Bioresource Technology 194 (2015) 3642

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Gasication of agricultural residues in a demonstrative plant: Vine

pruning and rice husks
Enrico Biagini a,, Federica Barontini a,b, Leonardo Tognotti a,b
a
Dipartimento di Ingegneria Civile e Industriale (DICI), Universit di Pisa, L. Lazzarino, 56126 Pisa, Italy
b
Centro di Ricerca Interuniversitario Biomasse da Energia (CRIBE), Via vecchia di Marina, 6, 56122 Pisa, Italy

h i g h l i g h t s g r a p h i c a l a b s t r a c t

 Experimental gasication tests are

carried out with two biomass
residues.
 The gasier is a downdraft reactor at
a demonstrative scale (350 kWth).
 Syngas composition, pressure drops,
material and energy balances are
provided.
 Plant operability and performance
indexes are compared with previous
results.
 A reactor diagram is dened
considering density, size and shape of
feedstocks.

a r t i c l e i n f o a b s t r a c t

Article history: Tests with vine pruning and rice husks were carried out in a demonstrative downdraft gasier (350 kW),
Received 17 June 2015 to prove the reactor operability, quantify the plant efciency, and thus extend the range of potential
Received in revised form 3 July 2015 energy feedstocks. Pressure drops, syngas ow rate and composition were monitored to study the mate-
Accepted 4 July 2015
rial and energy balances, and performance indexes. Interesting results were obtained for vine pruning
Available online 8 July 2015
(syngas heating value 5.7 MJ/m3, equivalent ratio 0.26, cold gas efciency 65%, power efciency 21%),
while poorer values were obtained for rice husks (syngas heating value 2.53.8 MJ/m3, equivalent ratio
Keywords:
0.4, cold gas efciency 3142%, power efciency 1013%). The work contains also a comparison with pre-
Biomass fuels
Bio-energy
vious results (wood pellets, corn cobs, Miscanthus) for dening an operating diagram, based on material
Thermo-chemical plant density and particle size and shape, and the critical zones (reactor obstruction, bridging, no bed buildup,
Power production combustion regime).
Cogeneration process 2015 Elsevier Ltd. All rights reserved.

1. Introduction as a soil amendment, recovery heat for cogeneration), the low

Gasication with air in a downdraft reactor coupled to an inter-
nal combustion engine is one of the most attracting
biomass-to-energy applications on a small scale, because of the
higher power efciency with respect to any direct combustion
based system, the possibility of utilizing the by-products (biochar
Corresponding author. Tel.: +39 050 2217840.
E-mail address: e.biagini@ing.unipi.it (E. Biagini).
emissions due to the combustion of a gas instead of a solid. The
open issues concern the limitations on feedstock specications
(low moisture, narrow dimensional range, as discussed in Simone
et al., 2009; Martinez et al., 2012), the high level required in the
syngas cleaning (especially for tar and dust), and the disposal of
waste water and/or lter media. Further drawbacks derive from
issues related to biomass fuels, such as seasonal availability, logis-
tics, low energy density.

