Technical Assessment and Flue Gases Emission

Monitoring of an Oil Palm Biomass–Biogas

Cofired Boiler
Abu Bakar Nasrin , Soh Kheang Loh, Mohamad Azri Sukiran, Nurul Adela Bukhari, and
Astimar Abdul Aziz
Engineering and Processing Research Division, Malaysian Palm Oil Board (MPOB), No. 6 Persiaran Institusi, B.B. Bangi, 43000,
Kajang Selangor, Malaysia; [email protected] (for correspondence)
Published online 21 February 2019 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ep.13189

Biogas generated from anaerobic digestion of palm oil mill Due to its high organic content with the biological oxygen
effluent (POME) can be a supplementary fuel in palm oil mill demand and chemical oxygen demand >30 g/L and >60 g/L,
biomass boilers. The effects of biogas cofiring with different types respectively, the industry deploys anaerobic digestion method for
of oil palm biomass-based conventional boiler fuel on the boiler POME treatment [4]. During this process, biogas is generated
efficiency and performance were investigated. Three different through a multistage metabolic reaction involving different micro-
types of fuel input were assessed, namely (1) a mixed biogas- organisms for degradation of organic matters, and emitted freely
mesocarp fiber (MF)-empty fruit bunch fiber (EFBF) (biogas- at the open ponds or digester tanks to the atmosphere [5]. For
MF-EFBF) fuel, (2) a mixed biogas-MF (biogas-MF) fuel, and each tonne (t) of fruit bunch processed, an approximate of 0.65 t
(3) MF only for a 40 t/h water tube boiler. It was found that the of POME and 18.2 m3 of biogas is generated [6]. Besides, many
boiler efficiencies and particulate emissions varied from 72 to studies have consistently reported that an approximately 28 m3 of
83% and 0.1375 to 0.6050 g/m3, respectively. The highest boiler biogas was generated for every m3 of POME digested containing
efficiency with the lowest particulate emission was attainable mainly methane (CH4, ~60–70%) and carbon dioxide (CO2,
when cofiring biogas with MF. This fuel mix combination con- ~35%) [4,6,7]. Hence, in 2017, 65.66 million t of POME and 1839
tributed to reduction of the fed solid biomass fuel into the boiler million m3 of biogas were generated from 101.02 million t of fresh
furnace and resulted in lesser incomplete combustion and ash fruit bunches processed in 454 palm oil mills nationwide [8]. CH4
generation. The findings demonstrated that the fuel selection and CO2 in biogas, both of which are greenhouse gases causing
and proper mixture of the type of biogas-biomass fuel are global warming, with CH4 about 25 times the global warming
among the important factors to ensure high palm oil mill boiler potential of CO2 [4,7]. This situation contributes to adverse effect
efficiency and low particulate emission in meeting the permit- on the sustainable image and marketability of the palm oil, partic-
ted limit as stipulated in the Malaysian Environmental Quality ularly in developed countries. Thus, the industry is urged to cap-
(Clean Air) Regulations, 2014. © 2019 American Institute of Chemi- ture and use the biogas as a renewable fuel for both the economic
cal Engineers Environ Prog, 38:e13189, 2019
and environmental benefits.
Keywords: biogas-biomass fuel, cofiring, palm oil mill efflu- Several commercial approaches on biogas utilization are
ent, particulate emission, biomass boiler
adopted currently e.g. cofiring, electricity generation for on-site
uses and grid connection, rural electrification, use in palm oil
INTRODUCTION refining and upgrading to bio-compressed natural gas [7,9,10].
Malaysia is the world’s second largest producer and Biogas cofiring consists of burning biogas simultaneously with
exporter of palm oil [1,2]. The industry produced 19.92 million other forms of oil palm biomass in the palm oil mill boiler [11].
tonnes of crude palm oil which represents 30% global produc- The biogas is captured in the closed anaerobic digester, sup-
tion in 2017 from 5.81 million hectares of oil palm plantations plied and burnt in the biomass boiler using gas burner
[1]. Besides, the industry generates a huge amount of non-oil (Figure 1). The burner is normally installed either at the tubing
biomass as solid and liquid by-products. At palm oil mills side of boiler wall or front side closed to fuel feeder. This pro-
where fresh fruit bunches are processed for oil, other residues cess offers an immediate and low-cost investment option for a
comprising the mesocarp fibers (MFs), palm kernel shells, direct and clean conversion of biogas to steam and electricity
empty fruit bunches and palm oil mill effluent (POME) are via an existing cogeneration plant available at palm oil mills.
produced. Solid biomass is the main source of fuels used to The cofiring of biogas has a great potential to reduce the bio-
produce steam and generate electricity using a combined heat mass fuel intake of the boiler, mainly palm shells which has
and power plant in palm oil mills [3]. Many opportunities still pretty high economic value for commercial applications. A sur-
exist in exploiting the palm oil milling by-products as energy, vey by the Malaysian Palm Oil Board (MPOB) conducted
especially the biogas from POME. under the Entry Point Project No. 5 of National Key Economic
Area (NKEA-EPP5) has shown that there are at least 12 palm
oil mills nationwide operating a biomass–biogas cofired
© 2019 American Institute of Chemical Engineers boiler [7,9].

