You are on page 1of 9

b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1

Available online at www.sciencedirect.com

http://www.elsevier.com/locate/biombioe

Influence of feeding ratio on steam gasification


of palm shells in a rotary kiln pilot plant.
Experimental and numerical investigations

P. Iovane*, A. Donatelli, A. Molino


ENEA, Italian National Agency for New Technologies, Energy and Sustainable Economic Development,
SS 106 Jonica km 419.500, 75026 Rotondella (MT), Italy

article info abstract

Article history: This paper presents experimental and numerical results on steam gasification of palm
Received 29 October 2012 shells in a rotary kiln pilot plant. Both the process performance and gas features have been
Received in revised form evaluated varying the steam-biomass ratio (SBR), defined as the mass of steam divided by
22 May 2013 the mass of palm shells. First, some experimental tests have been performed. Then, the
Accepted 24 May 2013 obtained experimental results have been used to verify the consistency of a numerical
Available online 25 June 2013 model developed with the aid of the commercial code ChemCAD.
Numerical results showed that the gas energy content decreases as the SBR increases as
Keywords: well, achieving a maximum value for SBR ¼ 0.6 that produced a gas which volumetric
Biomass composition N2 free is H2 ¼ 40.4%, CO ¼ 24.1%, CO2 ¼ 21.7%, CH4 ¼ 12.2%, C2H4 ¼ 1.7% and
Palm shells in correspondence of which the lower heating value (LHV) is equal to 12 MJ m3 in normal
Steam gasification conditions. SBR values higher than 0.6 do not produce a further increase of the gas yield,
Rotary kiln rather require a greater amount of input energy for heating the steam from the room
Syngas temperature to the process temperature.
Bioenergy ª 2013 Elsevier Ltd. All rights reserved.

1. Introduction carried out since antiquity is biomass which can be converted


through different physical processes (grinding and drying),
Many experts and decision markers agree that many envi- biochemical processes (as alcoholic fermentation and anaer-
ronmental problems will and has occurred because of climate obic digestion), and finally thermochemical processes carried
change. According to the Fourth Assessment Report (AR4) of out at high temperatures [3e9].
United Nation Intergovernmental Panel on Climate Change Within the thermochemical procedures three processes
(IPPC), released on 17 December 2007, observed global warm- can be distinguished, namely combustion, pyrolysis and
ing over last 50 years is likely due to increase of greenhouse gasification. Gasification, in particular, is a thermal process
gas emission such as CO2, coming from the use of fossil fuel that can convert carbonaceous materials, such as organic
and identified as the primary cause of global increment in waste or biomass, into CO and H2 with a controlled amount of
temperature [1,2]. Thus, the substitution of fossil fuel by oxidizing agent as oxygen, steam, CO2 or mix. The resulting
renewable energy source appears to be the best and necessary gas mixture, also called synthesis gas or syngas, is a fuel able
alternative to reduce emission of greenhouse gases and to to produce energy (bioenergy) in internal combustion engines,
promote greater energy efficiency. A renewable energy source turbines or fuel cells [10e12] or as starting point for the

* Corresponding author. Tel.: þ39 0835974423; fax: þ39 0835974284.


E-mail address: pierpaolo.iovane@enea.it (P. Iovane).
0961-9534/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biombioe.2013.05.025
424 b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1

Nomenclature B correction factor [a dimensional]


S slope of the kiln [ ]
SBR mass steam divided by mass of palm shells ratio
D kiln diameter [m]
[kg kg1]
n angular velocity [min1]
LHV lower heating value [MJ kg1]
DHr reaction enthalpy [kJ kg1]
HHV higher heating value [MJ kg1]
T Temperature [ C]
t solid residence time [min]
G mass flow rate [kg h1]
q response angle of matrix [ ]
P power from combustion of syngas [kW]
L length of the kiln [m]