http://dx.doi.org/10.1016/j.biortech.2015.07.016
0960-8524/ 2015 Elsevier Ltd. All rights reserved.
 E. Biagini et al. / Bioresource Technology 194 (2015) 3642
It is important to prove the operability of the gasier with feed-
stocks other than the reference material (wood chips) in order to
extend the diffusion of this technology to cheap and widespread
agricultural by-products and food residues. The valorization of
these lignocellulosic residues may increase the economic values
of agricultural wastes and represent an opportunity for farmers.
As a matter of fact, many of them represent out-of-specication
fuels and are not suitable for the direct gasication, due to high
moisture or ash content, presence of nes or long particles.
The work includes the results of the regional research project
Bio-Power in Tuscany funded by Regione Toscana (Italy), aiming
at proving the feasibility of a distributed energy generation in
small plants that use biomass sources on a regional area. It is also
part of the programmed activities on agricultural and forest bio-
masses to be tested in the gasication plant at CRIBE
(Biomass-to-Energy Interuniversity Research Center).
Previous works with corn cobs revealed that good results (good
operability, high plant efciency) can be obtained, almost compa-
rable with those obtained with wood chips (Biagini et al., 2014a).
This work continues the investigation with out-of-specication
agricultural residues and compares the results with previous ones
in the same plant to extend the range of potential energy feed-
stocks and dene an operability diagram of the gasier (Biagini
et al., 2014b). Vine pruning and rice husks were studied: the for-
mer derives from wine industry, which generates large amounts
of residues on a regional scale and is a woody waste to be disposed;
the latter derives from the cultivation of rice and is known to have
a high ash content and low energy density. Potential availability of
vine pruning is estimated to be 42 kt/y in Tuscany, and raises to
540 kt/y in Italy, while the availability of rice husks is around
280 kt/y in Italy (data from ENAMA, 2011).
Many studies can be found for the conversion of rice residues to
energy (see for instance the review by Lim et al., 2012), dealing
with thermochemical and biochemical processes. More limited
investigations can be found for the utilization of vine pruning
due to the smaller amounts and local availability. Data on such bio-
fuels are rare for gasication in pilot/demonstrative scale plants
(Natarajan et al., 1998; Yin et al., 2002; Ramrez Behainne and
Martinez, 2014 for rice husks, Brito et al., 2014 for vine pruning),
while some more frequent studies in laboratory scale apparatuses
can be found (see for instance Gan et al. (2006) for vine residues,
Zhao et al., 2009 for rice husks). In this paper the results of the
experimental campaigns in a downdraft demonstrative gasier
are provided and compared with previous data in the same plant.
2. Methods
2.1. Fuel characterization
The activities of the plant were supported by the laboratory at
DICI (Department of Industrial and Civil Engineering of the
University of Pisa Italy), for the characterization of the solid fuels.
The moisture content and bulk density was evaluated according to
the European standard test methods EN 14774-1 and EN 15103,
respectively. Volatile matter (VM), xed carbon (FC) and ash con-
tent were determined by thermogravimetric (TG) analysis with a
TA Q-500 thermobalance. Ultimate analysis was carried out with
a LECO TruSpec CHN Elemental Analyzer, according to EN 15104
test method. A LECO AC-500 Isoperibol Calorimeter was used for
determining the heating value (HV), according to EN 14918 test
method. The particle size distribution was obtained according to
EN 15149-1.
The results of material characterization are listed in Table 1. It
can be noted that the moisture content is relatively low for both
feedstocks (below 20% wt, which is the recommended
37
specication). The ash content of rice husks is very high
(16.6% wt on a dry basis), with repercussion on the heating value
(15.6 against 18.1 MJ/kg on a dry basis for rice husks and vine
pruning, respectively). The bulk density is low for vine pruning
(260 kg/m3) and very low for rice husks (140 kg/m3). The particle
size distributions of both feedstocks are compared in Fig. 1 with
specications recommended for an optimal operability of the feed-
ing lines and gasier. As can be seen, vine pruning is formed of a
certain amounts of out-of-specication particles, both too large
(a difference of 4.5% with respect to the recommended size fraction
above 63 mm) and too ne (a difference of 24% for the range 3.15
8 mm, a difference of 11% for the range below 3.15 mm). Rice husks
are formed of out-of-specication particles, almost exclusively
below 3.15 mm.
2.2. Description of the gasication facility
The Gastone gasication plant is sited at CRIBE center (Pisa
Italy) and has a maximum thermal input of 350 kW, corresponding
to 80 kg/h of biomass with a moisture content of 15%. The reactor
is a downdraft type, with air entering just above the internal
restriction for the depression caused by the blower at the end of
the plant. During gasication tests, the amount of material inside
the reactor was maintained constant by means of a level control
(Reed and Das, 1988). Biochar was removed from the bottom of
the reduction zone and samples were collected for ofine analysis.
Particulate, tar and most water were removed from the syngas in
the cleanup line. Differential manometers gave the pressure drops
across the gasier bed and clean-up units to monitor the plant
operability. The syngas temperature was continuously measured
at the exit of the gasier. A ow-meter at the end of the cleanup
line measured the syngas production. A Gas Chromatograph
(micro-GC Agilent 3000) and a Fourier Transformed Infrared
Spectrometer (Bruker Tensor 37 FTIR spectrometer), positioned at
the end of the cleanup line, was used for online gas analysis.
More details on the equipment, diagnostics and analysis proce-
dures can be found elsewhere (Simone et al., 2012, 2013; Biagini
et al., 2014a).
3. Results and discussion
Four experimental tests were carried out (tests 1 and 2 with
vine pruning, tests 3 and 4 with rice husks) by regulating the
blower head at different levels, and thus obtaining different pres-
sure drops across the gasication bed (DP) and different air and
biomass ow rates. The measurements of the syngas ow rate,
DP and syngas temperature at the gasier exit are shown in
Fig. 2, where results for all tests are plotted consecutively, exclud-
ing the transitory periods required for achieving stationary
conditions.
Tests 1 and 2 with vine pruning differed in the syngas ow rate,
87 and 112 m3/h (mean values of the stationary period, as follow-
ing reported data), respectively. Correspondingly, the pressure
drops were relatively low (14 mbar) for test 1 and relatively high
(20 mbar) for test 2, respectively. These values are indeed accept-
able and comparable with typical operation of the gasier (gener-
ally DP are in the range 1040 mbar). However, it can be noted that
pressure drops are more irregular for test 2, i.e. for relatively high
biomass feeding rate.
Tests 3 and 4 with rice husks differed in the syngas ow rate, 87
and 119 m3/h, respectively. In both cases, the pressure drops were
extremely low (13 mbar). These values are judged too small for
typical operation of the gasier, and imputable to the difculty
in building the xed bed on the bottom grid of the reactor. This
38 E. Biagini et al. / Bioresource Technology 194 (2015) 3642

