
fuel
)
Ctarðg=m3
NÞ
(estimated)
Ctarðg=m3
NÞ
(measured) CGE
RIL1 1 olivine 856 0.6 0.21 0 15.4 18.8 14.2 6.0 122 6800 3.30 0 — 0.74
2 olivine 843 0.6 0.17 0 15.7 20.4 15.2 6.6 137 7700 3.31 0 — 0.75
3 olivine 911 0.7 0.26 0 13.3 19.9 14.5 4.4 114 6200 3.47 0 — 0.78
4 olivine 860 0.4 0.29 0 14.6 16.5 13.3 4.2 76 5500 3.49 0 — 0.79
7 olivine 850 0.4 0.20 0 14.7 19.0 15.2 4.6 74 6300 3.14 0 — 0.71
5 olivine 832 0.7 0.28 0.90 18.2 11.9 17.5 4.2 78 5300 3.42 0 — 0.77
6 olivine 808 0.6 0.26 0.63 18.6 11.6 17.0 4.1 72 5200 3.12 0 — 0.70
8 olivine 847 0.7 0.20 0.63 17.2 15.8 21.1 5.1 85 6600 3.74 8 — 0.84
9 olivine 864 0.6 0.25 0 14.3 19.1 13.8 4.9 116 6200 3.33 0 — 0.76
10 sand 855 0.7 0.25 0 15.5 17.2 13.3 5.0 120 6100 3.32 0 — 0.76
RIL2 1 olivine 925 0.7 0.30 0 13.7 16.6 10.3 3.4 111 4800 3.90 2 4 0.69
2 olivine 907 0.7 0.26 0 13.3 17.2 11.8 3.7 115 5100 3.71 2 4 0.66
3 olivine 885 0.7 0.22 0 13.5 18.9 13.1 4.3 125 5900 4.02 4 1 0.71
4 sand 945 0.7 0.31 0 13.3 15.9 9.4 3.2 107 4500 3.68 9 7 0.65
5 sand 865 0.7 0.26 0 13.5 17.8 11.2 3.9 117 5400 3.99 9 1 0.71
7 sand 854 0.4 0.16 0 14.4 17.9 12.8 5.3 83 6200 3.31 70 — 0.59
6 sand 768 0.7 0.13 0.62 15.8 17.0 14.5 5.8 75 6700 3.20 — — 0.57
8 dolomite 923 0.7 0.28 0 11.0 19.4 13.1 3.3 117 5300 4.30 0.6 0.4 0.72
RIL3 1 olivine 881 0.7 0.25 0 14.7 16.3 12.3 3.2 116 4900 3.08 13 1.4 0.62
2 olivine 907 0.7 0.31 0 14.7 15.1 9.9 3.3 109 4600 3.43 0 1.9 0.69
3 olivine 856 0.7 0.21 0 14.3 17.6 14.8 4.3 128 5900 3.48 21 1.7 0.70
631
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

Gasifying Medium
The possibility of utilizing different gasifying agents, such as air, steam, steam–
oxygen, and carbon dioxide, greatly contributes to the recognized high flexibility of
gasification process. Each gasifying medium affects the selectivity of the gasification
reactions and then the composition and heating value of the producer gas (Basu,
2006; Juniper, 2001).
Data reported in Figures 1 and 2 (and further detailed in Table 5) refer to
experiments carried out with air as fluidizing gas and calcined olivine as bed
material. The utilization of olivine as an active bed material was suggested by its
good performance in fluidized bed gasifiers of biomass (Devi et al., 2005; Pfeifer
et al., 2004; Rapagna
`et al., 2000) and plastic waste (Arena et al., 2008b; Arena
et al., 2009; Mastellone and Arena, 2008). A specific study about the role of olivine
as a tar removal catalyst during the gasification of a plastic waste (Mastellone and
Arena, 2008) indicated that magnesium and iron, both largely present in the olivine
Figure 1 Concentration of some components in the syngas as measured downstream of the scrubber in
the experiments with all the biomass fuels and a bed of olivine (shaded symbols ¼air gasification at
U0.7 m=s; open symbols¼air–steam gasification at U 0.7 m=s; half-shaded symbols ¼air gasification
at U ¼0.4 m=s).