production of chemicals (biofuels) [13]. Steam gasification is researchers have studied this potential and evaluated the
one of the promising processes among all biomass conversion feasibility of biomass power plants in terms of technology
ones producing a gas with high H2 concentration and a ideal availability and economic feasibility [38e41]; the scenario of
[H2]/[CO] ratio for the synthesis of methanol and Fischer- installing of new Combined Heat and Power plants yielded a
Tropsch synthesis oil (about 2) [13e18]. number of benefits in terms of net energy savings, net avoided
This process has several advantages over traditional CO2 emission, and profits [42].
combustion: a limited formation of dioxins and of nitrous and In some cases, palm shells are used as the source of energy
sulfur oxides; a strong reduction in the process gas volume; for the processing mill itself which produce oil palm to
production of an energy carrier that can be integrated with generate heat and electricity via combustion cofiring [43].
combined cycle turbines or reciprocating engines; a reduced However, this is not always practical due to the high moisture
amount of secondary wastes; and the possibility to apply the content biomass and the huge amount of energy required for
process at a smaller scale [19,20]. On the other hand, during complete combustion, thus reducing the energy efficiency.
gasification, tars, heavy metals, halogens and alkaline com- Substantial modification of the furnaces of the mills is needed
pounds are released within the gas and can cause environ- if the conventional fuel is to be replaced by palm shells.
mental and operational problems so it is important an There are currently no regulations for the management of
improved syngas cleaning, able to meet defined specifications palm residues except for the general ban on open-air burning
[21e25]. as stipulated in 1998 amendments to the Environment Quality
The biomass gasification technology is still not commer- Act of 1974. The development of such biomass plants can
cialized in full scale but several demonstration plants are present an opportunity to produce bioenergy and will have a
under way, and hence studies aimed at developing energy- positive effect on the waste disposal problems faced by
efficient, economically viable and resource-efficient systems various mills, diverting away from incineration and reducing
are necessary. the amount of the residues used as mulch [44].
Typically, a gasification plant consists of a gasifier unit, a One advantage would be to enable discarded and/or excess
gas cleaning system and an energy recovery system. Gasifi- residues for stand-alone power generation via small-scale
cation reactors can be basically classified as fixed beds, flu- gasification in the nearby located villages in Malaysia [45].
idized beds, rotary kilns or entrained beds. In spite of the high potential of palm biomass residues too
Rotary kilns has some advantages as good mixing capa- few researches have been published regarding gasification of
bility, ability to use different materials and efficient heat this material in particular on pilot scale. Within this frame-
transfer; the indirect heat rotary meets successfully a bundle work, the current research aims to investigate the gasification
of standard gasifier specifications [26], especially for gasifying of palm shells in a rotary kiln pilot plant. In particular, the
moist materials [27]. influence of the feeding ratio between the gasifying agent
However, the cases of experimental tests of steam gasifi- (steam) and the material to be gasified on the performance of
cation of biomass in rotary kiln reactors are not abundant in the process, in terms of amount of gas produced and its
the literature. They refer basically to the case of steam gasi- composition and content of energy, was examined. The
fication of lignite briquettes [28], waste wood [29] and poplar experimental results were compared to an equilibrium
[30]. simulation model.
Among all biomass, palm can be a very promising alter-
native source of renewable energy being the dominant agri-
cultural crop in Malaysia with more than 3.88 million hectares 2. Experimental work
of land under cultivation [31].
Every year, the oil palm industry produces more than one 2.1. Feedstock and characterization
hundred million tonnes of residues worldwide and 1 ha of oil
palm plantation can produce about 50e70 tonnes of biomass The material chosen for experiments has been palm shells, a
residues [32,33]. A large quantity of this biomass are generally biomass residue obtained from extraction of the oil palm
dumped in open areas, left to rot, or disposed off in open (Elaeis guineensis) from empty fruit bunch coming from the
burning, generating pollutant gases [34]. Thus, Malaysia has crop of 2011executed in a mill factory in Kuala Langat located
the potential to be one of the major contributors of renewable in the state of Selangor in Peninsular Malaysia. This process
