JOURNAL

Journal of
OFOil
OILPalm
PALMResearch
RESEARCHVol.
18 (JUNE
18 June
2006)2006 p. 210-218

THERMOCHEMICAL BEHAVIOUR OF EMPTY

FRUIT BUNCHES AND OIL PALM SHELL WASTE
IN A CIRCULATING FLUIDIZED-BED
COMBUSTOR (CFBC)
AHMAD HUSSAIN*; FARID NASIR ANI*; AMER NORDIN DARUS*; HAMDAN MOKHTAR**; SAIFUL
AZAM** and AZEMAN MUSTAFA+
ABSTRACT
Circulating fluidized-bed (CFB) technology is considered to be one of the most suitable techniques to thermally
convert fuels into useful energy. However, practical experience is available for only a limited number of fuels
and conditions. This paper describes the results from a bench-scale circulating fluidized-bed combustor (CFBC),
installed at SIRIM Berhad, Shah Alam, for gasification and combustion of different biomass materials. The
purpose of the tests was to investigate the suitability of the selected fuels for energy production using CFBC
while taking care of the flue gas emissions. The experiments gave sufficient information on the main process
and flue gas characteristics. The measurements for temperatures and emissions were done for four different
biomass samples of empty fruit bunch (EFB) of palm shell and palm shell waste powders of 210-300, 425 and
600 microns. The concentrations of CO, NOx and CO 2 in the flue gas were measured continuously. The
combustion performances were evaluated by varying the primary gas flow through the CFBC tubular furnace
to identify the optimum operating conditions for the CFBC. The NOx content was from 20-164 ppm while
the CO emissions were high for some operating conditions.
Keywords: biomass, circulating fluidized-bed, oil palm solid wastes, emission characteristics, temperature profiles.
Date received: 29 August 2005; Sent for revision: 15 November 2005; Received in final form: 7 February 2006; Accepted: 8 March 2006.

INTRODUCTION
Biomass was the worlds first fuel and source of
energy, but when coal became widely available, to
be followed later by bio oil and natural gas, its use
declined. However, in recent years interest in it has
* Faculty of Mechanical Engineering,
Universiti Teknologi Malaysia,
81310 UTM,
Skudai, Johor, Malaysia.
E-mail: ahussain_a2000@yahoo.com
** Environment and Bioprocess Technology Centre,
SIRIM Berhad,
P. O. Box 7035,
40911 Shah Alam,
Selangor, Malaysia.
+
Faculty of Chemical and Natural Resources,
Universiti Teknologi Malaysia,
81310 UTM,
Skudai,
Johor, Malaysia.

been renewed, much of being focused on its efficient

conversion for energy. The importance of biomass
energy in developing countries today is indisputable.
However, in many countries, wood supply (as fuel)
can no longer meet the demand and very few
countries have excess wood for gasification or
charcoal production without serious impact on their
natural resources (Kate et al., 1997).
With respect to the global issues of sustainable
energy and reduction in greenhouse gases, biomass
is getting increased attention as a potential source
of renewable energy. However, biomass is not yet
competitive with fossil fuels. Fossil fuels contribute
to the major part of worlds total energy
consumption. According to the World Energy
Assessment report, 80% of the worlds primary
energy consumption is contributed by fossil fuels,
14% by renewable energy (out of which biomass
contributes 9.5%) and 6% by nuclear energy (Rogner
et al., 2000; Devi et al., 2003).
210

Malaysian Palm Oil Board 2006

THERMOCHEMICAL BEHAVIOUR OF EMPTY FRUIT BUNCHES AND OIL PALM SHELL WASTE IN A CIRCULATING FLUIDIZED-BED COMBUSTOR (CFBC)