632 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

particles, activate the endothermic decomposition reactions of hydrocarbon frag-
ments that are the first precursors of tar formation. In particular, magnesium cata-
lytically enhances the dehydrogenation and isomerization reactions of fragments
produced by thermal cracking: pC
x
H
y
!qC
n
H
m
þrH
2
; whereas iron catalytically
assists the dehydrogenation and carbonization reactions: C
n
H
m
!nC þm=2H
2
,
where C
x
H
y
represents the heavier fuel fragments and C
n
H
m
hydrocarbons with
a smaller number of carbon atoms and a larger degree of unsaturation than C
x
H
y
.
Data in Figure 1 show that the content of hydrogen, carbon monoxide, meth-
ane, and C
2
H
m
decreases as the equivalence ratio ER increases for all the biomass
fuels tested. This necessarily implies a reduction of lower heating value (LHV) of
the produced syngas but even an increase in the specific syngas yield (Figure 2)
and a positive reduction of tar content (Table 5 and shaded symbols in Figure 3),
as a consequence of the larger oxygen amount that can react with volatiles in the
pyrolysis zone. These opposite effects balance to each other producing no remark-
able variations in the specific energy contained in the syngas and in the chemical
energy of the fuel transferred to the syngas (also known as the cold gas efficiency
Figure 2 Parameters of process performance as obtained in the experiments with all the biomass fuels and
a bed of olivine (shaded symbols ¼air gasification at U 0.7 m=s; open symbols ¼air–steam gasification
at U 0.7 m=s; half-shaded symbols ¼air gasification at U ¼0.4 m=s).
GASIFICATION OF NATURAL AND WASTE BIOMASS 633
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

[CGE]). The syngas tar content and CGE are likely the most important parameters
of the gasification process performance. With reference to them, it is evident that
the values related to the natural biomass are remarkably better than those related
to the other two waste fuels.
It is noteworthy that the tar concentrations reported in Table 5 and Figure 3
refer to the on-line measurements made downstream of the scrubber unit and are
obtained by means of a mass balance on atomic species. Moreover, some off-line
analyses were made on samples taken upstream of the cyclone (i.e., just after the
reactor exit), in order to obtain the concentration and composition of tars upstream
of the cleaning devices. Data for runs RIL1–9 and RIL3–3 show that a fraction
between 80 and 90% of tar components has been detected as PAHs: this fraction
was essentially composed by naphthalene and indene (83% and 16% for RIL1–9
and 82% and 10% for RIL3–3, respectively). Both of these compounds can be easily
removed from the syngas by means of traditional secondary methods, such as
wet scrubbers.
The utilization of steam as the only gasifying medium requires complex design
for heat supply to the endothermic process. The utilization of some amount of oxy-
gen can provide the heat necessary to allow an autothermal process for biomass and
waste gasification (Babu, 2006; Juniper, 2001). Accordingly, steam has been added to
air in some runs, in order to obtain a steam–fuel ratio (S=F) from 0.6 to 0.9. Figure 4
clearly shows the sudden change occurring in syngas composition just after the steam
addition in the reactor: there is a reduction of CO content and an increase of CO
2
Figure 3 Tar concentration in the syngas downstream of the scrubber as estimated by material balances on
atomic species.
634 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

and H
2
contents, which indicates the predominant role of the water–gas shift
reaction: CO þH
2
O$CO
2
þH
2
. The analysis of results showed in Figures 1 and 2
(open symbols) confirms the increase of hydrogen and the decrease of carbon
monoxide concentration but also highlights the reduction of low heating values of
the produced syngas. The overall effect of steam injection, in the tested range of
S=F, is almost negligible in terms of specific energy and CGE.
Fluidizing Velocity
Some runs were also carried out at a lower value of superficial gas velocity in
the bed, equal to about 0.4 m=s. The data reported in Figures 1–3 and 5 indicate a
slight worsening of the main process performance parameters, accordingly with
results already obtained with plastic wastes (Arena et al., 2009; Mastellone and
Arena, 2008).