1 of 8 September/October 2019 Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Raw palm oil
Anaerobic closed Post POME
mill effluent Cooling Mixing
digester treatment
(POME)
Treated
POME
Biogas

Biomass Gas storage

Heat and power for mill boiler (if any) Flaring

Figure 1. Cofiring of biogas in palm oil mill biomass boiler [11].

For decades, palm oil mills are self-sufficient in energy for is used to generate 1250 kWh electricity using back pressure
milling process through combustion of a typical palm oil mill bio- turbine for mill’s own consumption.
mass boiler. Since the process is more for biomass disposal, the The performance of the biogas-biomass cofired boiler in terms
main concerns are on the efficiency of the boiler and emission of its efficiency, particulate emissions, flue gases compositions
generated from the combustion. The flue gases emissions from and dark smoke was assessed. Three different types of boiler fuel
the boilers are monitored by the Malaysian Department of Envi- were studied, namely two tests with fuel mixture and one test as
ronment (DOE) under the Environmental Quality (Clean Air) Reg- sole fuel. In Test 1, biogas was cofired with MF and empty fruit
ulation 2014, focusing on black smoke and particulate emissions. bunch fiber (EFBF) at MF:EFBF of 70:30 (weight ratio). Biogas
Under Regulation 13, the maximum limit of particulate emissions was cofired with MF in Test 2 while Test 3 employed MF as a sole
from solid fuel boiler is 0.15 g/m3 after corrected to 12% CO2. fuel (serves as positive control which represents current practice
The regulation has been enforced with immediate effect to new in palm oil mill). Table 2 shows the different types of fuel and mix-
boilers with a 5-year grace period for existing boilers [12,13]. ing ratio for the cofiring monitoring purpose.
Therefore, the maximum limit of 0.4 g/m3 (expressed in terms of The on-site monitoring was conducted during the steady-
12% CO2) enforced earlier under the Environmental Quality state operation of the boiler within 4 h, and required data were
(Clean Air) Regulations 1978 is still applicable until 2019. The cor- taken at every 20-min interval. The boiler operating parame-
rection factor either using 12% CO2 or O2 (typically 7% in the US ters during the monitoring were set as follows: (1) pressure =
and 11% in Europe and Asia) is defined by regulation [14,15]. By 32 barg, (2) steam flow rate = 28 t/h and (3) steam tempera-
using 12% CO2 correction factor, the measured concentration is ture = 285
C. The production of oil palm biomass and POME
corrected to a value corresponding to the flue gases having a was estimated based on weight ratio (wt/wt, %) of MF, PKS,
CO2 concentration of 12%. This approach is deployed to elimi- empty fruit bunch and POME to fresh fruit bunches processed
nate the effect of pollutant emission and flue gases dilution by air. which is 13%, 6%, 23% and 65%, respectively [7,25]. The indi-
As long as the same correction equation is used, the resulting data vidual fuel, i.e. MF and EFBF was collected and analyzed
will be comparable among them and the emission standard to determine their physico-chemical properties (wet basis)
applied will be uniform [15]. via proximate and ultimate analyses as described by Nasrin
To date, many studies have been conducted to determine the et al. [25] (Table 3). The biogas flow rate was taken from the
efficiency and emission of the conventional oil palm biomass control panel and flowmeter installed at the biogas plant and
fired boilers. The boilers efficiencies vary from 65 to 87% [16–19], its gas composition was measured using a portable gas ana-
while the particulate emissions range 1.11–11.6 g/m3 [20–23]. To lyzer (Brand MRU Optima 7 Biogas) at every 20-min interval
our knowledge, there is no detailed study yet in assessing the throughout the monitoring period.
performance of the biomass–biogas cofired boiler although this
has been commercially practiced for many years. Therefore, this Boiler Efficiency Assessment
article examined the effects of cofiring biogas with different types The efficiency of the boiler (μ%) cofiring biogas with differ-
of oil palm biomass-based conventional boiler fuel on palm oil ent biomass fuels was calculated based on Equation (1) by
mill biomass boilers. The boiler performance and the resulting Husain et al. [16].
particulate emissions were assessed. The technical and environ-
mental benefits of this approach were compared to those from ms ðh2 − h1 Þ
the typical biomass boilers used in palm oil mills. μ% = 

 ð1Þ
mf CV f + mg CV g
MATERIALS AND METHODS
where ms is the steam flow rate (kg/h), h1 is the enthalpy
Project Approach of feedwater (kJ/kg), h2 is the enthalpy of steam (kJ/kg), mf is
A biomass–biogas cofired boiler operated in a palm oil mill the biomass fuel flow rate (kg/h), LHVf is the gross calorific
situated in Negeri Sembilan was selected for on-site boiler per- value (CV) of biomass (kJ/kg), mg is the flow rate of biogas
formance monitoring, data and samples collection. Table 1 (m3/h), and CVg is the CV of biogas (kJ/m3).
depicts the technical data of the mill, the cogeneration and The assumptions as follows were used in the calculation:
biogas plants (capacity, design, operating parameters, etc.) for
this case study. The mill operates a 40 t/h water tube biomass • The mill used all biogas generated from the covered lagoon
boiler using oil palm biomass and biogas as fuel to process digester for cofiring.
60 t/h fresh fruit bunches. Biogas generated from anaerobic • 70–100% of MF generated was used as fuel for cofiring.
digestion of POME is captured using covered lagoon technol- • 20% of EFBF produced was used as fuel for cofiring.
ogy and cofired in the biomass boiler using gas burners as • CV of biogas is 20,000 kJ/m3 [8].
illustrated in Figures 2 and 3. During normal operation, the
boiler uses about 80% of the total MF produced and 100% of Stack Emissions Monitoring
the captured biogas to produce 28 t/h superheated steam at Sampling of particulate emissions from the chimney was
32 barg working pressure and 285
C. The high pressure steam conducted using an isokinetic method based on the Malaysian

Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep September/October 2019 2 of 8
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 1. The technical data of the monitoring site used for this study.

Parameter Description/mean value Parameter Description/mean value

Mill Steam turbine
Installed capacity 60 t/h No. of unit 2
Average annual FFB processed 288,521 t/a Designed capacity
Unit 1 2000 kW
Unit 2 1750 kW
Operating parameters
Electricity generation 1250  25 kWh
Inlet pressure 31.0  1.0 barg
Outlet pressure 2.9  0.3 barg

Boiler Biogas plant

No. of boiler (type) 2 unit (water tube boiler) Technology used Covered lagoon
Lagoon material Geo-membrane
Designed steam flow rate & No. of closed digester 1
pressure Total digester capacity 32,000 m3
Boiler 1 (stand by) 40 t/h, 21 barg Designed biogas flow rate 1000 m3/h
Boiler 2 (operation) 40 t/h, 38 barg Hydraulic retention time Min. 27 days
Operating parameters Estimated biogas yield 22–27 m3/m3 POME
Steam flow rate 28.0  1.0 t/h
Working pressure 32.0  1.0 barg Internal (auxiliary) load 50 kWh
Steam temperature 285  10
C

Abbreviations: FFB, fresh fruit bunch; POME, palm oil mill effluent.

Figure 2. Biomass–biogas cofiring system used for this study [24].

Standard MS 1596:2003 (determination of concentration and Ringelmann chart based on BS 2742(1969). The observation
mass flow of particulate matter in flue gas for stationary source was conducted for a period of 1 h at every 15-s interval. The
emissions). Dark smoke observation was carried out using the compositions of flue gases were sampled and analyzed using a

Figure 3. Covered lagoon anaerobic digester and biogas burners installed at palm oil mill biomass boiler. [Color figure can be
viewed at wileyonlinelibrary.com]

3 of 8 September/October 2019 Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 2. Fuel type and estimated ratio used for the monitoring purpose.

Mixing ratio (wt/wt, %)

Test Type of fuel Biomass Biogas
1 Biogas, mesocarp fiber (MF) and empty 70 (MF):30 (EFBF) As produced
fruit bunches fiber (EFBF)
2 Biogas and MF MF As produced
3 MF only MF –

Table 3. Characterization of physico-chemical properties of followed by sole fuel of MF and lastly, mixed biogas-MF-EFBF
mesocarp fiber and empty fruit bunch fiber. fuel. The cofiring of biogas with MF had slightly increased the
efficiency of the boiler (~2.5%) compared to burning just the
MF alone. However, the use of EFBF in the fuel mix contrib-
Standard Equipment (brand uted to the lowest boiler efficiency. This probably was due to
Analysis method and model) inconsistent size and high moisture content of EFBF which
Calorific value ASTM D5865 LECO AC 600 could result in non-homogeneity of fuel mixing with MF, sub-
Proximate analysis ASTM 5142 LECO TGA 701 sequently affecting the fuel feeding consistency and poor dis-
Ultimate analysis ASTM D5373 LECO CHNS 628 tribution on the furnace grate. Average size of MF and EFBF
was 2 cm and 10 cm in length, respectively (Figures 5 and 6).
This had led to incomplete combustion and high load of fuel
portable flue gas analyzer (Brand Quintox KM 9106) equipped was required to heat up and maintain steam demand and pres-
with electrochemical sensors. The flue gases were sampled for sure. According to Kun and Abdullah [22], pre-conditioning of
a period of 1 h at every 10-min interval. Figure 4 summarizes the biomass fuel is required for better combustion in the boiler
the methodology used in this study. furnace and higher boiler efficiency.
The finding was comparable to several other studies on the
RESULTS AND DISCUSSION efficiencies of oil palm biomass fired boilers [16–19]. Besides,
the specific energy used to produce 1 kg steam varied from
Production Rate and Fuel Properties of Biomass and 3393 to 3890 kJ which corresponded to the findings obtained
Biogas by Nasrin et al. [18]. Therefore, a direct positive impact in
Based on weight ratio (wt/wt, %) of oil palm biomass and using biogas as a supplementary fuel to boiler efficiency is
POME to fresh fruit bunches processed at 60 t/h capacity, an dependent on type of biomass fuel used. A possible explana-
approximate of 7.8 t/h MF, 13.8 t/h raw empty fruit bunches, tion for this is due to the following factors: (1) palm oil mill
3.6 t/h palm kernel shells and 46.2 m3/h POME were produced boiler is primary designed for biomass fuel, mainly a mixture
by the mill. 11.04 t/h EFBF was produced by shredding and of MF and palm kernel shells and (2) the captured biogas pro-
pressing of raw empty fruit bunches. Biogas production from vides direct replacement to palm kernel shells which has better
anaerobic digester was monitored and measured, and found in economic value. Factors such as air-fuel ratio and designing a
the range of 850–900 m3/h. Due to a full utilization of the cap- new boiler specifically for biomass–biogas fuel can be consid-
tured biogas, 100% of palm kernel shells is displaced and sold ered to further optimize the cofiring process.
to the nearby factory. The generated MF and EFBF contain
~37–47% of moisture, which have been typically used as a fuel Stack Emissions
in palm oil mill boilers. The biomass and biogas physico-
Table 6 summarizes the particulate emissions generated
chemical and fuel properties are summarized in Table 4. In the
from the three different fuels used for combustion test. It
study, the captured biogas containing approximately 59% CH4,
showed a varying particulate emission from 0.1375 to
38% CO2 and 2400 ppm hydrogen sulfide (H2S) was at accept-
0.6050 g/Nm3 after corrected to 12% CO2 (Figure 7). All results
able level for cofiring in biomass boiler and no pretreatment
complied with the DOE limit except for the particulate emis-
was required.
sion from the mixed biogas-MF-EFBF fuel which was higher
than the permitted limit of 0.4 g/m3 under the Malaysian Envi-
Boiler Efficiency ronmental Quality (Clean Air) Regulations 1978. Nevertheless,
Table 5 summarizes the efficiency of the boiler which used these findings were relatively lower compared to the previous
three different fuels during cofiring. The highest efficiency was studies conducted on the particulate emission from mixed MF-
obtained when the boiler used a mixed biogas-MF fuel, PKS fuel, with mean values ranging from 1.11 to 7.25 g/m3