Table 1
Properties of vine pruning and rice husks (VM: volatile matter; FC: xed carbon; LHV: low heating value; ar: as received; daf: dry and ash free).

Moisture VM FC Ash C H N LHV Bulk density

% wt % wt % wt % wt % wt % wt % wt MJ/kg kg/m3
ar dry dry dry daf daf daf dry ar
Vine pruning 17.6 80.84 16.54 2.62 50.84 5.82 0.88 18.1 260
Rice husks 12.5 67.95 15.45 16.60 49.44 6.25 0.54 15.6 140

1 inconvenience is related to the properties of the material (size and

density) and the explanation is discussed below.
0.8
As for the operability, tests with vine pruning were character-
maximum value of ized by high bed permeability, leading to relatively high syngas
recommended
mass fraction

specification ow rate. Some operating problems were observed, especially dur-

0.6 rice husks vine pruning
ing test 2 (after a signicant stationary period), for the formation of
bridges in the upper part of the reactor (revealed at the end of the
0.4 test). This inconvenience was ascribed to the presence of some
large particles (sticks) in the feedstock and the low density of the
material, favoring the deposition of the feeding stratus onto the
0.2
previous ones, instead of breaking down the formed bridge struc-
ture. Minor effects were observed during test 1, probably for the
0 lower biomass feeding rate. Some investigation carried out in the
<3.15 3.15-8 8-63 > 63 late period of test 2 (not shown here) conrmed that benecial
particle size range (mm) improvements could be obtained by increasing the frequency of
the feeding system, reducing the amount of biomass per each load,
Fig. 1. Particle size distribution of vine pruning (black bars) and rice husks (grey and increasing the vibration provided by the electric shaker on the
bars), compared with specication on maximum mass fractions (dotted bars) for
each particle size range.
external reactor wall. It is noted that a more stringent sieving pre-
treatment (excluding sticks and large particles) of the feedstock
might have resulted in a less difcult operation of the plant.
Rice husks had good handling properties and gave no signicant
problems in the screw conveyors. Some problems were observed in
the gasier, as the small dimension of the particles and the low
density of the material were unsuitable to buildup the gasication
bed in the reduction zone. These problems required to nd operat-
ing parameters (feeding frequency, level control, vibration, rota-
tion of the solid discharge system) completely different from the
reference case and assure stable operations. The tests were charac-
terized by extremely high permeability, leading to extremely low
values of the pressure drop (as seen in Fig. 2b), which proved the
scarce consistency of the gasication bed buildup.
Rapid gas analyses were carried out with the micro-GC, for
monitoring the syngas composition during the entire duration of
each test. The measurements of tests 1 and 4 are shown in
Fig. 3a and b for the gasication of vine pruning and rice husks,
respectively. It can be seen that syngas composition is quite stable
for both cases shown in the graphs, as well as for the other tests
(not shown here) that have similar trends. Based on the composi-
tion of measured gaseous compounds, the lower heating value
(LHV) of syngas is calculated and shown in the graphs.
Comparing the results for vine pruning and rice husks, it is evident
that syngas from the former has a LHV between 5 and 6 MJ/m3,
which is a typical value for biomass gasication in air. The LHV
of syngas from rice husks is denitely lower, that is between 2
and 4 MJ/m3. This is due to the greater content of nitrogen in tests
3 and 4, and this is due to the equivalent ratio (ER, expressed as the
ratio between the actual oxygen to the stoichiometric oxygen
required for complete combustion of the material), which exceeds
the typical values, as discussed below.
It is worth noting that the syngas composition as well as the
lower heating value was characterized by some uctuations, also
during stable operations of the gasier, likely due to the discontin-
uous feeding of the biomass and local variations in the
air-to-biomass ratio. As observed in previous works (Simone
et al., 2013; Biagini et al., 2014a), linear correlations between CO
Fig. 2. Syngas ow rate (Q), pressure drops across the reactor (DP) and syngas
temperature at the exit of the gasier (T) during the gasication tests of (a) vine
and CO2, and between C2H4 and CH4 can be observed. Fig. 4 shows
pruning and (b) rice husks. such correlations for the gasication of vine pruning and rice
 E. Biagini et al. / Bioresource Technology 194 (2015) 3642 39