Different biomass conversion processes produce

heat, electricity and fuels. Biomass integrated
gasification/combined cycle systems are of prime
importance as modern technologies (Bui et al., 1994).
Among all biomass conversion processes,
gasification is one of the promising ones. The energy
efficiency in the case of gasification is higher than
that of combustion. One of the major issues in
biomass gasification is how to deal with the tar
formed in the process.
Among the proven combustion technologies
(such as grate-fired systems and suspension-fired
systems, fluidized bed systems), the fluidized bed
is reported to be the most efficient and suitable for
converting agricultural and wood residues into
energy (Bridgwater, 1994). The emissions from
biomass combustion systems, including the products
of complete (CO2) and incomplete combustion (CO,
char particles, tar, polycyclic aromatic hydrocarbons
and other organic compounds), as well as NOx, SO2,
HCl and ash particles, are affected by the combustion
method as well as by the operating conditions and
fuel properties (Dornburg et al., 2001). In order to
control the emissions effectively (especially to
minimize the emissions of incomplete combustion
products), one has to know the effects of the
operating parameters on the emission behaviour. In
this research, experiments have been done, under
different operating conditions, to estimate the
emissions behaviour.
In the very near future, it will be necessary to
work more intensively on the replacement of fossil
fuels by high hydrogen content fuels and to shift to
natural gas and biogas. As we prepare for a world
of diminishing fossil fuel reserves, with increasing
demand for the protection of our environment, the
combustion community has two major goals to
fulfill: enhancement of combustion efficiency and
reduction of pollutant emission. However, it is to
be realized that the combustion phenomena are very
complex, multi-dimensional, unsteady and multidisciplinary.
The development of new combustion
technologies and their transfer to large plants or
factories are on the way, because the competition has
already became a worldwide issue in the
international economic actors. The transfer of these
novel technologies to small- and mid-sized
manufacturing industries could be essential for the
economies of developing countries. According to
recent developments in fundamental research, the
design of energy systems and the implementation
of the operations feedback will mainly rely on
progress on the conception of more reliable materials
(optimization of heat transfer, thermal and
mechanical resistance, lengthening of life time),
efficiency and reliability of energy production and
energy saving (reduction of fuel consumption,
alternate fuel, reduction of pollutant emissions) and
the control of combustion processes (Pierre, 2004).
211

Over the last 8-10 years, interest in large scale

biomass gasification for power generation has been
growing steadily. Efficiency well over 40% has been
predicted for such plants in the near future. For
capacity range lower than 5-10 MWe, new designs
and catalytic gas cleaning may make a new
generation of biomass power gasifier feasible in the
near future (Bhattacharya, 1997).

MATERIALS AND METHODS

This paper deals with the experimental study of
combustion of different biomass fuels, namely, oil
palm shell waste and empty fruit bunch (EFB) in a
circulating fluidized-bed combustor (CFBC) using
alumina particles of 200 microns as inert material.
The main objectives of this work were to study the
formation and reduction of the major gaseous
pollutants (CO and NOx) in the CFBC when firing
the selected biomass fuels, and to study the
combustion performance when fired with different
fuels for different operating conditions.
Biomass Materials Used for Experimental Work
Biomass, as a fuel, is characterized by high
moisture and volatile content, low bulk density, low
specific energy and normally low ash content. The
proximate and ultimate analysis of oil palm shell are
shown in Tables 1 and 2 (Ani, 1992).
TABLE 1. PROXIMATE ANALYSIS OF OIL PALM SHELL
WASTE

Solid waste

Palm shell

Proximate analysis %
Moisture

Volatile

Fixed
carbon

Ash

9.7

67

21.2

2.1

TABLE 2. ULTIMATE ANALYSIS OF OIL PALM SHELL

WASTE

Solid waste

Palm shells

Ultimate analysis %
C

47.62

6.2

0.7 43.38

GCV
S

MJ kg-1

19.1

The ultimate and promimate analysis of EFB was

also done. The results are being tabulated in Tables 3
and 4.
TABLE 3. PROXIMATE ANALYSIS OF EMPTY FRUIT
BUNCH (EFB)

EFB

Palm shell

Proximate analysis %
Moisture

Volatile

Fixed
carbon

Ash

6.8

77.4

19.3

3.3

Malaysian Palm Oil Board 2006

JOURNAL OF OIL PALM RESEARCH 18 (JUNE 2006)

TABLE 4. ULTIMATE ANALYSIS OF EMPTY FRUIT BUNCH
(EFB)

EFB

Ultimate analysis %
C

GCV
Cl MJ kg-1

Palm shells 49.5 5.9 0.5 40.6 0.10 0.20

appearance of EFB after sieving is shown in Figure

1b. Sieving of EFB improved the transfer of EFB from
the screw feeder.
Experimental