Boiler efficiency study (direct method):

- Fuel flow rate; - Pysico-chemical properties of biomass Stack monitoring:
- Biogas composition - Particulate emission
Isokinetic sampling : MS
1595 : 2003
- Dark smoke (Ringglemen
Test 1: biogas - MF - EFBF Chart- BS 2742(1969)
- Flue gases analysis
Fuels Water tube biomass boiler
used Test 2: biogas - MF

Test 3: MF only

Figure 4. Summary of the methodology used for this study. Abbreviations : MF, mesocarp fiber; EFBF, empty fruit bunch fiber.

Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep September/October 2019 4 of 8
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 4. Oil palm biomass and biogas fuels used for cofiring
in palm oil mill and their properties.

Empty fruit
Mesocarp bunch
Parameter fiber (MF)* fiber (EFBF)* Biogas
Flowrate (t/h) 7.80 11.04 855  40
(except for 13.80 (Raw
biogas (m3/h)) EFB)
Moisture (%) 37.2  5.2 47.0  3.2 –
Volatile matter (%) 46.92  0.16 45.41  0.25 –
Ash (%) 2.70  0.12 1.78  0.15 –
Fixed carbon (%) 10.18  0.32 8.80  0.32 –
Calorific value 11,888  216 10,716  236 20,000
(kJ/kg)
(except for
biogas (kJ/m3))
Carbon (%) 40.48  3.43 28.32  12.08 –
Hydrogen (%) 6.87  0.65 8.15  1.31 –
Nitrogen (%) 0.47  0.0 0.42  0.52 –
Sulfur (%) 0.11  0.02 0.05  0.04 –
Oxygen (%) 52.08  2.76 63.07  11.33 – Figure 5. Mesocarp fiber (MF). [Color figure can be viewed at
Methane (%) – – 58.9  2.1 wileyonlinelibrary.com]
Carbon dioxide (%) – – 38.2  3.2
Hydrogen sulfide – – 2402  210
(%)
into the boiler furnace which would have resulted in higher
ash generation and subsequently more particulate load to the
*Biomass fuels were analyzed as received. multi-cyclone for dust collection. Hence, cofiring biogas with
MF had contributed to the lowest particulate emission among
the mixture of fuels used. The amount of soot or particulate
released is proportional to the total mass of fuel burnt [27].
[20,23,26]. The lowest particulate emission was attainable
A higher boiler efficiency which reflects a better combus-
when cofiring biogas with MF. This was due to a significant
tion performance in the boiler furnace also can contribute to
reduction of the otherwise supposedly fed solid biomass fuel
lower particulate emission. In this case, the CO2 content in the
flue gases increases indicating a complete combustion, and
Table 5. Efficiency of the biomass–biogas cofired boiler. thus the resulting particulate emission would be lower when
expressed in terms of 12% CO2 [21,22,28]. Incomplete combus-
tion of biomass contributes about 0.2% to the particulate emis-
Test 1 Test 2 Test 3
sion [29]. Therefore, the percentage of CO2 in the flue gases
(biogas- (Biogas- (7.25–7.37%) from a particular fuel type is a vital parameter
Parameter/fuel used MF-EFBF) MF) (MF only) that determines the final particulate emission coming out from
a boiler. In Tests 1 and 2, higher CO2 contents in the flue gases
FFB processed (t/h) 60 60 60
Flowrate of fuel used:
Biogas (m3/h) 855 855 –
MF (kg/h) 5460 6240 7800
EFBF (kg/h) 2208 – –
Total biomass fuel (kg/h) 7668 6240 7800
Boiler working pressure 32 32 32
(barg)
Steam flow rate (t/h) 28 28 28
Steam temperature (
C) 285 285 285
Feed water enthalpy, 125.7 125.7 125.7
hf (kJ/kg)
Steam enthalpy, 2948.35 2948.35 2948.35
h1 (kJ/kg)
Total energy input 108.92 94.99 97.36
(kJ/h × 106)
Energy in steam 79.03 79.03 79.03
(kJ/h × 106)
Efficiency (%) 72.56 83.20 81.17
Specific energy used 3890 3393 3477
to produce 1 kg of
steam (kJ/kg)