energy in the world via palm biomass [35e37]. Some consists of four major steps: sterilization, stripping, screw
b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1 425

pressing and purification [46]. The sample, as received, is alkaline solution to remove acid gas. On both the quencher
about of 1e2 cm diameter particle size and, before feeding the and the scrubber there are two recirculation lines that pump
reactor via a screw feeder system for all experimental test, it the water and alkaline solution, respectively.
has been dried for 24 h at 105  C and stored under dry condi-
tions. To have a knowledge on the quantitative energy content 2.2.4. Procedure
of the biomass fuel, ultimate and proximate analysis were Palm shells were loaded into the feed hopper and a measured
carried out on palm shell according to ASTM Standards (ASTM quantity of water was introduced into the tank of the
D5373-08 for ultimate analysis, ASTM D3175-77 and ASTM quencher. The reactor was electrically heated to 850  C in N2
D2866-83 for proximate analysis) with an elementary analyzer atmosphere, then the steam was fed into the reactor. The
EA 1110 CHNS-O (ThermoQuest) and a thermo-balance 2950 steam was first produced into a heat exchanger by the com-
(TA Instruments). The results are shown, with the corre- bustion of LPG and of syngas and then it was sent to the
sponding higher heating value (HHV) (ASTM D5865) obtained reactor. The steam mass flow-rate (into the range
with calorimetric bomb IKA C5000, in Table 1. 2e3.5 kg h1) has been controlled by a valve and a volumetric
Between palm biomass and wood there is no significant counter, directly posed on the water adduction circuit, for
difference in elemental analysis except ash contents [30,47]. monitoring the set SBR value (about 0.6e1).
Palm shells passed from the hopper to the reactor by a
screw conveyor whose rotation was controlled by an inverter
2.2. Pilot plant and gasification process
in order to obtain the desired flow-rate (into the range
3e4 kg h1).
The experiments have been performed in a continuous pilot
During the experiments, a nitrogen flow-rate (about
plant, having a feeding capacity of 5 kg h1. The plant is a
1 m3 h1 in normal conditions) has been introduced in the
rotary kiln gasifier with an indirect electric heating system
plant in order to:
and its layout is sketched in Fig. 1 [48]; each component will be
described in detail in the following sections.
 create an inert atmosphere during the reactor heating
starting phase (washing);
2.2.1. Fuel system
 remove produced gases from the reactor (carrier).
The feeding system consists of a closed tank for fuel storage
and a screw conveyor which transports it directly into the ro-
Moreover, a secondary nitrogen flow-rate (about 0.5 m3 h1
tary kiln reactor. The 50 dm3 storage tank has an appropriate
in normal conditions) has been sent to the mechanical seals
system to prevent the formation of obstructions that impede
for preventing gas leakage from the reactor to the external.
the passage of the material. The reactor is fed by the top.
This mass has not been introduced into the reactor.
The char produced has been collected into a tank and then
2.2.2. Rotary kiln reactor
unloaded after the test. The produced gas crossed the
The reactor is an indirectly heated rotary kiln with an internal
quencher for cooling and condensing the excess steam, and
diameter of 400 mm and a length of 1000 mm and then has an
the washing section for removing acid components eventually
overall volume of about 126 dm3. The speed rotation can be
present in the gas. Then it crossed a hydraulic guard with
varied in the range of 0.5e3 min1 and the temperature in the
50 mm water column for assuring a small overpressure and
range 400e1100  C. The residence time of the solid matrix can
avoiding undesired air entrance. Finally, it has been analyzed
be varied changing appropriately the slope and the speed of
by using a microgas chromatograph Agilent A3000 and
the kiln. It can be analytically computed by the Equation (1)
quantified by using a volumetric counter. Once done, the gas
which reads [49]:
has been sent to a torch for realizing the complete oxidation. A
t ¼ ð1:77OqBLÞ=ðSnDÞ (1) hydraulic emergency guard, placed on the line between the
quencher and the reactor, assured the drainage of gas in the
The geometric features of the rotary kiln reactor are re-
torch in case of an undesired overpressure.
ported in Table 2.
In order to analyze the product features, different tests
have been carried out at different SBR. The process parame-
2.2.3. Gas treatment system
ters used in the experiments are shown in Table 3.
The produced gas treatment system consists of an outside
For each test the solid and gas yields so defined as be
gravity settler to collect the large char particles eventually
calculated:
present into syngas, a quencher with water to collect fine
particles, tars and cool down the gas and a scrubber with an solid yield ¼ char flow rate=palm shells flow rate (2)