18.1

The EFB was shredded and in this state it

contained fibrous EFB several centimetres long as
shown in Figure 1a. These fibrous EFB were easily
entangled and formed into lumps, which jammed
the screw feeder. It was found that such EFB could
not be dropped smoothly from the screw feeder to
the sample supply port. In order to remove the
fibrous EFB causing such a problem, the shredded
EFB were sieved through a 5 mm mesh screen. The

The experiments were carried out in SIRIM

Berhad, Shah Alam. The system was developed by
JFE, Japan under a joint-venture programme
between the Japanese Government and SIRIM
Berhad. The experimental apparatus consisted of a
circulating fluidized-bed (CFB) type experimental
gasifier, experimental sample supplying unit,
secondary combustion furnace, gas cooler, dust
collector, blower and control panel. A schematic
diagram of the test rig is shown in Figure 2.

(a)

(b)

Figure 1. Appearance of empty fruit bunches (EFB).

Figure 2. Layout of the experimental circulating fluidized-bed (CFB) gasifier at SIRIM.

212

Malaysian Palm Oil Board 2006

THERMOCHEMICAL BEHAVIOUR OF EMPTY FRUIT BUNCHES AND OIL PALM SHELL WASTE IN A CIRCULATING FLUIDIZED-BED COMBUSTOR (CFBC)

The CFB experimental gasifier is a vertical tubular

furnace having an inner diameter of about 50 mm
and a height of 2000 mm. The fluidizing air is
supplied from the bottom and the circulating
particles are separated from air by a cyclone and
the particles returned to tubular furnace through the
loop seal downstream of the furnace. The gasifier is
provided with three external electric heaters (upper,
centre and lower) to allow internal control the
temperature. A heater is installed on the air supply
line to pre-heat the air.
A gas sampling port is provided in the middle of
the duct leading to combustion chamber. Flue gas,
tar and moisture generated by the gasification can
be sampled through the port. For monitoring the
gasification condition, the concentrations of CO, CO2
and O2 in the gas produced were measured with a
gas analyser.
The biomass sample to be gasified was supplied
to the gasifier through the screw feeder and rotary
valve. The feed rate from the screw feeder can be
adjusted by changing the rotational speed of the
driving motor through inverter control. A
photograph of the experimental gasifier is shown in
Figure 3. The rotary valve is located after the screw
feeder to seal off the gasifier. The chute downstream
side of the rotary valve is cooled by aeration to

prevent fusing of the biomass sample by

overheating.
The secondary combustion furnace is designed
to completely burn the combustible gases, such as
CO and H2, in the gas produced by gasification. A
mixture of the gas produced and air is burnt
constantly in a LPG burner. The gas cooler after the
secondary combustion furnace cools the exhaust
combustion gas with water to protect the equipment
on the downstream side. However, in this paper, the
combustible gases were not burned in the LPG;
rather they were diluted and discharged through the
stack. The gases discharged were monitored for the
emissions.
Immediately before the induction fan (IDF), an
automatic valve is connected to the differential
pressure gauge that measures the pressure in the
gasifier. The valve controlled the pressure in the
furnace during the experiment. In case of any
excessive pressure build up in the gasifier, the
pressure control valve can relieve the pressure. The
operation of these devices, heater temperatures, air
pre-heat temperature, in-gasifier pressure and
sample feed rate is via the main control panel. The
secondary combustion chamber is controlled via the
auxiliary control panel on the equipment side.
The temperatures at various locations in the CFB
rig were continuously monitored and recorded using
a Yokogawa Hybrid Recorder (HR 1300). The furnace
pressure, supply air flow rate were also monitored.
The concentrations of CO, CO2 and O2 in the produced
gas were also recorded using a calibrated Madur Flue
Gas Analyser (GA-40 plus). The resolution for
measurement of O2 and CO2 was 0.01%, and for the
measurement of CO and NO, 1 ppm.
Method
Approximately 200 g of the circulating alumina
powder were loaded into the gasifier (height of static
bed: approximately 100 mm), and a CFB formed by
pumping in air and increasing the in-furnace
temperature. Formation of the CFB was observed by
monitoring the change in the temperature of the loop
seal. It took about 3-4 hr before the CFB attained the
pre-set temperatures. The pre-set temperatures were
selected using thermogravimetric analysis (TGA) of
the devolatization behaviour of EFB and palm shell
waste powders. After the CFBC temperature reached
the specified temperatures, constant feeding of
biomass can be started with in-furnace temperature
and the gas composition (CO, CO2 and O2)
monitored.
After the temperatures had stabilized, the
primary air flow rate of the fluidizing air was varied
and the produced gas analysed from the sampling
port after the cyclone and as well as from the exhaust.
The biomass feed rate varied depending on the
biomass used. Using the 5 mm EFB particles the flow