Bed Material
Olivine, quartz sand, and dolomite were used in a series of runs with the
biomass fuel RIL2 and in a single test with biomass fuel RIL1. The results are listed
in Table 5 and visualized in Figures 3 (for RIL1 and RIL2) and 5 (only for RIL2).
Data indicate that runs carried out at the same ER but with a bed of olivine (or dolo-
mite) have a lower content of tar in the produced syngas, even though no remarkable
effects of the three bed materials used were found on the other process performance
parameters, such as the hydrogen content in the producer gas. These results suggest
that the catalytic action of olivine is only partially present during air gasification of
waste biomass. It is likely that the catalytic support to the cracking and isomeriza-
tion due to the magnesium action is always active so that the heavier fragments
Figure 4 Time profiles of syngas composition moving from air gasification to air–steam gasification in a
test with RIL1 fuel.
GASIFICATION OF NATURAL AND WASTE BIOMASS 635
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

are broken and a number of unsatured hydrocarbons with two or three carbon
atoms are formed. On the contrary, the catalytic enhancement of the dehydrogena-
tion and carbonization determined by the potentially active sites of iron is absent, so
that the hydrogen content remains low and, for waste biomass fuels, the tar forma-
tion is only partially reduced.
This conclusion is supported by the values reported in Table 6 for the complete
analyses of inorganic fractions of the fines collected at the cyclone during runs
carried out under similar operating conditions with the three tested fuels in a bed
of olivine. In this table the enrichment factor Fen of an element in the ash is defined
as (Zevenhoven and Kilpinen, 2001): Fen ¼(element concentration in ash=element
concentration in fuel) (% ash in fuel)=100. Other studies with different refuse-
and packaging-derived fuels (Arena et al., 2010; Mastellone and Arena, 2008) indi-
cated that when the olivine is active the final result of the enhanced series and parallel
reactions is the production of molecular hydrogen and carbon coke. The latter links
to the elemental iron of olivine by forming coordination complexes (Cotton and
Wilkinson, 1999) that are entrained out of the reactor. As a consequence, when
Figure 5 Some important parameters of process performance as obtained in the experiments with RIL2
with different bed materials (shaded symbols ¼air gasification at U ¼0.7 m=s; open symbols ¼air–steam
gasification at U ¼0.7 m=s; half-shaded symbols ¼air gasification at U ¼0.4 m=s).
636 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

the catalytic action of olivine is present, the fines collected at the cyclone should
contain larger quantity of iron than that entering the reactor with the fuel.
On the basis of the data in Table 6, the ratio between the iron flow rate
that escapes the reactor as fines and that of the iron that enters into the reactor as
inorganic fraction of the fuel is 4.4 for RIL1, 3.4 for RIL2, and 0.8 for RIL3. Taking
in mind that when olivine is very active this ratio can reach values as high as 200 and
when olivine is completely inhibited it remains to values as low as 0.1 or less (Arena
et al., 2010), data obtained with the tested biomass support the hypothesis that there
is only a limited activity of olivine, in particular with the waste biomass fuels. Only
the reactions of production of low-molecular weight hydrocarbons occur, whereas
those of aromatization and coking are substantially inhibited. A possible expla-
nation is that the catalytic action determined by magnesium, which is largely present
inside olivine (that is made of MgO for about the 50%), is almost completely
developed. On the other hand, the catalytic action due to iron oxides is inhibited
or strongly limited. This effect could be due to the presence in the ash of the waste
biomass of larger fractions of ferrous and nonferrous metals (Table 3) that could act
as competitors of the iron of olivine particles.
CONCLUSIONS
The fluidized bed gasification process is technically feasible with all the
materials tested.