Table 1 e Analysis of palm shells.


Sample Proximate analysis [g g1] (dry basis) Ultimate analysis [g g1] (dry basis) HHV [MJ kg1]

Volatile matters Fixed carbon Ash N C H S Oa

Palm shells 71 21.5 7.5 0.6 48.3 5.5 <0.01 38.1 17.4

a The oxygen (O) content was determined by difference.


426 b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1

Fig. 1 e Schematic diagram of pilot plant.

species (C, H, and O) at equilibrium, considering the atomic


gas yield ¼ dry gas flow rate=palm shells flow rate (3)
species as invariant during the process. The remaining two
At the end of the experiments, a total mass balances have describe the reactions involving the species at equilibrium:
been done and the residual palm shells in the hopper were
unloaded and quantified. Ca Hb Oc )/ða  b=4  c=2ÞC þ b=4CH4 þ c=2CO2 DHr
1
(4)
¼ 218:9 kJ$mol

1
C þ CO2 )/2CO DHr ¼ 172:5 kJ$mol ðBoudouard reactionÞ
3. Numerical model
(5)
The influence of SBR on gasification process have been Regarding the steam gasification section, the components
numerically evaluated with the aid of a simulation tool considered at equilibrium have been six: H2, CH4, CO, CO2, C
created by ChemCAD. and H2O. Similarly, the linearly independent equations have
The numerical model is able to determine the gas compo- been the atomic balance of the species at equilibrium (C, H,
sition at equilibrium by following the two subsequent oper- and O) and three reactions of equilibrium as follows:
ating steps:
1
C þ H2 O)/CO þ H2 DHr ¼ 131:3 kJ$mol (6)
 palm shells pyrolysis at atmospheric pressure and 400  C,
under inert atmosphere conditions; 1
CH4 þ H2 O)/CO þ 3H2 DHr ¼ 205:8 kJ$mol
 pyrolysis gas steam gasification at atmospheric pressure 
ðCH4 steam reforming (7)
and 850  C.
1
COþH2 O)/CO2 þH2 DHr ¼ 50:8 kJ$mol ðCO shift reactionÞ
Reactions have been evaluated in both reactors (i.e. pyrol-
(8)
ysis and steam gasification), according with the simplification
assumptions at equilibrium. Regarding the pyrolysis section, The experiments showed that the steam gasification of a
only five components have been considered: H2, CO, CO2, CH4 matrix with organic components produces a gas containing at
and C so that five linearly independent equations have been least two additional components: C2H6 and C2H4. Moreover,
introduced. Three of them describe the balance on the atomic their fractions are smaller than those of the other components.

Table 2 e Geometric characteristics of the


4. Results and discussion
rotary kiln.
L [m] 1
4.1. Thermogravimetric analysis
D [m] 0.4
n [min1] 0.7e0.5
S [ ] 3.75 The thermal stability of palm shell was determined using
thermogravimetric analysis (TGA) under nitrogen atmosphere
b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1 427

Table 3 e Process parameters for experimental tests.