80 H2 N2 CH4 CO CO2 C2H4 LHV (right axis) 8

Volumetric fraction (%)

30
test 1 test 2
60 6

LHV (MJ/m3)
40 4 20

CO (%vol)
20 2
10

0 0
0 50 100 150
(a) Time (min)
0
0 5 10 15 20

80 8
(a) CO2 (%vol)
H2 N2 CH4 CO CO2 C2H4 LHV (right axis)

1
test 1 test 2
Volumetric fraction (%)

60 6
LHV (MJ/m3) 0.8

C2H4 (%vol)
40 4 0.6

0.4
20 2

0.2

0 0
200 250 300 350 400 0
(b) Time (min) 0 1 2 3 4
(b) CH4 (%vol)
Fig. 3. Syngas composition measured with micro-GC and calculated lower heating
value (LHV) during (a) gasication test 1 with vine pruning and (b) gasication test 30
test 3 test 4
4 with rice husks.

husks. Both correlations can be explained with the oxygen avail-

20
ability. When more oxygen is available, CO2 is favored with respect
CO (%vol)

to CO (see the case of test 4, which has a higher ER, as shown

below, with respect to the case of test 3). CH4 and C2H4 are propor-
tionally related, and this is due to the fact that they are products of 10
primary pyrolysis, so that when more oxygen is available, they are
more efciently oxidized in the oxidation zone.
Mean values of syngas composition and ow rate can be calcu-
lated for each test during the stationary period and results are 0
listed in Table 2, along with calculated mean values of performance 0 5 10 15 20
indexes. The biomass feeding rate is calculated as a mean value of (c) CO2 (%vol)
the material loads, which were weighted during each test. The val-
ues are shown in Table 2 on a dry basis. Air ow rate entering the 1
test 3 test 4
gasier is estimated from the balance on nitrogen, which was mea-
sured in the syngas. The equivalent ratio is then calculated from 0.8
the actual oxygen and the one required for complete combustion,
which is calculated from the ultimate analysis of the biomass
C2H4 (%vol)

0.6
(Table 1). Values in the range 2527% are estimated for gasication
tests with vine pruning, and these are in the typical range of bio-
mass gasication (generally between 25% and 30%). Results for 0.4
tests 3 and 4 exceed this range (38% and 41%, respectively), and
conrm the observation on the high nitrogen content and low 0.2
heating value of the syngas from rice husks gasication.
The specic syngas productivity is in the typical range (2
0
2.5 m3/kg) of biomass gasication, with negligible difference
0 1 2 3 4
between both feedstocks. It should be noted that this parameter
is calculated as volume of syngas produced from the gasication
(d) CH4 (%vol)

of 1 kg of dry biomass. In this case, considering the high ash con- Fig. 4. Correlations between the gaseous species measured in the syngas during the
tent of rice husks, the parameter can be expressed on a dry and gasication of (a and b) vine pruning and (c and d) rice husks.
40 E. Biagini et al. / Bioresource Technology 194 (2015) 3642