Abbreviations: FFB, fresh fruit bunch; MF, mesocarp fiber; Figure 6. Empty fruit bunch fiber (EFBF). [Color figure can be
EFBF, empty fruit bunch fiber. viewed at wileyonlinelibrary.com]

5 of 8 September/October 2019 Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 6. Boiler’s stack gases and particulate emissions.

Parameter Test 1 Test 2 Test 3

Type of fuel Biogas-MF-EFBF Biogas-MF MF only
Stack temperature (
C) 246.5  0.7 259.9  18.6 274.0  2.8
CO2 (%, v/v) 7.25  0.07 7.37  1.03 5.80  0.14
O2 (%, v/v) 12.25  0.35 12.17  1.13 13.50  0.14
N2 (%, v/v) 80.5  0.4 79.97  1.01 80.85  0.07
Stack gas velocity (m/s) 14.45  0.50 14.09  1.99 14.34  0.99
Actual particulate emission (PE) (g/m3) 0.3916  0.0002 0.0846  0.0481 0.1001  0.0157
PE corrected at 12% CO2 (g/m3) 0.6050  0.0015 0.1375  0.028 0.2080  0.0327

Abbreviations: MF, mesocarp fiber; EFBF, empty fruit bunch fiber.

0.7 separate study, the smoke was monitored continuously using a

smoke density sensor (in unit of opacity) during boiler opera-
0.6
tion [30]. It was found that the utilization of biogas can
particulate emission, g/m3

0.5 improve the boiler stack opacity by about 20% consistently

compared to about 40% without biogas [13]. Under Regulation
0.4 12 of the Environmental Quality (Clean Air) Regulation 2014,
the dark smoke of the solid fuel fired boiler shall be <shade
0.3
1 of the Ringelmann chart and <20% opacity. With the grace
0.2 period given by DOE for full compliance, this will be applica-
ble to all the existing boilers from 2019 onwards.
0.1 Besides oxygen (O2), CO2 and nitrogen (N2), other gases
detected from the boiler were carbon monoxide (CO), nitro-
0
Test 1 (biogas-MF-EFBF) Test 2 (biogas-MF) Test 3 (MF only) gen oxide (NO), nitrogen dioxide (NO2) and sulfur dioxide
(SO2) (Table 7 and Figure 8). Their emissions levels at all
Actual particulate emission (PE) PE corrected at 12% CO₂ tested conditions were relatively low, particularly for NO2 and
Figure 7. Actual particulate emission and particulate emission SO2. Though there was a concern on the effect of the deriving
after corrected to 12% CO2. H2S in the biogas to the combustion chamber, in this case H2S
would be fully oxidized into its combustion products, SO2
before it has any chance of corroding the metals. In addition,
the content of H2S and S present in biomass–biogas fuel were
found insignificant in generating high level of SO2 during com-
were observed from the mixed biomass–biogas fuels com- bustion. The low SO2 in flue gases was also attributed to the
pared to using a single biomass fuel (Test 3, 5.80%). Assuming formation of alkaline (e.g., potassium, K; natrium, Na; calcium,
that all MF and EFBF have been combusted, the higher CO2 Ca and magnesium, Mg) sulfates during biomass combustion
level in the mixed biomass–biogas fuels was probably came in the presence of S [31]. As reported by Mitchell et al. [27], the
from the higher CO2 in biogas. The higher CO2 level found in SO2 was retained as alkaline ash and only <22.5 ppm
the flue gases of Test 1 and Test 2 also showed that biogas (58.95 mg/m3) SO2 was detected from the tested fuels.
present in the mixed fuels would facilitate combustion process CO is one of the products of incomplete combustion. The
for better and complete combustion. Equations (2) and (3) quality of combustion could be indicated by the amount of CO
show the gas balance during the combustion of the mixed emitted, with a lower emission indicating a higher conversion
biomass–biogas fuels. of C into CO2, resulting in a complete combustion. According
to Demirbas [32], due to the nature of biomass, complete com-
C6 H12 O6 + 6O2 ! 6CO2 + 6H2 O ð2Þ bustion is not achievable at any biomass burning conditions.
CH4 + 2O2 ! CO2 + 2H2 O ð3Þ The level of CO in the flue gases measured at a range of
86–102 ppm (98–117 mg/m3) from all tests was very low com-
Based on Equations (2) and (3), cofiring biogas in biomass pared to the concentration limit of CO < 1000 mg/m3 set by
will emit an additional 1 mol of CO2 compared to sole biomass the DOE. This indicated that the combustion process in the
fuel. Co-combustion of two types of biomass (MF and EFBF) palm oil mill biomass boiler and the air-fuel ratio were well
in the presence of biogas tends to emit lesser CO2 but more optimized by the mill. The mill has installed an inverter unit to
CO as EFBF is less efficient as a fuel. Hence, the concentration control the air-fuel ratio to be proportionate to the required
of CO2 in decreasing order is Test 2 > Test 1 > Test 3. While pressure and steam of the boiler.
for CO concentration, the order is Test 3 > Test 1 > Test The CO emitted was rather similar in all the three different
2 (Table 7). fuel mixes. However, the combustion of biomass–biogas
Other factor that determines particulate emission of a bio- mixed fuel generated much lower CO compared to the com-
mass boiler is the efficiency of the employed dust collector. bustion of sole biomass fuel in the boiler. Compared to biogas,
Kun and Abdullah [22] proposed that a multi-cyclone with biomass is highly potential for incomplete combustion. The
>91% efficiency could meet the set particulate emission presence of biogas in the biomass boiler would reduce amount
by DOE. of biomass fuel, therefore facilitate complete combustion pro-
The boiler’s dark smoke observations showed that the flue cess and minimize CO formation generated from biomass–
gases emissions from all the three different fuel mix were biogas fuel. This was reflected by the higher CO2 generated
< shade 1 of the Ringelmann chart throughout the 1-h monitor- from the combustion of biomass–biogas fuel (7.25–7.37%)
ing period. All the results were within the limit stated in the compared to the sole biomass fuel (5.80%) (Table 7). High
Environmental Quality (Clean Air) Regulations 1978. In a content of noncombustible CO2 in biogas also increases CO2

Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep September/October 2019 6 of 8
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Table 7. Flue gases compositions.

Test 1 Test 2 Test 3

Gas (ppm, v/v) (biogas-MF-EFBF) (biogas-MF) (MF only) Allowable limit*, (mg/m3)**
89.7  12.2 86.3  41.0 102.2  23.3
CO (1000)
(102.76  13.98) (98.87  46.97) (117.08  26.69)
89.8  7.5 37.3  32.9 46.3  7.3 –
NO
(110.22  9.21) (45.78  40.38) (56.83  8.96)
1.5  1.0 1.0  0.0 2.2  1.1 –
NO2
(2.82  1.88) (1.88  0.0) (4.14  2.07)
91.3 38.3 48.5 –
NOx
(113.04) (47.67) (60.97)
7.5  6.0 13.3  10.4 22.5  8.2 –
SO2
(19.65  15.72) (34.85  27.25) (58.95  21.48)

Abbreviations: MF, mesocarp fiber; EFBF, empty fruit bunch fiber

*As stipulated in Regulation 13 of the Environmental Quality (Clean Air) Regulation 2014.
**Unit devoted to values in bracket.

120 1978, the concentration limit of CO <1000 mg/m3 has been

specified in the recent Environmental Quality (Clean Air) Reg-
100 ulation 2014 (Regulation 13).