Table 6 Composition of the inorganic fraction (expressed with reference to the dry fuel or, in the case of
chlorides, to the fuel as received) and the relative enrichment factor of fines collected at the cyclone, during
runs with the three tested biomass fuels in a bed of olivine (runs RIL1–9, RIL2–2, RIL3–1)
Element (mg=kg
ds
) RIL1 Fen
RIL1
RIL2 Fen
RIL2
RIL3 Fen
RIL3
Arsenic 7.8 1.03 0.1 0 0.94 0.17
Aluminium 18300 1.61 4637 0.42 15770 0.65
Antimony 6.6 0.04 17.3 0.36 152 0.06
Cadmium 1.1 0.29 2.65 0.31 4.57 0.22
Calcium 13915 0.10 33800 0.13 153900 0.37
Cobalt 39.9 3.10 25.3 1.26 10.4 0.38
Chrome 56.2 0.78 191 1.11 94.2 0.59
Iron 41021 1.87 47500 1.43 19820 1.19
Magnesium 117028 3.32 38650 0.76 19160 0.54
Manganese 356 0.31 594 0.12 392 0.37
Mercury 1.5 0.20 0.1 0.01 0.1 0.04
Nickel 192 7.23 519 4.41 146 1.43
Lead 98 1.07 196 0.57 1133 1.20
Potassium 7280 0.29 3150 0.30 8060 0.20
Copper 118 1.00 136 17.55 138 1.33
Sodium 2780 0.33 5120 0.36 3181 0.03
Tin 55.9 0.01 0.1 0 155 0.12
Thallium 0.01 0.01 0.25 0.01 0.25 0.04
Vanadium 58 7.67 7.8 0.50 36.4 0.02
Chlorides (mg=kg
ar
) 2.15 0.23 0.1 0 0.94 0.17
GASIFICATION OF NATURAL AND WASTE BIOMASS 637
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

The commercial potential of the process is interesting. The natural and waste
biomasses have all given acceptable performance, yielding a syngas of sufficient
quality for energy applications after an adequate downstream cleaning.
An olivine proved to be an interesting candidate to act as a bed catalyst for
the tar cracking reactions, even taking into account its low cost and outstanding
resistance to attrition in the fluidized bed. However, its catalytic action appears
remarkably reduced during gasification of waste biomass fuels, probably as a
consequence of the large fractions of metals in the fuel ash, which act as competitors
of the iron of olivine particles.
ACKNOWLEDGMENTS
The authors are indebted to CONAI, the Italian National Consortium
for Packaging, which financially supported a 3-year research program focused on
fluidized bed gasification of several alternative fuels. The biomass fuels were
provided by RILEGNO, the Italian Consortium for Wood Packaging Recycle.
The authors are also indebted to PhD students Maria Mallardo and Donato
Santoro, who performed the off-line analyses of the reported runs.
REFERENCES
Arena, U., and Mastellone, M.L. 2005. Fluidized pyrolysis and gasification of solid Wastes.
Proc. of Industrial Fluidization South Africa 2005, South African Institute of Mining
and Metallurgy, pp. 53–68.
Arena, U., and Mastellone, M.L. 2006. Fluidized bed pyrolysis of plastic wastes, chapter 16.
In Scheirs, J. and Kaminsky, W. (Eds.), Feedstock Recycling and Pyrolysis of Waste
Plastics, Wiley, Hoboken, NJ, pp. 435–474.
Arena, U., Romeo, E., and Mastellone, M.L. 2008a. Recursive operability analysis of a pilot
plant gasifier. J. Loss Prevention in the Process Industries,21, 50–65.
Arena, U., Zaccariello, L., and Mastellone, M.L. 2008b. Gasification of a plastic waste in a
fluidized bed of olivine. Proc. of CFB9, 9th Int. Conf. on Circulating Fluidized Beds,
Werther, J., Nowak, W., Wirth, K.-E., and Hartge, E.-U. (Eds.), pp. 691–696.
Arena, U., Zaccariello, L., and Mastellone, M.L. 2009. Tar removal during the fluidized bed
gasification of a plastic waste. Waste Management,29, 783–791.
Arena, U., Zaccariello, L., and Mastellone, M.L. 2010. Fluidized bed gasification of waste
derived fuels. Waste Management,30, 1212–1219.
Babu, S.P. 2006. Perspectives on biomass gasification: Thermal gasification of biomass. IEA
Bioenergy Agreement Task 33, Work Shop No. 1.
Baratieri, M., Baggio, P., Fiori, L., and Grigiante, M. 2008. Biomass as an energy source:
Thermodynamic constraints on the performance of the conversion process. Bioresource
Technology,99, 7063–7073.