Table 2 Woody feedstocks (chips and pellets) as well as corn cobs

Results of the gasication tests. exhibited good handling properties and allowed stable feeding
Test (1) Test (2) Test (3) Test (4) and continuous operation of the gasier. Rice husks had good han-
Vine Vine Rice Rice dling properties, but some operating problems arose as the small
pruning pruning husks husks size of the particles and low density of the material were unsuit-
Syngas composition (%vol) able to buildup the gasication bed in the reduction zone.
N2 45.07 46.36 60.50 66.54 Modications to the operation procedures adopted to assure stable
H2 17.06 18.44 7.97 5.78
CO 21.74 21.31 20.04 12.27
operation with this feedstock might have increased the
CO2 13.02 11.32 10.58 14.55 air-to-biomass ratio out of typical ranges. Minor problems in the
CH4 2.55 2.06 0.69 0.61 feeding lines were observed with vine pruning, along with more
C2H4 0.42 0.42 0.20 0.22 consistent ones in the operation of test 2 (at high biomass feeding
C2H6 0.06 0.04 0.01 0.01
rate). The formation of bridges inside the gasier stopped the
C2H2 0.06 0.03 0.01 0.01
biomass outow, and thus the feeding of the material. This prob-
Syngas LHV (MJ/m3) 5.82 5.70 3.76 2.53
lem can be related to the presence of large particles (especially
Syngas ow rate (m3/h) dry 87 112 87 119
Biomass feeding rate (kg/h) 44 52 50 63 long pieces) and the low bulk density of the biomass. More dra-
dry matic operations were experienced with Miscanthus (results not
ER (%) 25.1 27.4 38.2 41.4 shown here), having more than 90% of out-of-specication large
Specic productivity (m3/ 2.04 2.17 2.07 2.26 particles, for which stable operations were not achieved due to
kg)
the frequent formation of bridges impeding the continuous feeding
CGE (%) 64.2 66.8 41.6 30.6
Potential power production 45 57 29 27 of the material.
(kW) The pressure drops across the gasication bed are a measure of
Potential plant efciency 20.5 21.4 13.3 9.8 the acceptable owability: small pressure drops can be the symp-
(%)
tom of signicant by-pass in the reduction zone or inconsistency of
the bed buildup (giving high air-to-biomass ratios with risk of
shifting from gasication to combustion regime), while high values
ash free basis. The results for vine pruning vary to 2.10 and of pressure drops can be due to the obstruction of the reduction
2.23 m3/kg daf (test 1 and 2, respectively), while for rice husks they zone or inefciency of the solid discharge system. Typical values
become 2.48 and 2.71 m3/kg daf (test 3 and 4, respectively). This are in the range 1050 mbar, values between 20 and 30 mbar were
comparison gives a further evidence that a high air-to-biomass registered for tests with reference material (wood chips).
ratio is established during the gasication of rice husks. Fig. 5 shows most of results obtained in the same gasier with
The cold gas efciency (CGE) is dened as the ratio of the ther- different feedstocks. The abscissa represents the mean value of the
mal power of the produced syngas (given by multiplying the syn- syngas ow rate, while the ordinate represents the range of pres-
gas ow rate by its LHV) to the thermal power of the feedstock sure drops DP measured during the corresponding stationary per-
(given by multiplying the biomass feeding rate by its LHV). The iod of gasication. Tests with wood pellets showed the highest
results for test 1 and 2 (see Table 2) are in the typical range of bio- values of DP (2045 mbar) at relatively low gas ow rate (80
mass gasication (6080%) (Erlich and Fransson, 2011; Martinez 120 m3/h), and the greatest variations at a xed value of the gas
et al., 2012; Pathak et al., 2008; Zainal et al., 2002). Vice versa, ow rate. The values of DP were lower for tests with vine pruning
the values for tests 3 and 4 (3042%) are well below this range. (1228 mbar) in the same range of gas ow rate, although the vari-
The potential power production and plant efciency can be cal- ations at a xed value of the gas ow rate were signicant. DP for
culated according to the assumptions made in Biagini et al. rice husks gasication was extremely low, and this was imputed to
(2014a). In case the syngas is burnt in an internal combustion the different regime (ER around 0.4 instead of 0.250.3 for typical
engine (with a realistic efciency of 32%) for the production of biomass gasication). Values of DP comparable with those of wood
electricity, the gross electrical efciency of the plant (dened as pellet were found for corn cobs, although the gas ow rate was
the ratio of the electricity production to the thermal power of greater (120160 m3/h), and variations at a xed value of the gas
the feedstock) can be compared for the four tests. As expected from ow rate were moderate.
previous results on the composition and LHV of the syngas, the val-
ues obtained for tests with vine pruning are in line with typical val-
ues of biomass gasication (2025%, the highest values are
expected for the reference feedstock), while those for tests with
50
rice husks are well below (1013%). The results for rice husks are
only potential as the efcient combustion of the syngas from rice wood pellet
husks (having such a low LHV) should be proved. 40
A comparison of the results obtained in this work with previous
ones in the same plant (with corn cobs, since Biagini et al., 2014a;
DP (m bar)