No test T [ C] Gasifying GShells of palm nuts SBR [g g1] N2 flow-rate [m3 h1] Solid residence Gas residence
agent [kg h1] in normal conditions time [min] time [s]

1 850 Steam 3.4 0.6 0.9 6 37.4


2 850 Steam 4.1 0.7 0.9 6 33.9
3 850 Steam 3.4 1.0 0.9 6 40.7

until 900  C with a constant heating rate of 20  C min1 fol- [47]. Higher SBR does not influence appreciably the concen-
lowed by air combustion. The result is shown in Fig. 2 which tration of tars probably due to the high gas residence time and
presents the mass loss curve (TG) and derivative thermogra- process temperature.
vimetric analysis (DTG). Syngas were composed mainly by CO, CH4, CO2 and H2 with
The profile is a typical thermal degradation of lignocellu- C2H4 gases accounted from 0.2 to 3% and C2H6 was <0.1% in
losic material divided into three regions. The initial mass loss volume fraction. For each species, the trend is reported in
in the range of 50e100  C can be confidentially attributed to Fig. 3: H2 content increases from 40 to 52% as the SBR increases
the moisture drying (about 10% in mass); the other two pro- as well and correspondingly, CH4 content decreases from 12 to
nounced mass losses in the range of 240e380  C correspond to 6%; this is basically due to the high steam quantity which
the degradation of hemicelluloses and of the celluloses found makes possible the steam reforming reaction of CH4 Eq. (7),
in the structure of lignocellulosic biomass as reported in Ref. giving back H2. Comparable results have been obtained with
[50]. Thus, the main devolatilization of the lignocellulosic the catalytic steam gasification of oil palm wastes in a com-
samples takes place at the temperature range of 200e400  C, bined fixed bed reactor with SBR ¼ 1.33 at 800  C [53].
but the process slightly continues to 430e460  C probably due The percentage of H2þCO in the total gas evolved was as
to the degradation of lignic residues that it starts at low high as 64e74%. As SBR was increased, the amount of CO
temperature of 160e170  C and it prolongs at a low rate to decreased as the CO shift reaction proceeded Eq. (8): at SBR of
900  C [51]. The DTG-curve corresponds to the region where 1.0, the [H2]/[CO] was about 2.4, which is comparable to that of
the slope of TG-curve is almost constant depicted three peaks: the cedar wood and sufficient to produce biofuel [13]. CO and
the first is relative to the removal moisture, the other two CO2 contents were influenced by steam reaction Eq. (8) and
peaks in the range of 200e400  C the first, from the left, bottom Boudouard reaction Eq. (5).
and enlarged, corresponds to degradation of hemicellulose, The gas composition reported in graphs is nitrogen free, to
while for the second peak, corresponding to the celluloses, make able a direct comparison with the one obtained from
more narrow and high. At temperatures greater than 400  C, numerical simulation, when nitrogen is not present [54].
occurs the thermal degradation of the lignin associated with a A direct comparison among the experimental results ob-
very low peak and expanded, hardly distinguishable from the tained in the present work and those obtained by Galvagno
base line [52]. et al. on steam gasification of poplar in a rotary kiln reactor
[30] is shown in Table 4.
4.2. Effect of SBR The main difference consists in the type of biomass used
(poplar) and in the SBR value, while the gasification temper-
To study the effect of SBR on gasification performance, this ature used is the same. First of all in this work it noticed a
parameter was varied in the range of 0.6e1 while the process higher char yield and a lower gas yield with respect of those
temperature was fixed at 850  C. obtained from Galvagno et al. (2009), and this is due to a lower
Tar was recovered from quencher and its yield was low value of the residence time of biomass in the reactor and to a
(<0.1%) under all three conditions as well as reported in Ref.

Fig. 2 e Thermogravimetric analysis of palm shells. Fig. 3 e Gas volume fraction vs SBR.
428 b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1

Table 4 e Comparison with data from literature.


Galvagno This work
et al. (2009)

Proximate analysis [%]