Basu, P. 2006. Combustion and Gasification in Fluidized Beds. CRC Press, Boca Raton, FL.
Corella, J., Aznar, M.P., Gil, J., and Caballero, M.A. 1999. Biomass gasification in fluidised
bed: Where to locate the dolomite to improve gasification? Energy and Fuels,13,
1122–1127
Cotton, F.A., and Wilkinson, G. 1999. Advanced Inorganic Chemistry, 6th ed., Wiley,
Hoboken, NJ.
638 U. ARENA ET AL.
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010

Devi, L., Craje, M., Thune, P., Ptasinski, K.J., and Janssen, F.J.J.G. 2005a. Olivine as
tar removal catalyst for biomass gasifier: Catalyst characterization. Applied Catalysis,
294, 68–79.
Devi, L., Craje, M., Ptasinski, K.J., Janssen, F.J.J.G., van Paasen, S.V.B., Bergman, P.C.A.,
and Kiel, J.H.A. 2005b. Catalytic decomposition of biomass tars. Use of dolomite and
untreated olivine. Renewable Energy,30, 565–587.
Devi, L., Ptasinski, K.J., and Janssen, F.J.J.G. 2003. A review of the primary measures for
tar elimination in biomass gasification processes. Biomass and Bioenergy,24, 125–140.
Felder, R.M., and Rousseau, R.W. 2000. Elementary Principles of Chemical Processes, 3rd ed.,
Wiley, Hoboken, NJ.
Gil, J., Caballero, M.A., Martin, J.A., Aznar, M.P., and Corella, J. 1999. Biomass gasification
with air in a fluidized bed. Effect of the in-bed use of dolomite under different operating
conditions. Ind. Eng. Chem. Res.,38, 4226–4235.
Juniper Consultancy Services Ltd. 2001. Pyrolysis & Gasification of Waste. A Worldwide
Technology & Business Review, 2nd ed., Juniper Consultancy Services, Uley, England.
Kiel, J.H.A., van Paasen, S.V.B., Neeft, J.P.A., Devi, L., Ptasinski, K.J., Janssen, F.J.J.G.,
Meijer, R., Berends, R.H., Temmink, H.M.G., Brem, G., Padban, N., and Bramer,
E.A. 2004. Primary measures to reduce tar formation in a fluidised-bed biomass
gasifier. Report ECN-C–04–014, ECN. Petten, The Netherlands.
Malkow, T. 2004. Novel and innovative pyrolysis and gasification technologies for energy
efficient and environmentally sound MSW disposal. Waste Management,24, 53–79.
Mastellone, M.L., and Arena, U. 2008. Olivine as a tar removal catalyst during fluidized bed
gasification of plastic waste. AIChE Journal,54, 1656–1667.
Milne, T.A., Evans, R.J., and Abatzoglou, N. 1998. Biomass gasifier ‘tars’: Their nature,
formation and conversion. NREL Technical Report (NREL=TP-570-25357), National
Energy Laboratory, Golden, CO.
Narvaez, I., Orio, A., Aznar, M.P., and Corella, J. 1996. Biomass gasification with air in an
atmospheric bubbling fluidized bed. Effect of six operational variables on the quality
of produced raw gas. Ind. Eng. Chem. Res.,35, 2110–2120.
Pfeifer, C., Rauch, R., and Hofbauer, H. 2004. In-bed catalytic tar reduction in a dual
fluidized bed biomass steam gasifier. Ind. Eng. Chem. Res.,43, 1634–1640.
Rapagna
`, S., Jand, N., Kiennemann, A., and Foscolo, P.U. 2000. Steam-gasification of
biomass in a fluidised-bed of olivine particles. Biomass and Bioenergy,19, 187–197.
Zevenhoven, R., and Kilpinen, P. 2001. Control of Pollutants in Flue Gases and Fuel Gases.
A
˚bo Akademi University, Turku, Finland, available on www.hut.fi/rzevenho/gasbook
GASIFICATION OF NATURAL AND WASTE BIOMASS 639
Downloaded By: [EBSCOHost EJS Content Distribution - Superceded by 916427733] At: 08:55 23 June 2010