80
CONCLUSION
ppm (v/v)

Cofiring biogas with different types of oil palm biomass-based

60
conventional boiler fuel in a palm oil mill boiler showed varying
performance efficiencies. The highest boiler efficiency (83%) and
40
the lowest particulate emission (0.1375  0.028 g/m3) were
obtained from cofiring biogas with MF. Besides reducing the car-
20
bon footprint of crude palm oil production, the use of biogas in
0
palm oil mill boiler reduced or saved a significant amount of
Test 1 (biogas-MF-EFBF) Test 2 (biogas-MF) Test 3 (MF only) boiler-fed biomass fuel, in particular palm kernel shells which can
CO NO NO2 SO2
be a premium fuel for other nearby industries. Due to poor physi-
cal characteristics, the use of EFBF as a supplementary fuel poten-
Figure 8. Flue gases compositions of three different fuel of tially will have adverse effects to boiler performance and
the combustion test. particulate emissions. The flue gases content is closely related to
biomass fuel properties. With the proper fuel mix selection and
optimized cofiring process, the operation of a biomass–biogas
cofired boiler in palm oil mill could provide techno-economic and
in the flue gases. Interestingly, the lowest CO level obtained environmental benefits for the palm oil milling process.
when the boiler combusted the mixed biogas-MF fuel was in
consistent with the earlier findings on its highest boiler effi- ACKNOWLEDGMENTS
ciency using the same fuel mix. The authors thank the Director-General of MPOB for per-
The NO and NO2 emissions are known to be related to and mission to publish this article. Thanks are also due to all indi-
indicative of the presence of N in the fuel, phase of combus- viduals, industrial partners, biogas technology provider and
tion and thermal oxidation leading to nitrogen oxides (NOx) at palm oil miller who were involved in this study. The study was
higher temperature (>1300
C) [27,33]. This high temperature supported and funded by the MPOB.
range is unlikely reached judging at the current practices in
palm oil mill boilers. In this study, the boiler temperature of LITERATURE CITED
665–750
C was recorded. In addition, the N content in the bio- 1. Kushairi, A., Loh, S.K., Azman, I., Hishamuddin, E., Ong-
mass fuels used was also relatively low, and therefore insignifi- Abdullah, M., Izuddin, Z.B.M.N., Razmah, G., Sundram, S., &
cant for NOx emission. Nevertheless, the highest NO content Parveez, G.K.A. (2018). Oil palm economic performance in
in the flue gases was found in Test 1. It was due to the use of Malaysia and R&D progress in 2017, Journal of Oil Palm
two types of biomass fuel simultaneously, namely MF and Research, 30, 163–195.
EFBF in adequate quantity. This situation may also be contrib- 2. Ooi, Z.X., Teoh, Y.P., Kunasundari, B., & Shuit, S.H. (2017).
uted by the Fenimore mechanism where prompt NOx forma- Oil palm frond as a sustainable and promising biomass source
tion could occur from reactions involving elemental N from air in Malaysia: A review, Environmental Progress & Sustainable
and hydrocarbon fragments, preferably at lower temperatures Energy, 36, 1864–1874.
[34,35]. In total, the NO and NO2 emitted from the flue gases 3. Abnisa, F., Wan Daud, W.M.A., & Sahu, J.N. (2013). Pyrol-
of the biomass–biogas cofired boiler represented >95% and ysis of mixtures of palm shell and polystyrene: an optional
<5% of the NOx released to the atmosphere. These figures method to produce a high-grade of pyrolysis oil, Environ-
were comparable to the values obtained via other biomass mental Progress & Sustainable Energy, 33, 1026–1033.
combustion such as wood and energy crops [30], i.e. about 4. Ahmad, A., & Wahid, Z.A. (2015). Immobilized cement
90% and 10%, respectively of the total NOx. Although there kiln dust enhances biomass and neutralizing of palm oil
was not any permitted limits for CO, NO, NO2 and SO2 emis- mill effluent for biogas production, Environmental Pro-
sions under the Environmental Quality (Clean Air) Regulations, gress & Sustainable Energy, 34, 736–743.