Figure 3. Photograph of the experimental gasifier.

213

Malaysian Palm Oil Board 2006

JOURNAL OF OIL PALM RESEARCH 18 (JUNE 2006)

was not so smooth as when using to palm shell

particles. The EFB gave a pulsating flow while the
palm shell produced a smooth flow. However, the
different palm shell particle sizes showed different
flow behaviour through the screw feeder as shown
in Figure 4. The experimental conditions are being
summarized in Table 5.

RESULTS AND DISCUSSION

Combustion of EFB Particles
The 5 mm EFB particles were fed through the
screw feeder and rotary valve. The temperature at
the top of the CFB was about 650C. At this
temperature, the gas was sampled and its properties
analysed using gas chromatography. It was found
that the sampled produced gas contained about 5%
H2, 10% CO, several % CH4 and other hydrocarbon
(HC) components and its heating value exceeded

1000 kcal m-3. Based on these results, it was realized

that a usable gas containing H2, CO and CH4 can be
obtained by CFB gasification of EFB. The
concentrations of H2 and CH4 increased as the infurnace temperature became higher. However, the
gas heating value and carbon conversion ratio did
not change so much with in-furnace temperature.
Although the gas heating value exceeded 1000 kcal
m-3 with the lower air ratios, the carbon conversion
was less than 40%. The carbon conversion ratio was
low perhaps because a large quantity of EFB
remained unreacted owing to the dimensional
restrictions of the experimental apparatus and char
and soot were over generated.
Combustion of Palm Shell Waste Particles of 212300 microns
Oil palm shell waste was obtained from Kulai
Palm Oil Mill of the Federal Land Development
Authority (FELDA), Johor, Malaysia. It was crushed,

Figure 4. Flow behaviour of feeder for different palm shell waste particles.
TABLE 5. EXPERIMENTAL CONDITIONS IN THE CIRCULATING FLUIDIZED-BED (CFB) TEST RIG

EFB feed rate

Palm shell powder feed rate
Gasifying agent flow rates
Primary air
Secondary air
Loop seal air
Combustion air
Gas velocity
Gas residence time
In furnace temperature
214

2.4-60 kg hr-1
3.0-7.0 kg hr-1
1.0-3.5 m3 hr-1
1.0-1.5 m3 hr-1
0.1-0.75 m3 hr-1
10-28 m3 hr-1
1.0-2.0 m s-1
1.1-2.0 s
700C-850C

Malaysian Palm Oil Board 2006

THERMOCHEMICAL BEHAVIOUR OF EMPTY FRUIT BUNCHES AND OIL PALM SHELL WASTE IN A CIRCULATING FLUIDIZED-BED COMBUSTOR (CFBC)

grinded and sieved to different size samples. The

conditions for this experiment were similar to those
for the previous experiment. However, the particles
were too small and too easily pushed by the screw
feeder even at a low speed. This overfed the furnace.
The temperature at the top of the CFB increased
rapidly as the feeding started to reached about 709C
in a very short time as shown in Figure 5. In order to
evaluate the effect of the primary air flow through
the CFB, the flow rate was varied and the system
stabilized before any data were recorded. The
emission and experimental data are shown in Table
6.
Due to increased particle flow rate, the circulating
particle formed a dense bed. The palm shell particles
fed to this layer got into contact with the high
temperature circulating particles and were quickly
pyrolyzed. When a palm shell particle (PSP) came
into contact with a circulating particle at a higher
temperature, the heat conductance to the PSP
improved, and it was more quickly pyrolyzed. As
shown in Table 6, as the primary air flow was
increased, the CO concentration increased rapidly.
This was probably due to the very fine particle size