30
with wood pellet, since Simone et al., 2012) is made here for draw-
corn cobs
ing up some guidelines on the gasication of out-of-specication
feedstocks in the downdraft reactor. No consideration will be made 20
wood chips
on moisture content, as all tested feedstocks were in the optimal (ref.mat.)
vine prunings
range (<20% wt). 10
The rst aspect taken into account concerns reactor operability. rice husks
The reliability of the gasier operation depends on the handling
0
and owability of the biomass in the gasier. There is no standard
0 20 40 60 80 100 120 140 160 180
way to predict how a biomass ows inside a xed bed gasier, as
Syngas flow rate (m3/h)
this depends on particle size/shape distribution and bulk density,
and the occurrence of problems is related to the gasier Fig. 5. Pressure drop range versus syngas owrate during gasication tests with
conguration. different feedstocks.
 E. Biagini et al. / Bioresource Technology 194 (2015) 3642
Different strategies were adopted to maintain the pressure
drops in the acceptable range, by adjusting the feeding procedure,
level control, vibration, discharge system. However, most issues
were indeed related to the characteristics of the feedstock (particle
size/shape, density, mechanical resistance) and a good pretreat-
ment could remove most critical ones. Wood pellets (having uni-
form size and high density) were likely to break up in the
reduction zone of the gasier due to mechanical and thermal stres-
ses, thus turning into dust and giving a high resistance to the gas
ow, while the variations in the pressure drops could be caused
by the substitution of dusty char with fresh (and heavy) material,
which retained the original morphology. Therefore, high and
unstable pressure drops, with moderate syngas productivity,
should be expected during the gasication of pelletized biomass
in the downdraft gasier. Tests with corn cobs were characterized
by relatively low permeability, leading to high pressure drops and
high syngas ow-rate, and this could be due to the signicant frac-
tion of ne particles contained in the feedstock (exceeding the
specication required with a difference of 13% in the fraction of
nes). Tests with vine pruning were characterized by high bed per-
meability, leading to relatively high syngas ow-rate and low pres-
sure drop, with the formation of some bridges at high biomass
feeding rates. In this case the presence of long particles was more
critical than the signicant presence of nes (see Fig. 1). Finally,
tests with rice husks were characterized by extremely high perme-
ability, leading to extremely low pressure drops, that proved the
scarce consistency of the gasication bed buildup.
As for the syngas composition, the average values obtained dur-
ing gasication tests with ER in the typical range (2530%) were
quite similar, with a LHV in the range 56 MJ/m3, representing a
good result comparable with values found in literature (Martinez
et al., 2012). In the case of rice husks, the syngas had a LHV below
4 MJ/m3, which can be imputed to the high ER of those tests.
In all tests, the syngas composition as well as the lower heating
value was characterized by continuous uctuations also during
stable operations of the gasier. Correlations of some species were
conrmed: positive deviations of carbon monoxide corresponded
to negative deviations of carbon dioxide, and vice versa; a linear cor-
respondence was found between methane and ethylene (primary
products of pyrolysis). These correlations can be explained by instan-
taneous variations of the air-to-biomass ratio (and thus of ER) due to
the discontinuous feeding of the biomass. Comparing the case of rice
husks with the other tests at lower ER, lower values of CH4 and C2H4
can be observed (see Table 2) and this could be due to both the high
oxygen availability and the high temperature achieved in the reactor
(as already observed in Fig. 2), which likely favored the reactions of
oxidation and cracking of pyrolysis products.
As for the energy balance, the cold gas efciency was in the typ-
ical range of biomass gasication (6080%) for tests with ER
between 25% and 30%. Tests with wood pellets, corn cobs and vine
pruning showed values in the range 6570%, while those with rice
husks showed lower values. Correspondingly, the potential plant
efciency was between 20% and 22% for residues (corn cobs and
vine pruning) and wood pellets, while values below 14% were
assessed for gasication of rice husks.
Summarizing, the gasication plant demonstrated a versatile
operating range, with biomass feeding rates between 40 and
85 kg/h, with different out-of-specication feedstocks. Most crucial
limits were dictated by moisture content (maximum acceptable
value 20% wt) and particle size distribution (small content of ne
particles, no particles with a size greater than 100 mm). A detailed
characterization of the fuel (proximate and ultimate analysis, heat-
ing value) is thus important to evaluate the suitability of a biomass
for the downdraft gasier. A low moisture content is crucial to
assure an adequate heating value of the feedstock, provide the suf-