Volatile matter 81 71
Fixed carbon 17.3 21.5
Ash 1.7 7.5
Ultimate analysis [%]
C 47.5 48.3
H 6.1 5.5
N 0.1 0.6
S 0.0 < 0.01
Oa 44.6 38.1
HHV [MJ kg1] 19 17.4
Gasification parameters
Fig. 4 e Solid and gas mass fractions vs SBR.
T [ C] 850 850 850 850
SBR 2.2 0.6 0.7 1.0
Solid residence 15 6 6 6
times [min]
Fractions yields [%] 4.3. Comparison between the experimental and
Char yield 14.4 21.9 21.7 21.2
numerical data
Gas yield 89.9 62.2 62.4 62.7
Volumetric gas composition [%]
H2 45.5 40.4 39.4 52.2 The numerical model is validated by performing simulations
CH4 8.6 12.2 11.3 6.0 able to reproduce experimental conditions. In particular the
CO 23.1 24.1 23.5 22.1 numerical model has been tested by varying the SBR from 0.4
CO2 20.8 21.7 23.1 19.5 to 1, then in a range largest then that used in experimental
C2H4 1.9 1.7 2.7 0.2 work. Looking at the comparison between the experimental
C2H6 0.1 <0.1% <0.1% <0.1%
and numerical results, in Fig. 5, it can be observed that the
H2/CO 2 1.7 1.7 2.4
trend in numerical simulation is a little different from the
experimental data.
In fact, for all the values of feeding ratios adopted the H2
higher value of the fixed carbon content in palm shells with and CH4 contents in the experimental case are lower and
respect to that one in the poplar. higher, respectively, of these obtained in the numerical
Furthermore, regarding the composition of the gas there is simulation: this result can be easily justified by considering
not significant difference in spite of higher solid residence that the numerical model permits to obtain the gas compo-
time and SBR. sition at the equilibrium at the process temperature
It is not possible a direct comparison with the work of Klose (T ¼ 850  C), while the residence times of palm shells and of
et al. on lignite, since in that case they performed a first step of gas produced during the gasification do not permit to reach
coking of lignin which was followed by a steam gasification of this equilibrium. In particular the steam reforming of CH4 (Eq.
the solid residue and was conducted in batch mode [28]. (7)), is incomplete and not all of the CH4 is converted to H2.
Moreover, it is not possible an immediate comparison with
the results obtained by Chidi et al. because they performed a
pyrolysis of biomass in the form of waste wood and subse-
quently a catalytic steam reforming of the pyrolysis oils and
gases [29].
Fig. 4 shows that the solid and gas yields have practically
reached their limit values already with the lowest feeding
ratio adopted (SBR ¼ 0.6), also if gas yield increases and solid
decreases slightly for the advance of the steam reactions.
Consequently, from that value of steam biomass ratio, the
addition of gasifying agent to the process does not involve any
change in the amount of char and gas obtained, but on the
other hand it involves a higher input of energy for super-
heating of water steam till to the process temperature
(T ¼ 850  C).
As SBR increased from 0.6 to 1.0, more steam was introduced
into reactor, the increased steam quantity enhanced the con-
centration of steam in reactor to facilitate all steam reforming
reactions (Eqs. (6), (7) and (8)) which resulted in a major increase
of H2 content in gas and a ideal [H2]/[CO] ratio for the synthesis Fig. 5 e Comparison between experimental and numerical
of methanol and Fischer-Tropsch synthesis oil. gas composition.
b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1 429

Table 6 e Power generated from combustion of syngas


produced from 1 kg hL1 of palm shells.
No test SBR P [kW]

1 0.6 2.6
2 0.7 2.5
3 1 2.5

X
n
HHVgas ¼ rgas HHVi %vol iri (9)
i¼1

Fig. 6 e Comparison between experimental and numerical where “i” represents the component present in the gas phase
char yield. and “n” the total number of components and subtracting the
heat of vaporization of the water vapor:

mH2 mCH4 9 mC2 H4 9 mC2 H6 9
The non-completion of the steam reforming reaction im- LHVgas ¼ HHVgas  rgas 9þ þ þ
mgas mgas 4 mgas 7 mgas 5
plies also an experimental CO content less than that provided  
mH2 O
by the numerical simulation. þ DHvap
mgas
Moreover, at the process temperature the endothermic (10)
formation of CO due to the Boudouard reaction (Eq. (5)) is
dominant: also in this case the residence time in experimental where DHvap is the heat of vaporization of the water (equal to
work does not assure the complete conversion of CO2 in CO 2.257 MJ kg1) and r is the density of the syngas.
and in such a way the experimental tests imply a higher value Gas LHV ranged between 10 and 12 MJ m3 in normal
of CO2 and a lower value of CO than those obtained in the conditions comparable to data in literature about steam
numerical simulation where the equilibrium is considered. gasification of palm oil wastes [53] and also other common
Devolatilization of biomass gives high contents in CH4 and biomass [30]. These results are higher than those obtained
higher hydrocarbons (CxHy) which do not react completely to ones in palm biomass gasification with air [56], thus it dem-
equilibrium concentration of CO, CO2 and H2, according with onstrates the potential of this locally available biomass for
the literature data [55]. gasification purpose.
Regarding the solid yield, Fig. 6 shows a comparison be- It is noticed that both the experimental and numerical LHV
tween the experimental and numerical results when the decrease when SBR increases due to changing in gas compo-
steam biomass ratio varies from 0.4 to 1: in numerical simu- sition. The higher values of CH4 and H2 in experimental work
lation, it can be seen that when the SBR increases from 0.4 to 1 are responsible of the higher value of LHV obtained in the
the solid yield decreases from 14% to 9%. numerical simulation work. Thus, higher SBR favored H2
First of all, it can be noted that the values obtained in the production and biofuel production but didn’t favor gas heating
numerical simulation result are lower than those obtained in value degrading product gas quality.
the experimental work: this is due to the kinetics of the re-
action between solid and gas phases, which in experiments 4.4. Energy efficiency
slows down the removal of the volatile from palm shells. The
kinetics of reaction are also the cause of the constant value of For different values of SBR, in Table 6 it is reported the power
the char yield in experimental case, where, unlike what hap- that can be generated when the syngas produced starting
pens in the simulations, the residence time of the material in from 1 kg h1 of palm shells is combusted: it is noticed that the
the rotary kiln reactor does not allow the completion of sol- power slowly decreases when the steam biomass ratio in-
idegas reactions in spite the SBR increases. creases as well and this because of variation in the gas
Obtained data give back information on the gas energy composition that decreases the LHV of gas.
content. In such a way it can tell that the gasification using an SBR of
Table 5 shows the results about the LHV of gas obtained by 0.6 represents the optimum condition in terms of resulting
varying SBR in both the experimental and numerical works. energy, among those investigated, in virtue also of consider-
The first is calculated starting from HHV: ation that when SBR increases it is necessary more demand of
energy in input to overheat the steam till to the process
temperature.
Table 5 e Gas energy content.
No SBR Experimental Numerical LHV
test LHV [MJ m3] in [MJ m3] in normal 5. Conclusions
normal conditions conditions
In this study, it has been analyzed the key aspects on steam
1 0.6 11.9 10.4
gasification of palm shells in a rotary kiln pilot plant. A series
2 0.7 12.0 10.3
3 1 10.0 9.8 of experiments have been performed to explore the effect of
SBR and both the process performance and the gas features
430 b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1