and low residence time in the CFB furnace. This

resulted in poor combustion and a high
concentration of CO was found near the CFB top.
The emission was a heavy white smoke representing
incomplete and enhanced CO formation.
Combustion of Palm Shell Waste Particles of 425
microns Particles
Keeping in view the inadequacy of the screw
feeder to control the feed rate of the 212-300 micron
particles, it was decided to feed bigger PSP of 425
and 600 microns. The 425 micron particles had a
satisfactory feeding behaviour. A typical heating
behaviour at the top of the CFB rig for 425 microns
particles is shown in Figure 6.
The feeding was started after about 60 min of
stabilization with the temperature at 790C. As soon
as the feeding started the devolatisation began and
temperature of the bed started to rise rapidly. In
about 30 min the temperature at the top of the
furnace had risen to about 864C. The effects of
variation in the primary air flow rate on emissions
are shown in Table 7.

720

Temperatures (C)

680

Feeding Start
640

600

560

10

20

30

40

Time (min)
Figure 5. Variation of circulating fluidized-bed (CFB) temperature with time.
TABLE 6. EMISSION QUALITY WITH PRIMARY AIR FLOW RATE

Sr.
No.
1
2
3

Primary air
flow rate (m3 hr-1)

Exhaust
temperature (C)

CO
(ppm)

NOx
(ppm)

CFB top temp.

(C)

1.0
1.5
2.0

185
160
150

2950
4819
7000

75
49
38

709
685
680

215

Malaysian Palm Oil Board 2006

JOURNAL OF OIL PALM RESEARCH 18 (JUNE 2006)

900
Feeding Start

Temperature (C)

800

700

600

500
0

15

30

45

60

75

90

100

130

Time (min)
Figure 6. Variation of circulating fluidized-bed combustor (CFBC) temperature with time.
TABLE 7. EMISSION QUALITY WITH PRIMARY AIR FLOW RATE

Sr.
No.
1
2
3
4
5

Primary air
flow rate (m3 hr-1)

Exhaust
temperature (C)

CO
(ppm)

NOx
(ppm)

CFB top temp.

(C)

1.0
1.5
2.0
2.5
3.0

105
193
150
143
150

1094
6092
4420
8500
5354

38
127
90
52
40

752
837
864
842
809

It can be inferred from Table 7 that the primary

air flow rate of 1 m3 hr-1 was good for the 425 micron
particles. Increasing the flow rate above that reduced
the particle residence and resulted in incomplete
combustion and hence, high CO emission. A similar
trend was also shown by the 600 micron particles
fluidization. However, the temperature at the top of
the CFB rose to about 888C and the CO emissions
was about 7000 ppm.
It is important to understand that as the primary
air flow rate is changed, the fluidization behaviour
in the gasifier also changes abruptly, causing the
devolatisation behaviour to change. This can be
shown by plotting the average gasifier temperature
with the height of the gasifier as shown in Figure 7.
This shows a sudden jump in temperature after
the middle section of the gasifier and which is
expected to change the pyrolysis and emission
behaviour.
The gas residence time in the furnace calculated
from the air feed rate is short 1.1 to 2.0 s. The
residence time is even shorter when the gas flow rate

is increased by gasification of PSP. Particularly, under

high temperature conditions, the residence time may
be extremely short - less than 1 s. Although the
sample was to be gasified by the quick heat
decomposition, that is a feature of the CFB, the
reaction did not advance much because of the
insufficient resident time and soot and char were
generated in quantity. Soot and char were also found
in the duct of the apparatus. This proved insufficient
residence time under the high temperature
conditions
In a CFB furnace, heat decomposition of the
matter fed into the furnace advances quickly. In
addition, since the particle concentration in the
furnace is high and the efficiency of contact of the
circulating particles with the feedstuff and gas also
high, high heat conductivity and the resultant high
speed progress of the reaction can be expected. These
features of the CFB are regarded as one of the factors
for low tar generation. In this experiment, the PSP
sizes 212-300 and 425 microns were sufficiently small
for their quick heat decomposition. The palm shell
216