cient energy and reach high temperatures in the gasier.
41
Another important requirement was dictated by the
uid-dynamics inside the reactor. It is necessary to determine
whether the feedstock can provide a sufcient amount of xed car-
bon, which is related to the possibility of forming a char bed and
providing carbon for the gasication reactions, mechanically sup-
port the feeding material and assure the air ow. These two last
points are important for the proper operation of the gasier, there-
fore the feedstock should be characterized also for its size and
shape distribution, bulk density, thermal and mechanical consis-
tency. Fine particles reduce the void fraction, by reducing the bed
permeability to air and giving a low syngas production and even
reactor obstruction. Long particles (sticks > 100 mm) compromise
the feeding system and form bridges inside the reactor, and thus
increase the bed permeability, favoring a combustion regime.
Bridging is likely to occur in the feeding hopper, where long parti-
cles may form a stable layer for the accumulation of further mate-
rial. In addition, the lower the density of the biomass, the higher
the tendency to form bridges, as the weight of the biomass above
the bridge is not enough to break down the structure (Mattsson
and Kofman, 2002).
Besides the feedstock characterization, it is necessary to moni-
tor the pressure drop across the bed during operations. High pres-
sure drops indicate a low bed permeability, with a consequent
reduction of the syngas production or even obstruction of the gasi-
er. On the other hand, low pressure drops indicate a too high bed
permeability, or presence of by-pass channels, with a poor mixing
of the gas and solid phases, resulting in a low biomass conversion
and residual pyrolysis products (tar) in the syngas. Low pressure
drops may also indicate a low biomass content within the gasier
(with scarce consistency of the xed bed), which favors a combus-
tion regime (as in the case of rice husks). As a consequence, the
heating value of the syngas is low, as well as the plant efciency.
All results from tests with out-of-specication feedstocks are
used to evaluate the range of operability of the plant and the pro-
cedures (setup of feeding devices, vibration at the reactor wall, fre-
quency of solid discharge system) to deal with reactor problems
(obstructions, combustion regime, bridging). Thanks to the results
obtained in this work and previous ones, some specications are
dened also for material density and particle shape. The represen-
tation shown in Fig. 6 is semi-quantitative as the limits depends on
the reactor conguration, efcacy of pretreatments and technical
adjustments for assuring the continuous operation of the plant.
The critical zones are highlighted with the possible inconveniences
(e.g., low conversion, bridge formation, reactor obstruction, bed
combustion). Every feedstock is represented with the proper bulk
density and a size and shape factor (SSF) dened with two contri-
butions. SSF for ne particles is dened as:
1000
HIGH PRESSURE DROPS corn cobs wood pellets
low syngas production
rice husks vine pruning
reactor obstruction
800 wood chips miscanthus
density (kg/m3)
600
400
200
0
0 50 100 150 200
LOW PRESSURE DROPS BRIDGING TENDENCY
low syngas HV, combustion regime SSF feeding obstruction
Fig. 6. Specications on the morphological and density properties of the feedstock
for continuous operation in the downdraft gasier.
42 E. Biagini et al. / Bioresource Technology 194 (2015) 3642
SSF fines 100  mf 1 Acknowledgement
where mf is the mass percentage of nes exceeding the specica-
tions. SSF for large particles is dened as:
SSF sticks 100 ms AR 2 References
where ms is the mass percentage of sticks exceeding the specica-
tions and AR the aspect ratio, given by the ratio of length and width
of the particles.
As can be seen in Fig. 6, the reference material is in the center of
the graph. Rice husks have low density and SSF, and therefore are
positioned in the low left corner, where low pressure drops and
combustion regime are expected. The wood pellets are positioned
in proximity of the high left corner of the graph, with risk of high
pressure drops and reactor obstruction. This occurrence worsens
when considering the tendency of pellets to turn into dust.
Miscanthus is in the bridging zone (low right corner) due to the
presence of numerous long sticks and the low density. Vine prun-
ing contains both nes and sticks, but only the presence of these
latter determines the risk of bridging. Corn cobs have a quite uni-
form size and shape so minor risks should be expected.
For feedstocks and tests within the zone of continuous opera-
tion, the use of out-of-specication materials can give stable gasi-
cation and plant efciency was proved to be just few points lower
than that with reference material (that is 2022% instead of 22
25%). Results obtained here do not preclude the possibility of an
efcient gasication also with rice husks, Miscanthus and further
problematic feedstocks. Of course, different reactor congurations
can be more successfully used (Natarajan et al., 1998; Yin et al.,