7 of 8 September/October 2019 Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep
 19447450, 2019, 5, Downloaded from https://aiche.onlinelibrary.wiley.com/doi/10.1002/ep.13189 by Nat Prov Indonesia, Wiley Online Library on [24/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
5. Salihu, A., & Alam, M.,.Z. (2015). Upgrading strategies for 20. Rozainee, M., Ramli, M., & Rashid, M. (1998). A field evalu-
effective utilization of biogas, Environmental Progress & ation of particulate emissions from palm oil mills, Jurnal
Sustainable Energy, 34, 1512–1520. Teknologi, 29, 1–6.
6. Vijaya, S., Ma, A.N., Choo, Y.M., & Nik Meriam, N.S. 21. Rashid, M., & Rozainee, M. (1993). Particulate emissions
(2008). Life cycle inventory of the production of crude from a palm oil mill plant – a case study, Jurnal Teknologi,
palm oil – a gate to gate case study of 12 palm oil mills, 22, 19–24.
Journal of Oil Palm Research, 20, 484–494. 22. Kun, Y.C., & Abdullah, A.M. (2013). Simulation of total dust
7. Loh, S.K., Nasrin, A.B., Mohammad Azri, S., Nurul Adela, B., emission from palm oil mills in Malaysia using biomass fuel
Muzzammil, N., Daryl Jay, T., Stasha Eleanor, R.A., Lim, W.S., composition and dust collector efficiency models, Interna-
Choo, Y.M., & Kaltschmitt, M. (2017). First report on Malay- tional Journal of Energy and Environmental Engineering,
sia’s experiences and development in biogas capture and uti- 4, 1–13.
lization from palm oil mill effluent under the economic 23. Rashid, M., Syahirah, M.M., Norruwaida, J., & Huda, N.
transformation programme: current and future perspectives, (2014). Characteristics of particulate emission from a bio-
Renewable and Sustainable Energy Reviews, 74, 1257–1274. mass fired boiler, Agriculture and Agriculture Science Pro-
8. MALAYSIAN PALM OIL BOARD (2017). Sectoral status cedia, 2, 265–271.
2017. http://bepi.mpob.gov.my /index.php/en/Statistics/- 24. GREEN LAGOON Technology (2016). Waste to energy solu-
sectoral-status/170-sectoral-status-2017.html/, accessed on tions for your environmental liabilities. (http://www.glt.
January 4, 2019. my/downloads/glt_iuasb.pdf) [accessed November 2016]
9. Nasrin, A. B., Loh, S. K., Mohammad Azri, S., Nurul Adela, B., 25. Nasrin, A.B., Vijaya, S., Loh, S.K., Astimar, A.A., & Lim, W.
Muzzammil, N., Daryl Jay, T., Stasha Eleanor, R. A., S. (2017b). Quality compliance and environmental impact
Astimar, A. A., Mohamed Fazil, M. S., Mohd. Kamahl, M. K., assessment of commercial empty fruit bunch (EFB) pellet
Lew, Y. S., & Lim, D. Y. (2017). Update on biogas capture and fuel in Malaysia, Journal of Oil Palm Research, 29,
utilization under Economic Transformation Programme. Proc. 570–578.
of the PIPOC 2017 Int. P.O Cong. – Chemistry, Processing 26. Rashid, M., Chong, W.C., Ramli, M., Zainura, Z.N., & Norruwaida,
Technology and Bioenergy. MPOB, Bangi. pp. 128–134. J. (2013). Evaluation of particulate emission from a palm oil mill
10. Heidari, A., Hajinezhad, A., & Aslani, A. (2018). A sustain- boiler, Sains Malaysiana, 42, 1289–1292.
able power supply system, Iran’s opportunities via bioe- 27. Mitchell, E.J.S., Lea-Langton, A.R., Jones, J.M., Williams, A.,
nergy, Environmental Progress & Sustainable Energy, 38, Layden, P., & Johnson, R. (2016). The impact of fuel prop-
171–188. https://doi.org/10.1002/ep.12937. erties on the emissions from the combustion of biomass
11. Nasrin, A.B., Lim, W.S., Sukiran, M.A., Loh, S.K., & Bukhari, N. and other solid fuel in a fixed bed domestic stove, Fuel
A. (2016). Cofiring of biogas in palm oil mill biomass boilers, Processing Technology, 142, 115–123.
Palm Oil Engineering Bulletin, 120, 23–26. 28. DEPARTMENT OF ENVIRONMENT (DOE) (1978). Environ-
12. DEPARTMENT OF ENVIRONMENT (DOE) (2014). Environ- ment Quality (Clean Air) Regulations 1978 – Regulation 43.
ment Quality (Clean Air) Regulations 2014. https://www.doe. http;//cp.doe.gov.my/cpvc/wp-content/uploads/2011/Reg
gov.my/portalv1/wp-content/uploads/2015/01/Peraturan- ulations/Environment_Quality(Clean_Air)pdf, accessed on
peraturan_kualiti_alam_sekeliling_udara_bersih_2014.pdf/, February 5, 2018.
accessed on January 3, 2018. 29. Pennise, D. (2018). Biomass pollution basics. http://www.
13. Ismail, I. (2015). Regulatory requirements for biogas who.int/interventions/antiguamod21.pdf, accessed on
plants, effluent discharge and flue gases emissions for January 17, 2018.
palm oil mills. Paper session presentation at the PIPOC 30. Yusoff, A.R., & Aziz, I.A. (2009). Predicting boiler emission
2015 international palm oil congress – chemistry, proces- by using artificial neural networks, Jurnal Teknologi, 50,
sing technology and bioenergy conference, Kuala Lumpur, 15–28.
Malaysia, October 6–8, 2015. 31. Fournel, S., Palacious, J.H., Moriiette, R., Villeneuve, J.,
14. Niessen, W.R. (2002). Combustion and incineration pro- Godbout, S., Heitz, M., & Savoie, P. (2015). Influence of
cesses. (3rd Edition), New York: Marcel Dekker Inc. biomass properties on technical and environmental perfor-
15. Jahnke, J.A. (2000). Continuous emission monitoring. (2nd mance of a multi-fuel boiler during on-farm combustion of
Edition), New York: John Willey & Sons. energy crops, Applied Energy, 141, 247–259.
16. Husain, Z., Zainal, Z.A., & Abdullah, M.Z. (2002). Analysis 32. Demirbas, A. (2007). Hazardous emissions from combus-
of biomass residue- based cogeneration system in palm oil tion of biomass, Energy Sources, Part A: Recovery, Utiliza-
mills, Biomass and Bioenergy, 24, 117–114. tion, and Environmental Effects, 30, 170–178.
17. Mohamed, H., Sapuan, S.M., Megat Ahmad, M.M.H., & 33. Anzola, M. (2012). Biomass boiler emissions and chimney
Fuad, A. (2008). Numerical study of heat loss from boiler height – A review of practice in the UK and other EU
using different ratios of fibre-to-shell from palm oil wastes, countries. Master thesis, University of Strathclyde, United
Journal of Scientific & Industrial Research, 67, 440–444. Kingdom. http://www.esru.strath.ac.uk/Documents/Msc.
18. Nasrin, A.B., Ravi, N., Lim, W.S., Choo, Y.M., & Fadzil, A. 2012/Anzola.pdf, accessed on November 14, 2017.
M. (2011). Assessment of the performance and potential 34. Glarborg, P., Jensen, A.D., & Johnson, J.E. (2003). Fuel
export renewable energy from typical cogeneration plants nitrogen conversion in solid fuel fired system, Progress in
used in palm oil mills, Journal of Engineering and Applied Energy and Combustion Science, 29, 89–113.
Sciences, 6, 433–439. 35. Popa, E., & Iordache, I. (2007). Dynamics of nitrogen-
19. Nazri, T., & Mohd Amin, A.M. (2016). Thermodynamic analy- bound fuel conversion to nitrogen oxides by biomass com-
sis on oil palm biomass cogeneration plant, ARPN Journal of bustion, Revue Roumaine des Sciences Techniques - Serie
Engineering and Applied Sciences, 11, 12959–12966. Électrotechnique et Énergétique, 52, 121–128.

Environmental Progress & Sustainable Energy (Vol.38, No.5) DOI 10.1002/ep September/October 2019 8 of 8