have been experimentally and numerically evaluated by [4] National Renewable Energy Laboratory. Biomass cofiring: a
varying SBR. renewable alternative for utilities. U.S. Department of
The experimental results show that: Energy; June 2000. (DOE).DOE-102000e1055.
[5] Wisconsin Biorefining development initiative.
Thermochemical liquefaction. Available from: http://www.
B the H2 content in the syngas increases until 52% as the SBR wisbiorefine.org/proc/thermo.pdf; 2004.
increases as well and correspondingly, CH4 content de- [6] Zahari MAKM, Zakaria MR, Ariffin H, Mokhtar MN, Salihon J,
creases until 6%; Shirai Y, et al. Renewable sugars from oil palm frond juice as
B the solid and gas yields have practically reached their limit an alternative novel fermentation feedstock for value-added
values already with the lowest feeding ratio adopted products. Bioresour Technol 2012;110:566e71.
(SBR ¼ 0.6); [7] Junker H. Cofiring of 500 MW coal-fired power plant with 10%
EFB bales or 5% shells and as a 2015 scenario 10% cofiring of
B at SBR ¼ 1.0, [H2]/[CO] ratio is equal to 2.4 which is ideal for
POFF. Available from: http://cdm.unfccc.int/Projects/DB/
the synthesis of methanol and Fischer-Tropsch synthesis oil; DNV-CUK1174646628.13/ReviewInitialComments/
B gas LHV ranged between 10 and 12 MJ m3 in normal L098VJ5PJ700XB67UXNM4BFJMQEG2W; 2005.
conditions. [8] Ringer M, Putdche V, Scahill J. Large-scale pyrolysis oil
production: a technology assessment and economic analysis.
The experiments demonstrated that there are some dif- National Renewable Energy Laboratory. Available from:
http://www.nrel.gov/docs/fy00osti/37779.pdf; 2006.
ferences when comparing experimental and numerical gas
[9] Abnisa F, Wan Daud WMA, Husin WNW, Sahu JN. Utilization
composition because the theoretical trend takes into account
possibilities of palm shell as a source of biomass energy in
the values at thermodynamic equilibrium, therefore in the Malaysia by producing bio-oil in pyrolysis process. Biomass
numerical simulation it is to be noticed: Bioenergy 2011;35(5):1863e72.
[10] Ganan J, Abdulla Al-Kassir, Miranda AB, Turegano J,
B a higher content of H2 and a lower content of CH4 (almost zero) Correia S, Cuerda EM. Energy production by means of
with respect to those obtained in the experimental results; gasification process of residuals sourced in Extremadura
(Spain). Renew Energy 2005;30(11):1759e69.
B a lower value of CO2 and a higher value of CO than those
[11] Karamarkovic R, Karamarkovic V. Energy and exergy
obtained in the experimental tests; analysis of biomass gasification at different temperatures.
B a lower char yield than those obtained in the experiment Energy 2010;35(2):537e49.
due to the kinetics of the reaction between solid phase and [12] Higman C, Van Der Burgt M. Gasification 2003.
gas phase, which in experiments slow down the removal of [13] Holmgren Kristina M, Berntsson T, Andersson E, Rydberg T.
the volatile from palm shells. System aspects of biomass gasification with methanol
synthesis e process concepts and energy analysis. Energy
2012;45(1):817e28.
From the energy point of view, higher SBR does not favor
[14] He M, Hu Z, Xiao B, Li J. Hydrogen-rich gas from catalytic
gas heating value and consequently the power generated from steam gasification of municipal solid waste (MSW): influence
combustion of syngas slowly decreases. of catalyst and temperature on yield and product
Finally, it can tell that there is a big potential of this locally composition. Int J Hydrogen Energy 2009;34(1):195e203.
available biomass for gasification purpose using an SBR of 0.6 [15] Mahishi MR, Goswami DY. An experimental study of
which represents the optimum condition in terms of resulting hydrogen production by gasification of biomass in the
presence of a carbon dioxide sorbent. Int J Hydrogen Energy
energy, among those investigated, in virtue also of consider-
2007;32(14):2803e8.
ation that when SBR increases it is necessary more demand of
[16] Turn S, Kinoshita C, Zhang Z, Ishimura D, Zhou J. An
energy in input to overheat the steam till to the process experimental investigation of hydrogen production from
temperature. On the other hand, a value SBR equal to 1.0 is biomass gasification. Int J Hydrogen Energy 1998;23(8):641e8.
preferable if steam gasification is devoted in producing a gas [17] Larocca V, Molino A, Petrone MT, Barisano D, Giordano G,
which has to be used for the synthesis of methanol or Fischer- Braccio G. Synthesis of methanol from biomass-derived
Tropsch synthesis oil. syngas: modeling and sizing of bench-scale reactor. Int J
Energy Technol 2010;2(2):1e7.
[18] Rapagna S, Provendier H, Petit C, Kiennemann A, Foscolo PU.
Development of catalysts suitable for hydrogen or syn-gas
references production from biomass gasification. Biomass Bioenergy
2002;22(5):377e88.
[19] Klein A, Whiting K, Archer E, Schwage R. Gasification and
[1] Solomon S, Qin D, Manning M, Chen Z, Marquis M, pyrolysis: what is the current situation for waste
Averyt KB, et al. Intergovernmental Panel on Climate Change management? Waste Manag World 2004:71e5.
(IPPC). Climate change. The physical science basis, summary [20] Arena U, Mastellone ML. Fluidized pyrolysis and gasification
for policymakers. Contribution of working group I to the of solid wastes. In: Proceedings of industrial fluidization
Fourth Assessment Report of the Intergovernmental Panel South Africa 2005. The South African Institute of Mining and
on Climate Change. Cambridge University Press; 2007. Metallurgy; 2005. p. 53e68. ISBN: 1-919782-83-0.
[2] Dentener F, Derwent R, Dlugokencky E, Holland E, Isaksen I, [21] Dayton DA. Review of the literature on catalytic biomass tar
Katima J, et al. Intergovernmental Panel on Climate Change destruction, milestone completion report. NREL/TP-
(IPPC). Climate change. The scientific basis. Contribution of 510e32815 (Project No. DE-AC36-99-GO10337). Golden.
working group I to the Third Assessment Report of the Colorado USA: National Energy Laboratory; 2002.
Intergovernmental Panel on Climate Change 2001. [22] Prisciandaro M, Lancia A, Musmarra D. Calcium sulphate
[3] Lora ES, Andrade RV. Biomass as energy source in Brazil. dehydrate nucleation in the presence of calcium and sodium
Renew Sustain Energy Rev 2009;13(4):777e88. chloride salts. Ind Eng Chem Res 2001;40:2335e9.
b i o m a s s a n d b i o e n e r g y 5 6 ( 2 0 1 3 ) 4 2 3 e4 3 1 431