2002; Brito et al., 2014; Risberg et al., 2014; Makwana et al.,
2015). However, such materials could be used also in simple down-
draft gasiers after optimized pretreatments, such as densication
for dusty low-density materials (Yoon et al., 2012) and specic
chipping for sticks and long particles, which are actually required
in order to fall out of the critical zones of Fig. 6. Further investiga-
tion and model interpretation, required for giving a more quantita-
tive denition of such zones, are in progress.
4. Conclusions
The gasication of vine pruning and rice husks was studied in a
downdraft gasier (350 kW). The operation was proven to be con-
tinuous after minor modications of the operating parameters. The
process efciency was high for vine pruning (2022%) and rela-
tively low for rice husks (1014%), due to the high equivalent ratio
and low heating value of syngas produced. Comparison of the
results obtained with other biomasses (wood pellets, corn cobs,
Miscanthus) allowed to draw the reactor operating diagram, based
on density, size and shape of the feedstocks, and remark the critical
zones (obstructions, bridging, combustion regime).
The work is partly funded by the Regional project Biopower in
Tuscany (POR CReO FESR 20072013, Regione Toscana Italy).
Biagini, E., Barontini, F., Tognotti, L., 2014a. Gasication of agricultural residues in a
demonstrative plant: corn cobs. Bioresour. Technol. 173, 110116.
Biagini, E., Barontini, F., Tognotti, L., 2014b. Gasication of agricultural residues in a
demonstrative plant. Chem. Eng. Trans. 37, 151156.
Brito, P.S.D., Oliveira, S.O., Rodrigues, L.F., 2014. Energy valorization of solid vines
pruning by thermal gasication in a pilot plant. Waste Biomass Valorization 5,
181187.
ENAMA (Ente Nazionale per la Meccanizzazione Agricola), Biomasse ed energia
Disponibilit delle biomasse, 2011. Document in Italian available at: <http://
www.enama.it/it/pdf/biomasse/Studio/p1c2.pdf> (last accessed June 2015).
Erlich, C., Fransson, T.H., 2011. Downdraft gasication of pellets made of wood,
palm-oil residues respective bagasse: experimental study. Appl. Energy 88,
899908.
Gan, J., Al-Kassir Abdulla, A., Cuerda Correa, E.M., Macas-Garca, A., 2006.
Energetic exploitation of vine shoot by gasication processes. A preliminary
study. Fuel Process. Technol. 87, 891897.
Lim, J.S., Manan, Z.A., Wan Alwi, S.R., Hashim, H., 2012. A review on utilisation of
biomass from rice industry as a source of renewable energy. Renew. Sustain.
Energy Rev. 16, 30843094.
Makwana, J.P., Joshi, A.K., Athawale, G., Singh, D., Mohanty, P., 2015. Air gasication
of rice husk in bubbling uidized bed reactor with bed heating by conventional
charcoal. Bioresour. Technol. 178, 4552.
Martinez, J.D., Mahkamov, K., Andrade, R.V., Silva Lora, E.E., 2012. Syngas production
in downdraft biomass gasiers and its application using internal combustion
engines. Renew. Energy 38, 19.
Mattsson, J.E., Kofman, P.D., 2002. Method and apparatus for measuring the
tendency of solid biofuels to bridge over openings. Biomass Bioenergy 22, 179
185.
Natarajan, E., Nordin, A., Rao, A.N., 1998. Overview of combustion and gasication of
rice husk in uidized bed reactors. Biomass Bioenergy 14, 533546.
Pathak, B.S., Patel, S.R., Bhave, A.G., Bhoi, P.R., Sharma, A.M., Shah, N.P., 2008.
Performance evaluation of an agricultural residue-based modular throat-type
down-draft gasier for thermal application. Biomass Bioenergy 32, 7277.
Ramrez Behainne, J.J., Martinez, J.D., 2014. Performance analysis of an air-blown
pilot uidized bed gasier for rice husk. Energy Sustainable Dev. 18, 7582.
Reed, T.B., Das, A., 1988. Handbook of Biomass Downdraft Gasier Engine Systems.
Biomass Energy Foundation Press, Golden, Colorado.
Risberg, M., Ohrman, O.G.W., Gebart, B.R., Nilsson, P.T., Gudmundsson, A., Sanati, M.,
2014. Inuence from fuel type on the performance of an air-blown cyclone
gasier. Fuel 116, 751759.
Simone, M., Guerrazzi, E., Biagini, E., Nicolella, C., Tognotti, L., 2009. Technological
barriers of biomass gasication. Int. J. Heat Technol. 27, 127132.
Simone, M., Barontini, F., Nicolella, C., Tognotti, L., 2012. Gasication of pelletized
biomass in a pilot scale downdraft gasier. Bioresour. Technol. 116, 403412.
Simone, M., Barontini, F., Nicolella, C., Tognotti, L., 2013. Assessment of syngas
composition variability in a pilot-scale downdraft biomass gasier by an
extended equilibrium model. Bioresour. Technol. 140, 4352.
Yin, L.Y., Wu, C.Z., Zheng, S.P., Chen, Y., 2002. Design and operation of a CFB
gasication and power generation system for rice husks. Biomass Bioenergy 23,
181187.
Yoon, S.J., Son, Y.I., Kim, Y.K., Lee, J.G., 2012. Gasication and power generation
characteristics of rice husks and rice husk pellet using a downdraft xed-bed
gasier. Renewable Energy 42, 163167.
Zainal, Z.A., Rifau, A., Quadir, G.A., Seetharamu, K.N., 2002. Experimental
investigation of a downdraft biomass gasier. Biomass Bioenergy 23, 283289.
Zhao, Y., Sun, S., Hongming Tian, H., Qian, J., Su, F., Ling, F., 2009. Characteristics of
rice husk gasication in an entrained ow reactor. Bioresour. Technol. 100,
60406044.