Malaysian Palm Oil Board 2006

THERMOCHEMICAL BEHAVIOUR OF EMPTY FRUIT BUNCHES AND OIL PALM SHELL WASTE IN A CIRCULATING FLUIDIZED-BED COMBUSTOR (CFBC)

Figure 7. Variation of circulating fluidized-bed combustor (CFBC) average temperature with gasifier height.
powders used were therefore characterized by easy
heat decomposition and low tar generation.
However, excessive tar generation was found for
the PSP of 600 micron. Reduced tar generation is one
of major goals in biomass gasification systems. It is
necessary to identify the optimum conditions for the
experiments. Good heat transfer between the
fluidizing particles and palm shell waste can be
achieved by having the appropriate amount of
alumina particles in the CFBC and using a low
primary air flow rate. In order to reduce the tar
formation, it is suggested to co-fire palm shell waste
with coal.

The authors would like thank the Malaysian

Government Commonwealth Secretariat, Public
Service Department Malaysia, Putrajaya for funding
the Ph.D programme. Thanks are due to the Japanese
Government and SIRIM Environment and
Bioprocess Technology Centre for permission to use
their CFB experimental facility. Thanks are also due
to Universiti Teknologi Malaysia (UTM) for
providing financial support for developing and
building the cold CFB test facility in the Department
of Mechanical Engineering, UTM, Skudai, Johor.

CONCLUSION

REFERENCES

The gasification experiment was successfully

conducted using the CFB. Gas containing H2, CO and
CH4 was produced with a heating value of 1000 kcal
Nm-3 or more under proper conditions. The tar
generation was low which may be due to quick heat
decomposition of the EFB on the CFB. On the other
hand, the generation of soot and char was high which
may also be the cause of the low tar generation. The
dependence of the emission quantity on the infurnace temperature and primary air flow was
identified and relationships between them derived.
By this research, we have examined the gasification/
combustion properties of palm shell waste powder
which is found in abundance in Malaysia. Also
fundamental data on gasification of palm shell waste
has been obtained using a CFB.

ANI, F N (1992). Thermal characteristics of oil palm

fruit shells. Proc. of the Second International Energy
Conference - Energy from Biomass Residues. Paper 7.
Kuala Lumpur, Malaysia.

217

ACKNOWLEDGEMENTS

BHATTACHARYA, S C (1997). State of the art of

biomass gasification. Proc. of the International
Symposium on Advances in Alternative and Renewable
Energy. Organized by Universiti Teknologi Malaysia,
Johor, Malaysia. p. 50-57.
BRIDGWATER, A V (1994). Advances in
Thermochemical Biomass Conversion. Volume 1, Blakie
Academic & Professional, New York. p. 397-404.

Malaysian Palm Oil Board 2006

JOURNAL OF OIL PALM RESEARCH 18 (JUNE 2006)

BUI, T; LOOF, R and BHATTACHARYA, S C (1994).

Multi-stage reactor for thermal gasification of wood.
Energy, 19(4): 397-404.
DEVI, L; PTASINSKI, K J and JANSSEN, F J J (2003).
A review of the primary measures for tar elimination
in biomass gasification processes. Biomass &
Bioenergy, 24: 125-140.
DORNBURG, V and FAAIJ, A (2001). Efficiency and
economy of wood fired biomass energy systems in
relation to scale regarding heat and power generation
using combustion and gasification technologies.
Biomass & Bioenergy, 21: 91-108.

KATE, R; DESHMUKH, G and TANDALE, M S

(1997). Energy from biomass a perspective under
Indian conditions. Proc. of the International Symposium
on Advances in Alternative and Renewable Energy.
Organized by Universiti Teknologi Malaysia, Johor,
Malaysia. p. 63-69.
PIERRE JOULAIN (2004). The efficient combustion
of biomass and its derivatives: a major challenge for
the future. Proc. of the Joint International Conference
on Sustainable Energy and Environment (SEE).
Thailand. p. 362-367.
ROGNER, H H and POPESCU, A (2000). An
introduction to energy. World energy assessment energy and challenge of sustainability. United Nations
Development Programme, New York. p. 30-37.

218

Malaysian Palm Oil Board 2006