[23] Lancia A, Karatza D, Musmarra D, Pepe F. Adsorption of technology using oil palm residue in Malaysia. Jpn Soc
mercuric chloride from simulated incinerator exhaust gas by Energy Resour 2008;29(5):1e7.
means of Sorbalit particles. J Chem Eng Jpn 1996;29:939e46. [41] Corporation EX. Methane and EFB power generation project
[24] Di Natale F, Erto A, Lancia A, Musmarra D. Mercury in Terengganu states, Malaysia. Reports of CDM/JI Feasibility
adsorption on granular activated carbon in aqueous Studies: FY 2005. Available from: http://gec.jp/main.nsf/en/
solutions containing nitrates and chlorides. J Hazard Mater Activities-CDMJI_FS_Programme-FS200509; 2006 Mar.
2011;192(3):1842e50. [42] Chiew YL, Iawata T, Shimada S. System analysis for effective
[25] Lancia A, Musmarra D, Pepe F. Modeling of SO2 absorption use of palm oil waste as energy resources. Biomass
into limestone suspensions. Ind Eng Chem Res Bioenergy 2011;35(7):2925e35.
1997;36:197e203. [43] Husain Z, Zainal ZA, Abdullah MZ. Analysis of biomass-
[26] Hatzilyberis KS, Androutsopoulos GP. Lignite chemical residue-based cogeneration system in palm oil mills.
conversion in an indirect heat rotary kiln gasifier. Therm Sci Biomass Bioenergy 2003;24(2):117e24.
2006;10:181e97. [44] Mahlia TMI, Abdulmin MZ, Alamsyah TMI, Mukhlishein D.
[27] Hatzilyberis KS. Design of an indirect heat rotary kiln An alternative energy source from palm waste industry for
gasifier. Fuel Process Technol 2011;92(12):2429e54. Malaysia and Indonesia. Energy Conv Manag
[28] Klose E, Born M. Partial gasification of lignite coke with 2001;42(18):2109e18.
steam in a rotary kiln for activated carbon production. Fuel [45] Erlich C. Comparative study of residue pellets from cane
1985;64(9):1313e6. sugar and palm-oil industries with commercial wood pellets,
[29] Efika CE, Wu C, Williams PT. Syngas production from pyrolysis- applied in downdraft gasification [Doctoral thesis].
catalytic steam reforming of waste biomass in a continuous Stockholm, Sweden: Royal Institute of Technology, ISBN 978-
screw kiln reactor. J Anal Appl Pyrolysis 2012;95:87e94. 91-7415-455-9; 2009.
[30] Galvagno S, Casciaro G, Casu S, Martino M, Mingazzini C. [46] Prasertsan S, Prasertsan P. Biomass residues from palm oil
Steam gasification of tyre waste, poplar, and refuse-derived millas in Thailand: an overview on quantity and potential
fuel: a comparative analysis. Waste Manag 2009;29(2):678e89. usage. Biomass Bioenergy 1996;11:387e95.
[31] Idris S, Rahman N, Ismail K, Alias A, Rashid Z, Aris M. [47] Ogi T, Nakanishi M, Fukuda Y, Matsumoto K. Gasification of
Investigation on thermochemical behaviour of low rank oil palm residues (empty fruit bunch) in an entrained-flow
Malaysian coal, oil palm biomass and their blends during gasifier. Fuel 2013;104:28e35.
pyrolysis via thermogravimetric analysis (TGA). Bioresour [48] Donatelli A, Iovane P, Molino A. High energy syngas
Technol 2010;101(12):4584e92. production by waste tyres steam gasification in a rotary kiln
[32] Saka S. Whole efficient utilization of oil palm to value-added pilot plant. Experimental and numerical investigations. Fuel
products. In: Proceedings of JSPS-VCC natural resources & 2010;89(10):2721e8.
energy environment seminar, Kyoto, Japan 2005. [49] Xiao YL, Eckehard S. Mean residence time and hold-up of
[33] Saxena RC, Seal D, Kumar S, Goyal HB. Thermo-chemical solids in rotary kilns. Chem Eng Sci 2006;61(15):5176e81.
routes for hydrogen rich gas from biomass: a review. Renew [50] Mohamed AR, Mohammaddi M, Darzi GN. Preparation of
Sustain Energy Rev 2008;12(7):1909e27. carbon molecular sieve from lignocellulosic biomass: a
[34] Yusoff S. Renewable energy from palm oil innovation on review. Renew Sustain Energy Rev 2010;14(6):1591e9.
effective utilization of waste. J Clean Prod 2006;14(1):87e93. [51] Muradov N, Fidalgo B, Gujar AC, T-Raissi A. Pyrolysis of fast-
[35] Shuit SH, Tan KT, Lee KT, Kamaruddin AH. Oil palm biomass growing aquatic biomass-Lemns minor (duckweed):
as a sustainable energy source: a Malaysian case study. characterization of pyrolysis products. Bioresour Technol
Energy 2009;34(9):1225e35. 2010;101(21):8424e8.
[36] Sumanthi S, Chai SP, Mohamed AR. Utilization of oil palm as [52] Hussain A, Ani FN, Darus AN, Ahmed Z. Thermogravimetric
a source of renewable energy in Malaysia. Renew Sustain and thermochemical studies of Malaysian oil palm shell
Energy Rev 2008;12(9):2404e21. waste. J Tech 2006;45(A):43e53.
[37] Sulaiman F, Abdullah N, Gerhauser H, Shariff A. An outlook [53] Jianfen L, Yanfang Y, Xuanming Z, Jianjun L, Rong Y.
of Malaysian energy, oil palm industry and its utilization of Hydrogen rich gas production. Int J Hydrogen Energy
wastes as useful resources. Biomass Bioenergy 2009;34(22):9108e15.
2011;35(9):3775e86. [54] Molino A, Braccio G, Fiorenza G, Marraffa FA, Lamonaca S,
[38] Chow MC. Assessment of potential and availability of palm Giordano G, et al. Classification procedure of the explosion risk
biomass for bioconversion to bioethanol. In A report areas in presence of hydrogen-rich syngas: biomass gasifier and
prepared under Malaysian-Danish Environment Cooperation molten carbonate fuel cell integrated plant. Fuel
Programme. Renewable energy and energy efficiency 2012;99:245e53.
component. Malaysia Energy Centre (PTM); 2005. [55] Molino A, Giordano G, Motola V, Fiorenza G, Nanna F,
[39] Dora RT. Biomass-fired CHP in palm oil mills. In A report Braccio G. Electricity production by biomass steam
prepared under Malaysian-Danish Environment Cooperation gasification using a high efficiency technology and low
Programme. Renewable energy and energy efficiency environmental impact. Fuel 2013;103:179e92.
component. Malaysia Energy Centre (PTM); 2004 Dec. [56] Lahijani P, Zainal ZA. Gasification of palm empty fruit bunch
[40] Nagatomi Y, Yamamoto H, Yamaji K, Iwasaki H, Yamada K. in a bubbling fluidized bed: a performance and
A system analysis of energy utilization and competing agglomeration study. Bioresour Technol 2011;102(2):2068e76.